User login
Research and Reviews for the Practicing Oncologist
Exemestane for postmenopausal women at increased risk of breast cancer
Report prepared by Matt Stenger
The selective estrogen-receptor modulators tamoxifen and raloxifene (Evista) are available for chemoprevention of breast cancer in women at increased risk of disease but are rarely used due to fears of serious adverse effects; there is an increased risk of endometrial cancer with tamoxifen and an increased risk of thromboembolism with both agents. The aromatase inhibitor exemestane (Aromasin) was recently shown to reduce the risk of invasive cancer in women at moderate risk, with no serious toxic effects and minimal changes in quality of life (QOL).1
In a double-blind trial, 4,560 postmenopausal women aged 35 years or older who were at moderate risk of breast cancer were randomized to receive exemestane (25 mg/day; n = 2,285) or placebo (n = 2,275).1 Patients had to have at least one of the following risk factors: age ≥ 60 years, Gail 5-year risk score (chance in 100 of developing invasive breast cancer) > 1.66%, prior atypical ductal hyperplasia (ADH) or lobular hyperplasia (ALH) or lobular carcinoma in situ (LCIS), or ductal carcinoma in situ (DCIS) treated with mastectomy. The primary endpoint of the trial was incidence of invasive breast cancer, with the trial being designed to detect a 65% relative reduction in risk with exemestane. The trial was event-driven, with patients continuing to receive study treatment for a planned maximum duration of 5 years or until occurrence of a breast event, neoplastic disease, cardiovascular event, or unacceptable toxicity.
The two study groups were well matched for baseline characteristics and risk factors. The median age was 62.5 years in the exemestane group vs 62.4 years in the placebo group, with 67.6% vs 69.1% being ≥ 60 years of age. The median body mass index values were 27.9 vs 28.1 kg/m2. The Gail model score was > 1.66% in 40.7% vs 39.8%; 48.8% vs 49.5% were aged ≥ 60 years (as a risk factor); 8.1% vs 8.3% had ADH, ALH, or LCIS on biopsy; and 2.5% vs 2.5% had DCIS treated with mastectomy. The median Gail score was 2.3% in both groups, with 57.8% vs 57.1% having a score > 2.0%. Prior use of menopausal hormone therapy; bone mineral density; history of clinical fracture; cardiovascular risk factors; and use of bisphosphonates, lipid-lowering drugs, and cardiovascular drugs were similar in the two groups. At the time of the clinical data cutoff, 33% of women randomized to receive exemestane and 29% of those randomized to receive placebo were no longer taking the study medication. About 5% in each group had completed treatment. The primary reason for discontinuation of treatment was toxicity, which resulted in discontinuation in 15.4% of exemestane recipients and 10.8% of placebo recipients (P < 0.0001). Patient refusal of treatment resulted in discontinuation in 6.9% vs 6.0% (P = not significant).
At a median follow-up of 35 months, invasive breast cancer had occurred in 11 exemestane recipients and 32 placebo recipients, representing a significant 65% reduction in annual incidence with exemestane(0.19% vs 0.55%; hazard ratio [HR], 0.35; P = 0.002; Figure 1). Exemestane was also associated with a lower incidence in the combined endpoint of invasive breast cancer or DCIS (20 vs 44 cases), with a significant 53% reduction in annual incidence (0.35% vs 0.77%; HR, 0.47; P = 0.004). Fewer cases of DCIS (9 vs 14) and the combined endpoint of ADH, ALH, or LCIS (4 vs 11 cases) occurred in the exemestane group, although the reduction in annual incidence of these outcomes did not reach statistical significance (0.16% vs 0.24%; HR, 0.65; P = 0.31 for DCIS; 0.07% vs 0.20%; HR, 0.36; P = 0.08 for ADH, ALH, or LCIS). Exemestane appeared to demonstrate consistent superiority over placebo in the subgroups included in the planned subgroup analysis (Figure 1). The number needed to treat to prevent 1 case of invasive breast cancer with exemestane was 94 over 3 years and 26 over 5 years (although few women had completed 5 years of therapy at data cutoff).
Adverse events of any grade were slightly more common with exemestane (88% vs 85%; P = 0.003). By grade, 21% vs 25% were grade 1; 42% vs 39% were grade 2; 24% vs 19% were grade 3; and 1% vs 1% were grade 4. Table 1 shows the incidence of adverse events that occurred in 5% or more of women, with a difference between groups of 1% or more, and the incidence of toxicities specified as secondary endpoints. Although there were statistically significant differences between groups for some adverse events, the absolute percentage differences were usually small. Menopausal symptoms were frequent and somewhat more common with exemestane. The most frequent adverse events were hot flashes (40% vs 32%; P < 0.001) and joint pain (30% vs 27%; P = 0.04); arthritis was also more common with exemestane (11% vs 9%; P = 0.01). There were no differences between groups in clinical fracture rates, new diagnoses of osteoporosis, new prescriptions for bisphosphonates, cardiovascular events, or rates of other cancers or time to detection of other cancers. There were 19 deaths in each group, with causes consisting of breast cancer in 1 exemestane recipient vs 0 placebo recipients, other cancers in 10 patients vs 12 patients, cardiovascular causes in 5 vs 4, and other causes in 3 vs 3. None of the deaths was considered treatment-related.
Health-related and menopausespecific QOL were assessed by the Medical Outcomes Study 36-item Short-Form Health Survey (SF-36) and the Menopause-Specific QOL questionnaire. When distributions of worsened, stable, and improved scores were compared, no differences in health-related QOL between groups were found. Women taking exemestane had an overall 7% worsened menopause-related QOL compared with those taking placebo.
Reference
1. Goss PE, Ingle JN, Alés-Martínez JE, et al. Exemestane for breast-cancer prevention in postmenopausal women. N Engl J Med 2011;364:2381–2391.
COMMENTARY
An aromatase inhibitor for breast cancer prevention: a promising option with barriers to resolve
D. Lawrence Wickerham, MD, Drexel University College of Medicine, Philadelphia, PA
The first results of the National Cancer Institute of Canada Clinical Trials Group (NCIC–CTG) MAP-3 trial, a randomized doubleblind placebo-controlled study of the aromatase inhibitor (AI) exemestane, was presented at the June 2011 meeting of the American Society of Oncology1 and was published in the New England Journal of Medicine.2 They represent good news for postmenopausal women at increased risk for developing breast cancer.
With a median follow-up of 35 months, the exemestane-treated women had a 65% (0.19% vs 0.55%; hazard ratio, 0.35; 95% confidence interval, 0.18 to 0.70; P = 0.002) reduction in the incidence of invasive breast cancer. A total of 4,560 postmenopausal women from the United States, Canada, Spain, and France entered this study between 2004 and 2010. All the women were at increased risk, which was determined based on age 60 years or older, prior breast biopsy showing atypical hyperplasia or lobular carcinoma in situ (LCIS), prior ductal carcinoma in situ (DCIS) treated by mastectomy, or a Gail model risk score of greater than 1.66% of developing breast cancer over the next 5 years.
Concerns and barriers amid the good news
Although I have heard my colleagues express some concerns about the trial results, I find most of their issues to be relatively minor. The median follow-up is only about 3 years, and relatively few of the trial participants have completed 5 years of therapy. However, the results are highly statistically significant and are in keeping with the data in adjuvant therapy trials, where the AIs appear to have a durable benefit in reducing new primary cancers of the opposite breast. At the time of analysis, only 43 invasive breast cancers had been diagnosed— 11 in the exemestane-treated group and 32 in the placebo group.
With the estimated reduction in invasive breast cancer of 65%, again estimated from treatment trials and compared with an untreated control, it takes relatively few events to confirm the reduction in invasive breast cancer, which was the primary endpoint. The women were carefully followed and had annual mammograms. The cancers that did occur in both groups were small, early-stage tumors, making a survival benefit almost impossible to document; however, the study was not designed to demonstrate such a finding.
In addition, the treatment assignments were unblinded, the women were informed of their current treatment, and the placebo group was offered the opportunity to cross over to exemestane. These actions make follow-up analyses less meaningful, but the study officials should be applauded for this approach. Although individuals enter trials such as MAP- 3 with the hope that it will benefit them, they understand that if new information concerning their care options becomes available, they will be informed.
There will be barriers to the routine use of exemestane for risk reduction. First, it has not yet been approved for this use by the US Food and Drug Administration, so insurance coverage may not apply. The drug is scheduled to become available generically in the near future, but it currently costs $300-plus per month.
Second, AIs reduce estrogen levels in postmenopausal women to almost zero; that is an effective way to treat and now prevent breast cancer. However, AIs are associated with potential side effects and toxicities, which can be a barrier to their use. Use of AIs can reduce bone density and may result in bone fractures. In MAP-3, bone density measures were obtained at study entry but not routinely during the trial. During treatment, reports of a diagnosis of osteoporosis were balanced in both groups, and fracture rates were similar. Here, the short follow-up and patient selection may have been factors. With prolonged use of AIs, the bone loss can be cumulative. The development of osteoporosis in women who start with normal bone density is low, but for women who already have lowbone mass, other prevention options may be a better first choice. Following AI-treated women with periodic bone density studies and the use of bisphosphonates to blunt the bone impact is an additional approach but further adds to the cost of treatment.
Third, the biggest problem with exemestane may be that to obtain the 65% risk reduction, the drug must be taken on a daily basis. Among the most common side effects of the AIs are troubling arthralgias and myalgias; these problems are the most common reason for women discontinuing AIs during the treatment of breast cancer, and in MAP-3, almost one-third of the exemestane-treated women had stopped taking their medication. That fact does not impact the trial results, which achieved the risk reduction despite the drop-offs, but for those women who stop their medication early, it is unlikely they will achieve substantial risk reduction. In addition, medication adherence within trials is often better than in the real world, due to patients volunteering and being committed to the trial plus the trial investigators monitoring and encouraging adherence. Similar adherence rates can be achieved outside research studies, but they require effort and reinforcement at each follow-up visit.
Patients with DCIS
Participants with a history of DCIS treated by mastectomy were eligible for MAP-3 and appear to have a risk reduction benefit similar in magnitude to other trial participants. However, it would be premature to begin to use exemestane in the adjuvant treatment of patients with receptor-positive DCIS. That step should await the results of two adjuvant AI trials in DCIS that have completed accrual—IBIS II (International Breast Cancer Intervention Study) and NSABP (National Surgical Adjuvant Breast and Bowel Project) B-35—both of which compare anastrozole with tamoxifen in patients with receptor-positive DCIS treated by lumpectomy. Exemestane did reduce the number of cases of DCIS that were diagnosed, although that reduction did not reach statistical significance. However, given the overall sample size and the median follow-up, the lack of significance is not surprising and is unlikely to be clinically important.
Overall, the MAP-3 results are impressive and demonstrate that exemestane should be included with tamoxifen and raloxifene as an effective option for breast cancer prevention.
References
1. Goss PE, Ingle JN, Ales-Martinez J, et al. Exemestane for primary prevention of breast cancer in postmenopausal women: NCIC CTG MAP.3—a randomized, placebo-controlled clinical trial. J Clin Oncol 2011;29:LBA504.
2. Goss PE, Ingle JN, Alés-Martínez JE, et al. Exemestane for breast-cancer prevention in postmenopausal women. N Engl J Med 2011;364:2381–2391.
WHAT'S NEW, WHAT'S IMPORTANT
Jame Abraham, MD, Editor
Breast cancer is the most common cancer in women. The National Surgical Adjuvant Breast and Bowel Project P-1 and P-2 studies both showed that selective inhibition of estrogen receptors with such drugs as tamoxifen or raloxifene (Evista) can decrease the incidence of breast cancer by about 50% in women who are at high risk of developing breast cancer based on the Gail risk model. Several studies in postmenopausal patients have shown that aromatase inhibitors are effective agents for the treatment of breast cancer when used in the metastatic, neoadjuvant, or adjuvant setting. The study by Paul Goss and colleagues described here has shown, for the first time, that exemestane, a steroidal aromatize inhibitor, can decrease the incidence of breast cancer by 65% in high-risk, postmenopausal patients.
So what do we tell our patients? In postmenopausal women who are at high risk of developing breast cancer, per the Gail risk criteria, we now have three drugs available: tamoxifen, raloxifene, and exemestane. When we consider any preventive intervention, we need to look at both the risk and the benefit expected. Because the median followup in the Goss study was only 3 years, the long-term side effects of exemestane are not well established in this setting.
Report prepared by Matt Stenger
The selective estrogen-receptor modulators tamoxifen and raloxifene (Evista) are available for chemoprevention of breast cancer in women at increased risk of disease but are rarely used due to fears of serious adverse effects; there is an increased risk of endometrial cancer with tamoxifen and an increased risk of thromboembolism with both agents. The aromatase inhibitor exemestane (Aromasin) was recently shown to reduce the risk of invasive cancer in women at moderate risk, with no serious toxic effects and minimal changes in quality of life (QOL).1
In a double-blind trial, 4,560 postmenopausal women aged 35 years or older who were at moderate risk of breast cancer were randomized to receive exemestane (25 mg/day; n = 2,285) or placebo (n = 2,275).1 Patients had to have at least one of the following risk factors: age ≥ 60 years, Gail 5-year risk score (chance in 100 of developing invasive breast cancer) > 1.66%, prior atypical ductal hyperplasia (ADH) or lobular hyperplasia (ALH) or lobular carcinoma in situ (LCIS), or ductal carcinoma in situ (DCIS) treated with mastectomy. The primary endpoint of the trial was incidence of invasive breast cancer, with the trial being designed to detect a 65% relative reduction in risk with exemestane. The trial was event-driven, with patients continuing to receive study treatment for a planned maximum duration of 5 years or until occurrence of a breast event, neoplastic disease, cardiovascular event, or unacceptable toxicity.
The two study groups were well matched for baseline characteristics and risk factors. The median age was 62.5 years in the exemestane group vs 62.4 years in the placebo group, with 67.6% vs 69.1% being ≥ 60 years of age. The median body mass index values were 27.9 vs 28.1 kg/m2. The Gail model score was > 1.66% in 40.7% vs 39.8%; 48.8% vs 49.5% were aged ≥ 60 years (as a risk factor); 8.1% vs 8.3% had ADH, ALH, or LCIS on biopsy; and 2.5% vs 2.5% had DCIS treated with mastectomy. The median Gail score was 2.3% in both groups, with 57.8% vs 57.1% having a score > 2.0%. Prior use of menopausal hormone therapy; bone mineral density; history of clinical fracture; cardiovascular risk factors; and use of bisphosphonates, lipid-lowering drugs, and cardiovascular drugs were similar in the two groups. At the time of the clinical data cutoff, 33% of women randomized to receive exemestane and 29% of those randomized to receive placebo were no longer taking the study medication. About 5% in each group had completed treatment. The primary reason for discontinuation of treatment was toxicity, which resulted in discontinuation in 15.4% of exemestane recipients and 10.8% of placebo recipients (P < 0.0001). Patient refusal of treatment resulted in discontinuation in 6.9% vs 6.0% (P = not significant).
At a median follow-up of 35 months, invasive breast cancer had occurred in 11 exemestane recipients and 32 placebo recipients, representing a significant 65% reduction in annual incidence with exemestane(0.19% vs 0.55%; hazard ratio [HR], 0.35; P = 0.002; Figure 1). Exemestane was also associated with a lower incidence in the combined endpoint of invasive breast cancer or DCIS (20 vs 44 cases), with a significant 53% reduction in annual incidence (0.35% vs 0.77%; HR, 0.47; P = 0.004). Fewer cases of DCIS (9 vs 14) and the combined endpoint of ADH, ALH, or LCIS (4 vs 11 cases) occurred in the exemestane group, although the reduction in annual incidence of these outcomes did not reach statistical significance (0.16% vs 0.24%; HR, 0.65; P = 0.31 for DCIS; 0.07% vs 0.20%; HR, 0.36; P = 0.08 for ADH, ALH, or LCIS). Exemestane appeared to demonstrate consistent superiority over placebo in the subgroups included in the planned subgroup analysis (Figure 1). The number needed to treat to prevent 1 case of invasive breast cancer with exemestane was 94 over 3 years and 26 over 5 years (although few women had completed 5 years of therapy at data cutoff).
Adverse events of any grade were slightly more common with exemestane (88% vs 85%; P = 0.003). By grade, 21% vs 25% were grade 1; 42% vs 39% were grade 2; 24% vs 19% were grade 3; and 1% vs 1% were grade 4. Table 1 shows the incidence of adverse events that occurred in 5% or more of women, with a difference between groups of 1% or more, and the incidence of toxicities specified as secondary endpoints. Although there were statistically significant differences between groups for some adverse events, the absolute percentage differences were usually small. Menopausal symptoms were frequent and somewhat more common with exemestane. The most frequent adverse events were hot flashes (40% vs 32%; P < 0.001) and joint pain (30% vs 27%; P = 0.04); arthritis was also more common with exemestane (11% vs 9%; P = 0.01). There were no differences between groups in clinical fracture rates, new diagnoses of osteoporosis, new prescriptions for bisphosphonates, cardiovascular events, or rates of other cancers or time to detection of other cancers. There were 19 deaths in each group, with causes consisting of breast cancer in 1 exemestane recipient vs 0 placebo recipients, other cancers in 10 patients vs 12 patients, cardiovascular causes in 5 vs 4, and other causes in 3 vs 3. None of the deaths was considered treatment-related.
Health-related and menopausespecific QOL were assessed by the Medical Outcomes Study 36-item Short-Form Health Survey (SF-36) and the Menopause-Specific QOL questionnaire. When distributions of worsened, stable, and improved scores were compared, no differences in health-related QOL between groups were found. Women taking exemestane had an overall 7% worsened menopause-related QOL compared with those taking placebo.
Reference
1. Goss PE, Ingle JN, Alés-Martínez JE, et al. Exemestane for breast-cancer prevention in postmenopausal women. N Engl J Med 2011;364:2381–2391.
COMMENTARY
An aromatase inhibitor for breast cancer prevention: a promising option with barriers to resolve
D. Lawrence Wickerham, MD, Drexel University College of Medicine, Philadelphia, PA
The first results of the National Cancer Institute of Canada Clinical Trials Group (NCIC–CTG) MAP-3 trial, a randomized doubleblind placebo-controlled study of the aromatase inhibitor (AI) exemestane, was presented at the June 2011 meeting of the American Society of Oncology1 and was published in the New England Journal of Medicine.2 They represent good news for postmenopausal women at increased risk for developing breast cancer.
With a median follow-up of 35 months, the exemestane-treated women had a 65% (0.19% vs 0.55%; hazard ratio, 0.35; 95% confidence interval, 0.18 to 0.70; P = 0.002) reduction in the incidence of invasive breast cancer. A total of 4,560 postmenopausal women from the United States, Canada, Spain, and France entered this study between 2004 and 2010. All the women were at increased risk, which was determined based on age 60 years or older, prior breast biopsy showing atypical hyperplasia or lobular carcinoma in situ (LCIS), prior ductal carcinoma in situ (DCIS) treated by mastectomy, or a Gail model risk score of greater than 1.66% of developing breast cancer over the next 5 years.
Concerns and barriers amid the good news
Although I have heard my colleagues express some concerns about the trial results, I find most of their issues to be relatively minor. The median follow-up is only about 3 years, and relatively few of the trial participants have completed 5 years of therapy. However, the results are highly statistically significant and are in keeping with the data in adjuvant therapy trials, where the AIs appear to have a durable benefit in reducing new primary cancers of the opposite breast. At the time of analysis, only 43 invasive breast cancers had been diagnosed— 11 in the exemestane-treated group and 32 in the placebo group.
With the estimated reduction in invasive breast cancer of 65%, again estimated from treatment trials and compared with an untreated control, it takes relatively few events to confirm the reduction in invasive breast cancer, which was the primary endpoint. The women were carefully followed and had annual mammograms. The cancers that did occur in both groups were small, early-stage tumors, making a survival benefit almost impossible to document; however, the study was not designed to demonstrate such a finding.
In addition, the treatment assignments were unblinded, the women were informed of their current treatment, and the placebo group was offered the opportunity to cross over to exemestane. These actions make follow-up analyses less meaningful, but the study officials should be applauded for this approach. Although individuals enter trials such as MAP- 3 with the hope that it will benefit them, they understand that if new information concerning their care options becomes available, they will be informed.
There will be barriers to the routine use of exemestane for risk reduction. First, it has not yet been approved for this use by the US Food and Drug Administration, so insurance coverage may not apply. The drug is scheduled to become available generically in the near future, but it currently costs $300-plus per month.
Second, AIs reduce estrogen levels in postmenopausal women to almost zero; that is an effective way to treat and now prevent breast cancer. However, AIs are associated with potential side effects and toxicities, which can be a barrier to their use. Use of AIs can reduce bone density and may result in bone fractures. In MAP-3, bone density measures were obtained at study entry but not routinely during the trial. During treatment, reports of a diagnosis of osteoporosis were balanced in both groups, and fracture rates were similar. Here, the short follow-up and patient selection may have been factors. With prolonged use of AIs, the bone loss can be cumulative. The development of osteoporosis in women who start with normal bone density is low, but for women who already have lowbone mass, other prevention options may be a better first choice. Following AI-treated women with periodic bone density studies and the use of bisphosphonates to blunt the bone impact is an additional approach but further adds to the cost of treatment.
Third, the biggest problem with exemestane may be that to obtain the 65% risk reduction, the drug must be taken on a daily basis. Among the most common side effects of the AIs are troubling arthralgias and myalgias; these problems are the most common reason for women discontinuing AIs during the treatment of breast cancer, and in MAP-3, almost one-third of the exemestane-treated women had stopped taking their medication. That fact does not impact the trial results, which achieved the risk reduction despite the drop-offs, but for those women who stop their medication early, it is unlikely they will achieve substantial risk reduction. In addition, medication adherence within trials is often better than in the real world, due to patients volunteering and being committed to the trial plus the trial investigators monitoring and encouraging adherence. Similar adherence rates can be achieved outside research studies, but they require effort and reinforcement at each follow-up visit.
Patients with DCIS
Participants with a history of DCIS treated by mastectomy were eligible for MAP-3 and appear to have a risk reduction benefit similar in magnitude to other trial participants. However, it would be premature to begin to use exemestane in the adjuvant treatment of patients with receptor-positive DCIS. That step should await the results of two adjuvant AI trials in DCIS that have completed accrual—IBIS II (International Breast Cancer Intervention Study) and NSABP (National Surgical Adjuvant Breast and Bowel Project) B-35—both of which compare anastrozole with tamoxifen in patients with receptor-positive DCIS treated by lumpectomy. Exemestane did reduce the number of cases of DCIS that were diagnosed, although that reduction did not reach statistical significance. However, given the overall sample size and the median follow-up, the lack of significance is not surprising and is unlikely to be clinically important.
Overall, the MAP-3 results are impressive and demonstrate that exemestane should be included with tamoxifen and raloxifene as an effective option for breast cancer prevention.
References
1. Goss PE, Ingle JN, Ales-Martinez J, et al. Exemestane for primary prevention of breast cancer in postmenopausal women: NCIC CTG MAP.3—a randomized, placebo-controlled clinical trial. J Clin Oncol 2011;29:LBA504.
2. Goss PE, Ingle JN, Alés-Martínez JE, et al. Exemestane for breast-cancer prevention in postmenopausal women. N Engl J Med 2011;364:2381–2391.
WHAT'S NEW, WHAT'S IMPORTANT
Jame Abraham, MD, Editor
Breast cancer is the most common cancer in women. The National Surgical Adjuvant Breast and Bowel Project P-1 and P-2 studies both showed that selective inhibition of estrogen receptors with such drugs as tamoxifen or raloxifene (Evista) can decrease the incidence of breast cancer by about 50% in women who are at high risk of developing breast cancer based on the Gail risk model. Several studies in postmenopausal patients have shown that aromatase inhibitors are effective agents for the treatment of breast cancer when used in the metastatic, neoadjuvant, or adjuvant setting. The study by Paul Goss and colleagues described here has shown, for the first time, that exemestane, a steroidal aromatize inhibitor, can decrease the incidence of breast cancer by 65% in high-risk, postmenopausal patients.
So what do we tell our patients? In postmenopausal women who are at high risk of developing breast cancer, per the Gail risk criteria, we now have three drugs available: tamoxifen, raloxifene, and exemestane. When we consider any preventive intervention, we need to look at both the risk and the benefit expected. Because the median followup in the Goss study was only 3 years, the long-term side effects of exemestane are not well established in this setting.
Report prepared by Matt Stenger
The selective estrogen-receptor modulators tamoxifen and raloxifene (Evista) are available for chemoprevention of breast cancer in women at increased risk of disease but are rarely used due to fears of serious adverse effects; there is an increased risk of endometrial cancer with tamoxifen and an increased risk of thromboembolism with both agents. The aromatase inhibitor exemestane (Aromasin) was recently shown to reduce the risk of invasive cancer in women at moderate risk, with no serious toxic effects and minimal changes in quality of life (QOL).1
In a double-blind trial, 4,560 postmenopausal women aged 35 years or older who were at moderate risk of breast cancer were randomized to receive exemestane (25 mg/day; n = 2,285) or placebo (n = 2,275).1 Patients had to have at least one of the following risk factors: age ≥ 60 years, Gail 5-year risk score (chance in 100 of developing invasive breast cancer) > 1.66%, prior atypical ductal hyperplasia (ADH) or lobular hyperplasia (ALH) or lobular carcinoma in situ (LCIS), or ductal carcinoma in situ (DCIS) treated with mastectomy. The primary endpoint of the trial was incidence of invasive breast cancer, with the trial being designed to detect a 65% relative reduction in risk with exemestane. The trial was event-driven, with patients continuing to receive study treatment for a planned maximum duration of 5 years or until occurrence of a breast event, neoplastic disease, cardiovascular event, or unacceptable toxicity.
The two study groups were well matched for baseline characteristics and risk factors. The median age was 62.5 years in the exemestane group vs 62.4 years in the placebo group, with 67.6% vs 69.1% being ≥ 60 years of age. The median body mass index values were 27.9 vs 28.1 kg/m2. The Gail model score was > 1.66% in 40.7% vs 39.8%; 48.8% vs 49.5% were aged ≥ 60 years (as a risk factor); 8.1% vs 8.3% had ADH, ALH, or LCIS on biopsy; and 2.5% vs 2.5% had DCIS treated with mastectomy. The median Gail score was 2.3% in both groups, with 57.8% vs 57.1% having a score > 2.0%. Prior use of menopausal hormone therapy; bone mineral density; history of clinical fracture; cardiovascular risk factors; and use of bisphosphonates, lipid-lowering drugs, and cardiovascular drugs were similar in the two groups. At the time of the clinical data cutoff, 33% of women randomized to receive exemestane and 29% of those randomized to receive placebo were no longer taking the study medication. About 5% in each group had completed treatment. The primary reason for discontinuation of treatment was toxicity, which resulted in discontinuation in 15.4% of exemestane recipients and 10.8% of placebo recipients (P < 0.0001). Patient refusal of treatment resulted in discontinuation in 6.9% vs 6.0% (P = not significant).
At a median follow-up of 35 months, invasive breast cancer had occurred in 11 exemestane recipients and 32 placebo recipients, representing a significant 65% reduction in annual incidence with exemestane(0.19% vs 0.55%; hazard ratio [HR], 0.35; P = 0.002; Figure 1). Exemestane was also associated with a lower incidence in the combined endpoint of invasive breast cancer or DCIS (20 vs 44 cases), with a significant 53% reduction in annual incidence (0.35% vs 0.77%; HR, 0.47; P = 0.004). Fewer cases of DCIS (9 vs 14) and the combined endpoint of ADH, ALH, or LCIS (4 vs 11 cases) occurred in the exemestane group, although the reduction in annual incidence of these outcomes did not reach statistical significance (0.16% vs 0.24%; HR, 0.65; P = 0.31 for DCIS; 0.07% vs 0.20%; HR, 0.36; P = 0.08 for ADH, ALH, or LCIS). Exemestane appeared to demonstrate consistent superiority over placebo in the subgroups included in the planned subgroup analysis (Figure 1). The number needed to treat to prevent 1 case of invasive breast cancer with exemestane was 94 over 3 years and 26 over 5 years (although few women had completed 5 years of therapy at data cutoff).
Adverse events of any grade were slightly more common with exemestane (88% vs 85%; P = 0.003). By grade, 21% vs 25% were grade 1; 42% vs 39% were grade 2; 24% vs 19% were grade 3; and 1% vs 1% were grade 4. Table 1 shows the incidence of adverse events that occurred in 5% or more of women, with a difference between groups of 1% or more, and the incidence of toxicities specified as secondary endpoints. Although there were statistically significant differences between groups for some adverse events, the absolute percentage differences were usually small. Menopausal symptoms were frequent and somewhat more common with exemestane. The most frequent adverse events were hot flashes (40% vs 32%; P < 0.001) and joint pain (30% vs 27%; P = 0.04); arthritis was also more common with exemestane (11% vs 9%; P = 0.01). There were no differences between groups in clinical fracture rates, new diagnoses of osteoporosis, new prescriptions for bisphosphonates, cardiovascular events, or rates of other cancers or time to detection of other cancers. There were 19 deaths in each group, with causes consisting of breast cancer in 1 exemestane recipient vs 0 placebo recipients, other cancers in 10 patients vs 12 patients, cardiovascular causes in 5 vs 4, and other causes in 3 vs 3. None of the deaths was considered treatment-related.
Health-related and menopausespecific QOL were assessed by the Medical Outcomes Study 36-item Short-Form Health Survey (SF-36) and the Menopause-Specific QOL questionnaire. When distributions of worsened, stable, and improved scores were compared, no differences in health-related QOL between groups were found. Women taking exemestane had an overall 7% worsened menopause-related QOL compared with those taking placebo.
Reference
1. Goss PE, Ingle JN, Alés-Martínez JE, et al. Exemestane for breast-cancer prevention in postmenopausal women. N Engl J Med 2011;364:2381–2391.
COMMENTARY
An aromatase inhibitor for breast cancer prevention: a promising option with barriers to resolve
D. Lawrence Wickerham, MD, Drexel University College of Medicine, Philadelphia, PA
The first results of the National Cancer Institute of Canada Clinical Trials Group (NCIC–CTG) MAP-3 trial, a randomized doubleblind placebo-controlled study of the aromatase inhibitor (AI) exemestane, was presented at the June 2011 meeting of the American Society of Oncology1 and was published in the New England Journal of Medicine.2 They represent good news for postmenopausal women at increased risk for developing breast cancer.
With a median follow-up of 35 months, the exemestane-treated women had a 65% (0.19% vs 0.55%; hazard ratio, 0.35; 95% confidence interval, 0.18 to 0.70; P = 0.002) reduction in the incidence of invasive breast cancer. A total of 4,560 postmenopausal women from the United States, Canada, Spain, and France entered this study between 2004 and 2010. All the women were at increased risk, which was determined based on age 60 years or older, prior breast biopsy showing atypical hyperplasia or lobular carcinoma in situ (LCIS), prior ductal carcinoma in situ (DCIS) treated by mastectomy, or a Gail model risk score of greater than 1.66% of developing breast cancer over the next 5 years.
Concerns and barriers amid the good news
Although I have heard my colleagues express some concerns about the trial results, I find most of their issues to be relatively minor. The median follow-up is only about 3 years, and relatively few of the trial participants have completed 5 years of therapy. However, the results are highly statistically significant and are in keeping with the data in adjuvant therapy trials, where the AIs appear to have a durable benefit in reducing new primary cancers of the opposite breast. At the time of analysis, only 43 invasive breast cancers had been diagnosed— 11 in the exemestane-treated group and 32 in the placebo group.
With the estimated reduction in invasive breast cancer of 65%, again estimated from treatment trials and compared with an untreated control, it takes relatively few events to confirm the reduction in invasive breast cancer, which was the primary endpoint. The women were carefully followed and had annual mammograms. The cancers that did occur in both groups were small, early-stage tumors, making a survival benefit almost impossible to document; however, the study was not designed to demonstrate such a finding.
In addition, the treatment assignments were unblinded, the women were informed of their current treatment, and the placebo group was offered the opportunity to cross over to exemestane. These actions make follow-up analyses less meaningful, but the study officials should be applauded for this approach. Although individuals enter trials such as MAP- 3 with the hope that it will benefit them, they understand that if new information concerning their care options becomes available, they will be informed.
There will be barriers to the routine use of exemestane for risk reduction. First, it has not yet been approved for this use by the US Food and Drug Administration, so insurance coverage may not apply. The drug is scheduled to become available generically in the near future, but it currently costs $300-plus per month.
Second, AIs reduce estrogen levels in postmenopausal women to almost zero; that is an effective way to treat and now prevent breast cancer. However, AIs are associated with potential side effects and toxicities, which can be a barrier to their use. Use of AIs can reduce bone density and may result in bone fractures. In MAP-3, bone density measures were obtained at study entry but not routinely during the trial. During treatment, reports of a diagnosis of osteoporosis were balanced in both groups, and fracture rates were similar. Here, the short follow-up and patient selection may have been factors. With prolonged use of AIs, the bone loss can be cumulative. The development of osteoporosis in women who start with normal bone density is low, but for women who already have lowbone mass, other prevention options may be a better first choice. Following AI-treated women with periodic bone density studies and the use of bisphosphonates to blunt the bone impact is an additional approach but further adds to the cost of treatment.
Third, the biggest problem with exemestane may be that to obtain the 65% risk reduction, the drug must be taken on a daily basis. Among the most common side effects of the AIs are troubling arthralgias and myalgias; these problems are the most common reason for women discontinuing AIs during the treatment of breast cancer, and in MAP-3, almost one-third of the exemestane-treated women had stopped taking their medication. That fact does not impact the trial results, which achieved the risk reduction despite the drop-offs, but for those women who stop their medication early, it is unlikely they will achieve substantial risk reduction. In addition, medication adherence within trials is often better than in the real world, due to patients volunteering and being committed to the trial plus the trial investigators monitoring and encouraging adherence. Similar adherence rates can be achieved outside research studies, but they require effort and reinforcement at each follow-up visit.
Patients with DCIS
Participants with a history of DCIS treated by mastectomy were eligible for MAP-3 and appear to have a risk reduction benefit similar in magnitude to other trial participants. However, it would be premature to begin to use exemestane in the adjuvant treatment of patients with receptor-positive DCIS. That step should await the results of two adjuvant AI trials in DCIS that have completed accrual—IBIS II (International Breast Cancer Intervention Study) and NSABP (National Surgical Adjuvant Breast and Bowel Project) B-35—both of which compare anastrozole with tamoxifen in patients with receptor-positive DCIS treated by lumpectomy. Exemestane did reduce the number of cases of DCIS that were diagnosed, although that reduction did not reach statistical significance. However, given the overall sample size and the median follow-up, the lack of significance is not surprising and is unlikely to be clinically important.
Overall, the MAP-3 results are impressive and demonstrate that exemestane should be included with tamoxifen and raloxifene as an effective option for breast cancer prevention.
References
1. Goss PE, Ingle JN, Ales-Martinez J, et al. Exemestane for primary prevention of breast cancer in postmenopausal women: NCIC CTG MAP.3—a randomized, placebo-controlled clinical trial. J Clin Oncol 2011;29:LBA504.
2. Goss PE, Ingle JN, Alés-Martínez JE, et al. Exemestane for breast-cancer prevention in postmenopausal women. N Engl J Med 2011;364:2381–2391.
WHAT'S NEW, WHAT'S IMPORTANT
Jame Abraham, MD, Editor
Breast cancer is the most common cancer in women. The National Surgical Adjuvant Breast and Bowel Project P-1 and P-2 studies both showed that selective inhibition of estrogen receptors with such drugs as tamoxifen or raloxifene (Evista) can decrease the incidence of breast cancer by about 50% in women who are at high risk of developing breast cancer based on the Gail risk model. Several studies in postmenopausal patients have shown that aromatase inhibitors are effective agents for the treatment of breast cancer when used in the metastatic, neoadjuvant, or adjuvant setting. The study by Paul Goss and colleagues described here has shown, for the first time, that exemestane, a steroidal aromatize inhibitor, can decrease the incidence of breast cancer by 65% in high-risk, postmenopausal patients.
So what do we tell our patients? In postmenopausal women who are at high risk of developing breast cancer, per the Gail risk criteria, we now have three drugs available: tamoxifen, raloxifene, and exemestane. When we consider any preventive intervention, we need to look at both the risk and the benefit expected. Because the median followup in the Goss study was only 3 years, the long-term side effects of exemestane are not well established in this setting.
IV iron sucrose for cancer and/or chemotherapy-induced anemia in patients treated with erythropoiesis-stimulating agents
Lowell B. Anthony, MD,1 Nashat Y. Gabrail, MD,2 Hassan Ghazal, MD,3, Donald V. Woytowitz, MD,4 Marshall S. Flam, MD,5 Anibal Drelichman, MD,6, David M. Loesch, MD,7, Demi A. Niforos, MS,8, and Antoinette Mangione, MD, PharmD9; for the Iron Sucrose Study Group*
1 Stanley S. Scott Cancer Center, Louisiana State University Health Sciences Center, New Orleans, LA; 2 Nashat Cancer Center, Canton, OH; 3 Kentucky Cancer Clinic, Hazard, KY; 4 Florida Cancer Specialists, Fort Myers, FL; 5 Hematology/Oncology Group of Fresno, Fresno, CA; 6 Newland Medical Associates, Southfield, MI; 7 Oncology/Hematology Associates, Indianapolis, IN; 8 AAI Pharma, Inc., Natick, MA; and 9 Luitpold Pharmaceuticals/American Regent, Inc., Norristown, PA
Manuscript received January 2, 2011; accepted June 16, 2011.
This work was presented at the 43rd Annual Meeting of the American Society of Clinical Oncology; June 1–5, 2007 in Chicago, IL, and was supported by Luitpold Pharmaceuticals/American Regent, Inc., Shirley, NY.
Correspondence to: Lowell B. Anthony, MD, LSUHSC New Orleans, Ochsner Kenner Medical Center, 200 West Esplanade, Kenner, LA 70065; e-mail: [email protected].
Conflicts of interest: Ms. Niforos was a fulltime salaried employee of AAI Pharma, Inc., contracted to perform all biostatistical services for the clinical trial. Dr. Mangione was a fulltime salaried employee of the trial sponsor, Luitpold Pharmaceuticals/American Regent, Inc. Drs. Anthony, Gabrail, Ghazal, Woytowitz, Flam, Drelichman, and Loesch have nothing to disclose.
Mild-to-moderate anemia occurs in up to 75% of cancer patients undergoing either single- or multimodality therapy and may contribute to an increased morbidity and reduced quality of life (QOL).1–4 This form of anemia resembles anemia of chronic disease, with a blunted erythropoietin response and inadequate erythropoietin production.5 Increasing hemoglobin (Hgb) concentrations and reducing red blood cell (RBC) transfusions while improving QOL and tolerance to cancer therapies are the treatment-related goals.
Intravenous (IV) iron is commonly administered with ESAs in CKD-associated anemia.12,13 Most studies regarding IV iron replacement in cancer and/or chemotherapy-induced anemia (CCIA) are positive, with one exception: Steensma et al14 reported no benefit in adding IV ferric gluconate to an ESA in a phase III randomized trial in which an oral placebo and iron were used as comparators. Practice guidelines are inconsistent, as the National Comprehensive Cancer Network (NCCN) recommends the IV route when iron is prescribed,6 and the American Society of Hematology/ American Society of Clinical Oncology considers the evidence insufficient to support routine IV iron use.15,16 Auerbach et al17 demonstrated that IV iron dextran results in a greater Hgb level increase than oral iron in ESAtreated patients. Approved formulations of IV iron in the United States include iron dextran, iron sucrose, and ferric gluconate, with the majority of published data with iron dextran.15,18,19 However, the iron dextrans have black-box warnings, and test doses are recommended. Henry et al20 reported that IV ferric gluconate significantly increased Hgb response when compared with oral iron or no iron and was well tolerated in CCIA.
Early work with IV iron sucrose includes a trial evaluating 67 lymphoma patients randomized between ESA or ESA with IV iron sucrose.21 Despite adequate bone marrow iron stores, the Hgb response was greater (91% vs 54%) and the time to reach a Hgb level > 2 g/dL was less (6 vs 12 weeks) in the IV iron-treated group.21 Another trial randomized 398 CCIA patients between fixed IV iron doses (mean weekly dose, 64.8 mg) with ESA versus standard practice (2% received IV iron).22 IV iron resulted in a trend toward a higher ferritin level, but transferrin saturation (TSAT) remained similar between the two groups.22 A study in patients with noniron-deficient anemic solid tumors receiving chemotherapy also demonstrated an increase in hemoglobin levels statistically favoring the darbepoetin alfa (Aranesp)/iron group.23 As additional information is needed, this study was performed to determine whether IV iron sucrose combined with ESA increases Hgb levels in CCIA patients who have been previously treated with an ESA.
Patients and methods
Patient eligibility
his was an open-label, phase III, randomized, institutional review board-approved, multicenter study at 56 US centers. After signing informed consent, patients ≥ 18 years of age with a histologic diagnosis of cancer (acute leukemia or myeloproliferative syndrome excluded) receiving ongoing or planned chemotherapy, with a Hgb level ≤ 10.0 g/dL, body weight > 50 kg, and a Karnofsky performance status of ≥ 60%, were eligible. Patients were excluded if they had iron depletion, active infection, myelophthisic bone marrow (except for hematologic malignancy), hypoplastic bone marrow, uncontrolled hypertension, bleeding, or planned surgery. To ensure a stable baseline Hgb value, no IV iron within 2 months of consent or RBC transfusions within 3 weeks of randomization were allowed.
Treatment
After 8 weeks of fixed ESA doses in stage 1, patients were classified as either ESA responders (≥ 1 g/dL Hgb level increase from baseline) or nonresponders (< 1 g/dL Hgb level increase from baseline), with each group separately randomized centrally using block randomization to receive either IV iron sucrose or no iron treatment (Figure 1). At the time of randomization (beginning of stage 2), patients were stratified according to malignancy type (solid tumor vs hematologic) and Hgb level (< 12 g/dL vs ≥ 12 g/dL for ESA responders; < 9.5 g/dL vs ≥ 9.5 g/dL for ESA nonresponders).
The calculated dose of the study drug (iron sucrose [Venofer]; 7 mg/kg up to 500 mg maximum) was added to 500 mL of normal or half-normal saline and administered IV over 4 hours.24 Patients randomized to receive iron sucrose were scheduled to receive up to three infusions at 1- to 3-week intervals during the first 9 weeks of stage 2, with the first dose administered as soon as possible after randomization. The last dose of ESA was given on or before week 12 of stage 2.
Outcome measures
The primary endpoint for efficacy was the change from baseline (end of stage 1) to the maximum Hgb level achieved during stage 2 in patients who responded to ESA. Major secondary endpoints included changes in Hgb levels when iron sucrose was added to ESA nonresponders as well as the percentage of all randomized patients with Hgb level increases > 1 g/dL, > 2 g/dL, and > 3 g/dL; changes in Hgb levels and iron indices from baseline at each visit; and changes in the 13-item Functional Assessment of Chronic Illness Therapy (FACIT) fatigue scale. Hgb levels were obtained weekly, and iron indices were measured every 3 weeks. The FACIT fatigue scale was measured during stage 1 at consent, weeks 4, and 8 and during stage 2 weeks 3, 6, 9, and at the end of the study.
Adverse events were recorded hourly during iron sucrose administration and from the day of randomization through study completion or 30 days following the last dose of study drug, whichever was later. Investigators provided the date of onset, severity, relationship, date of resolution, action taken, and adverse event outcome. Adverse drug events were events considered by the investigator to be possibly, probably, or definitely related to the study drug.
Statistical method
The sample size was based on the hypothesis that iron-treated ESA responders (group A) would have a 1.0 g/dL or higher mean increase in Hgb levels than would ESA responders who did not receive iron (group B). The standard deviation (SD) of the difference was assumed to be ≤ 1.5 g/dL. Targeting a 1.0 g/dL change in Hgb level to be significant, 49 patients/ group were required (alpha = 0.05; beta = 0.10). Assuming that the ESA response rate in stage I was at least 40% and that the stage I and stage 2 dropout rates were no more than 10% and 25%, respectively, 325 patients were the targeted number for stage I enrollment, with adjustments made by monitoring the stage I response rate.
The intent-to-treat (ITT) population included patients randomized into stage 2 based on actual treatment. The evaluable population included ITT patients who completed at least 10 weeks of stage 2 or who had interventions (RBC transfusions or nonstudy iron) prior to week 10.
Continuous variables were assessed using analysis of covariance and t-tests. Ordinal responses were analyzed with the Fisher’s exact test and Cochran-Mantel-Haenszel statistics. Changes from baseline to each visit for all FACIT scores were assessed for treatment groups with the unpaired two-sample t-test.
Results
Patient disposition and demographics
Of the 375 patients enrolled during the run-in stage 1 period (between July 2003 and October 2005), 132 patients discontinued treatment (the most common reasons were a required intervention [50], withdrawn consent [23], and adverse events [17]). Fourteen patients completed stage 1 but did not enter stage 2. Figure 2 shows the numbers of patients who were randomly assigned to the two treatment groups and were evaluated for safety and efficacy as well as reasons for study discontinuation. Table 1 shows the patient numbers assigned to the various treatment groups (A to D) based on ESA response in stage I and the study population; it also demonstrates the similar baseline demographic characteristics between the treatment groups. At baseline (ie, prior to randomization), there were no statistically significant differences in Hgb level, TSAT, and ferritin level between the ESA responders (A vs B) and nonresponders (C vs D).
Efficacy of iron sucrose
Mean maximum improvement in Hgb levels (Table 2). Among ESA responders (groups A and B), a statistically significantly greater mean maximum Hgb level increase was observed among patients who received iron sucrose (group A) than among those who did not (group B), achieving the primary endpoint (ITT, P = 0.004; evaluable, P = 0.008). A statistically significant greater increase in the mean maximum Hgb level was observed following iron sucrose (groups A and C) when compared with no iron treatment (groups B and D), regardless of prior ESA response. In the ESA nonresponder group, a significant increase (P = 0.027) in the mean maximum Hgb level was observed between those who received iron sucrose (group C) and those who did not (group D) in the ITT population; a statistical difference was not seen in the evaluable population (P = 0.082).
With regard to tumor subtypes, breast cancer and other tumor types, but not lung cancer, were associated with statistically significant increases in maximum Hgb levels following iron sucrose, regardless of prior ESA response.
Absolute increases in Hgb levels (Table 2). A greater proportion of patients assigned to IV iron sucrose achieved a ≥ 2 g/dL and ≥ 3 g/dL increase in Hgb level during the study than did those who did not receive iron. These differences were statistically significant for all the groups except for the evaluable ≥ 3 g/dL nonresponder group. The only statistically significant difference in the proportion achieving a ≥ 1 g/dL Hgb level increase occurred in the ESA nonresponder groups. In addition, baseline hematologic characteristics and iron indices did not predict the efficacy of IV iron treatment (as defined by a > 1 g/dL or > 2 g/dL increase in Hgb level). In the IV iron sucrose-treated group, there was no statistical difference in these baseline characteristics in the patients who demonstrated a > 1 g/dL (data not shown) or a > 2 g/dL treatment response to IV iron.
Changes from baseline in Hgb and ferritin levels and in TSAT. Figure 3 summarizes the Hgb level, ferritin level, and TSAT responses by study visit after IV iron sucrose compared with no iron in the ITT population. Between treatment groups, statistically significant differences (P < 0.05) were present by weeks 7, 3, and 13 for Hgb level, ferritin level, and TSAT, respectively. At the end of the study, week 13, the mean Hgb level increase from baseline was 2.3 g/dL versus 1.2 g/dL (P < 0.002), the mean ferritin level increase from baseline was 419 ng/mL versus a decrease of 50 ng/mL (P < 0.001), and the mean TSAT increase from baseline was 8.8% versus 0.2% (P < 0.005) in the iron sucrose versus no iron group.
Changes in fatigue levels (FACIT fatigue scale). There was a statistically significant decrease in the level of fatigue at the end of the study compared with at baseline (end of stage 1) in the iron sucrose-treated patients in the ITT but not in the evaluable population (–3.3 iron sucrose/–2.1 no iron, P = 0.022 ITT; –3.0 iron sucrose/–1.7 no iron, P = 0.058 evaluable population). No significant decrease in the level of fatigue was experienced by the patients who received no iron. There were no statistically significant differences between the groups in changes from baseline at each visit..
Safety of iron sucrose
Extent of exposure. In the ITT population, the mean per patient total dose of iron sucrose administered was 1,123 (SD, 402) mg in group A (responders) and 1,113 (SD, 387) mg in group C (nonresponders).
Adverse drug events (ADEs). All safety analyses were performed using the ITT population. Serious ADEs were experienced by three patients in the iron sucrose group (chest pain, hypersensitivity, and hypotension, one patient each) and by no patients in the ESA-only group. One ESA-only group patient (arthralgia) and four iron sucrose patients (hypersensitivity; abdominal pain; arthralgia and muscle cramps; myalgia, nausea, and vomiting) were prematurely discontinued from the study drug due to the occurrence of an ADE.
At least one ADE was experienced by 37.4% of the patients in the iron sucrose group and 0.8% in the control group. The most common (³ 5%) ADEs were nausea (8.1%), dysgeusia (8.1%), back pain (6.1%), arthralgia (6.1%), muscle cramp (6.1%), and peripheral edema (5.1%). Within the ESA-only group, the only ADE reported was hypertension (one subject, 0.8%).
Eleven grade 3 (National Institutes of Health/National Cancer Institute– Common Terminology Criteria, version 2.0) ADEs occurred in iron sucrose-treated patients and included nausea (2.0%), hypotension (2.0%), abdominal pain (1.0%), chest pain (1.0%), hypersensitivity (1.0%), arthralgia (1.0%), dizziness (1.0%), dyspnea (1.0%), and hypertension (1.0%). A serious grade 3 hypotensive event occurred in a 49-year-old woman weighing 50 kg who experienced dizziness, nausea, vomiting, and transient hypotension (110/60 mm Hg to 70/40 mm Hg) after her first iron sucrose dose of 375 mg. Ninety minutes later, following IV steroids, iron sucrose was restarted and the hypotension recurred. The patient received two subsequent lower iron sucrose doses (200 mg over 4 hours), with no further adverse reactions.
Deaths and thrombotic events. These events are summarized in Table 3. None of these events was judged by the investigators to be related to the study drug.
Laboratory results. Statistically greater mean increases in ferritin levels, TSAT, Hgb levels, hematocrit, mean corpuscular hemoglobin, mean corpuscular volume, and monocytes oc curred in the iron sucrose-treated group. There were no significant differences between treatment groups in clinical chemistry safety laboratory results.
Discussion
This study is the first to evaluate IV iron in CCIA patients who have received prior ESA therapy. IV iron sucrose administered with ESAs significantly increased Hgb levels in CCIA patients. Prior ESA response did not predict Hgb level response to iron sucrose, as benefit was demonstrated in both ESA responders and nonresponders. Baseline hematologic/ iron indices also did not predict IV iron responsiveness, as these characteristics were similar in IV iron responders and nonresponders. Improvement in QOL, as measured by fatigue levels at study completion, was also observed after IV iron but not in the no iron group. IV iron studies are commonly open-label because of the difficulty in blinding iron’s viscous dark-colored solution.
This study design limits the significance of QOL measurements in IV iron studies, where primary endpoints are typically objective measurements. Even though transfusion rates were lower in the IV iron groups (5.1% in groups A and C [A = 1.7%; C = 10%]) than in the no iron groups (10.4% in groups B and D [B = 2.6%; D = 22.9%]), this difference was not statistically significant (Fisher’s exact test, P = 0.215). Our findings support the prior observations that IV iron replacement in combination with ESAs effectively increases Hgb levels and is safe.17,20,21,25,26
Combining IV iron with ESA increases the Hgb level response and may either shorten the time to response and/or decrease the ESA requirement. Approximately 30%–50% of patients are nonresponders after 12–24 weeks of ESA therapy.8,9,17,27,28 Iron deficiency may be a major factor accounting for ESA resistance. Decreased ESA responsiveness in the dialysis population can be corrected by providing adequate iron supplementation. 11,18 Also, ESA nonresponders may become responders with IV iron replacement while continuing the ESA. ESA treatment in responders can produce a functional iron deficiency, because the ESA produces a rapid initiation of erythropoiesis. Inducing functional iron deficiency with ESA therapy implies that the iron supply to the erythron may be the rate-limiting step in erythropoiesis, and the IV iron dose may be important.25 As ESA responders and nonresponders experienced improvement in Hgb levels with IV iron therapy in this trial, IV iron supplementation may be required to achieve and/or maintain a response to ESA therapy.
Iron available for erythropoiesis is derived from the balance between dietary sources and that in the usable pool within the reticuloendothelial system.29 ESA therapy can result in RBC production that exceeds the rate of iron mobilization, even with adequate iron stores. Inflammatory cytokines may also hinder the release of stored iron from macrophages by inducing hepcidin and thus further contribute to an inadequate rate of RBC production.30–34
Of note, baseline ferritin levels were higher in the ESA nonresponders (groups C and D) than in the ESA responders (groups A and B), although these differences were not statistically significant. This finding may be consistent with elevated inflammatory cytokines impairing the availability of iron, leading to a failed ESA response. ESA resistance is multifactorial, with these factors contributing to the rapid depletion of the usable iron pool, thus blunting the ESA response. Identifying factors that allow for maximizing ESA therapy in CCIA patients may result in greater ESA efficiency. The IV route of iron replacement is superior to oral administration and accounts for one of these variables.17,21,25,26
Safely administering IV iron is an important factor that influences the choice of iron preparations. In the United States, the only IV iron indicated for iron deficiency anemia is iron dextran. The risk of allergic reactions and the need for test doses may account for practitioners limiting the use of iron dextran, despite a compelling medical need for rapid, reliable, and safe replenishment of body iron in populations such as those with CKD35–37 and CCIA. The non–dextran- containing IV irons (iron sucrose, ferric gluconate) are currently only FDA approved for CKD indications at doses of 100–200 mg over 2–5 minutes or up to 400 mg over 2.5 hours for iron sucrose and only 125 mg over 10 minutes for ferric gluconate. 18,19
This study supports other findings that IV iron sucrose is generally well tolerated at doses of 7 mg/kg, up to a maximum of 500 mg over 4 hours, in CCIA. Caution should be exercised, however, especially in patients with a lower body weight. This concern is supported by a study of iron sucrose in nondialysis CKD, where hypotension occurred in two patients < 65 kg after 500 mg doses were administered over 4 hours.38
Conclusion
This study’s primary objective was to determine whether prior response to ESA treatment would influence response to IV iron, not to detect differences between functional and absolute iron deficiency. Our findings support that administration of IV iron while continuing ESA treatment may correct functional, as well as absolute, iron deficiency in CCIA. Baseline iron indices did not predict responsiveness to iron sucrose. Without additional data identifying predictors of ESA responsiveness in CCIA, a more proactive approach that includes IV iron may be warranted, as in CKDrelated anemia. As a better understanding of functional iron deficiency evolves, it is becoming apparent that IV iron is important to optimize the response to ESAs for CCIA. Additional studies are needed to understand the mechanisms responsible for functional iron deficiency in CCIA and to assist in identifying the optimal IV iron administration schedule.
Acknowledgments: The authors wish to thank the study coordinators; the patients at each of the participating centers; and Drs. Perry Rigby and Robert Means, for reviewing the manuscript.
*Additional members of the Iron Sucrose Study Group include Ali Ben-Jacob, MD, Cache Valley Cancer Treatment and Research Clinic, Inc., Logan, UT; Amol Rakkar, MD, Hope Center, Terre Haute, IN; Philip Chatham, MD, Granada Hills, CA; Ahmed Maqbool, MD, Welborn Clinic, Research Center, Evansville, IN; Timothy Pluard, MD, Washington University, Medical Oncology, St. Peters, MO; Nafisa Burhani, MD, Joliet Oncology- Hematology Associates, LTD, Joliet, IL; David Henry, MD, Pennsylvania Hematology and Oncology Associates, Philadelphia, PA; David Watkins, MD, Allison Cancer Center, Midland, TX; Howard Ozer, MD, University of Oklahoma Health Science Center-Hematology Oncology Section, Oklahoma City, OK; Leo Orr, MD, Leo E. Orr, Inc., Los Angeles, CA; Billy Clowney, MD, Santee Hematology Oncology, Sumter, SC, Rene Rothestein-Rubin, MD, Rittenhouse Hematology/ Oncology, Philadelphia, PA; Peter Eisenberg, MD, California Cancer Care, Greenbrae, CA; Rosalba Rodriguez, MD, Chula Vista, CA; Kumar Kapisthalam, MD, United Professional Center, Pasco Hernando Oncology, New Port Richey, FL; Jennifer Caskey, MD, Wheat Ridge, CO; Sayed E. Ahmend, MD, Sebring, FL; Patricia Braly, MD, Hematology and Oncology Specialties, New Orleans, LA; Donald Flemming, MD, Medical Center of Vincennes, The Bierhaus Center, Vincennes, IN; William Tester, MD, Albert Einstein Cancer Center, Philadelphia, PA; William Solomon, MD, SUNY Downstate Medical Center, Brooklyn, NY; Mark Hancock, MD, Mile Hile Oncology, Denver, CO; Youssef Hanna, MD, Huron Medical Center, Port Huron, MI; Scot Sorensen, MD, Prairie View Clinic, Lincoln, NE; and Mark Yoffe, MD, Raleigh, NC.
References
1. Birgegård G, Gascón P, Ludwig H. Evaluation of anaemia in patients with multiple myeloma and lymphoma: findings of the European Cancer Anaemia Survey. Eur J Haematol 2006;77:378–386.
2. Groopman JE, Itri LM. Chemotherapyinduced anemia in adults: incidence and treatment. J Natl Cancer Inst 1999;91:1616–1634.
3. Ludwig H, Van Belle S, Barrett-Lee P, et al. The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey defining the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 2004;40:2293–2306.
4. Williams B, Tannous R, Gupta S, et al. Chemotherapy-induced anemia in geriatric patients with non-Hodgkin’s lymphoma. Consultant Pharmacist 2002;17:131–140.
5. Miller CB, Jones RJ, Piantadosi S, Abeloff MD, Spivak JL. Decreased erythropoietin response in patients with the anemia of cancer. N Engl J Med 1990;322:1689–1692.
6. NCCN Clinical Practice Guidelines in Oncology: Cancer- and Chemotherapy-Induced Anemia, Version 2.2011. National Comprehensive Cancer Network Web site. http:// www.nccn.org/professionals/physician_gls/ pdf/anemia.pdf. Accessed June 16, 2011.
7. Rizzo JD, Brouwers M, Hurley P, et al. American Society of Hematology/American Society of Clinical Oncology clinical practice guideline update on the use of epoetin and darbepoetin in adult patients with cancer. Blood 2010;116:4045–4059.
8. Demetri GD, Kris M, Wade J, Degos L, Cella D. Quality-of-life benefit in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 1998;16:3412–3425.
9. Gabrilove JL, Cleeland CS, Livingston RB, Sarokhan B, Winer E, Einhorn LH. Clinical evaluation of once-weekly dosing of epoetin alfa in chemotherapy patients: improvements in hemoglobin and quality of life are similar to three-times-weekly dosing. J Clin Oncol 2001;19:2875–2882.
10. Glaspy J, Bukowski R, Steinberg D, Taylor C, Tchekmedyian S, Vadhan-Raj S. Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. Procrit Study Group. J Clin Oncol 1997;15:1218–1234.
11. Drüeke TB, Bárány P, Cazzola M, et al. Management of iron deficiency in renal anemia: guidelines for the optimal therapeutic approach in erythropoietin-treated patients. Clin Nephrol 1997;48:1–8.
12. Fishbane S, Frei GL, Maesaka J. Reduction in recombinant human erythropoietin doses by the use of chronic intravenous iron supplementation. Am J Kidney Dis 1995;26:41–46.
13. Van Wyck DB, Roppolo M, Martinez CO, et al. A randomized, controlled trial comparing IV iron sucrose to oral iron in anemic patients with nondialysis-dependent CKD. Kidney Int 2005;68:2846–2856.
14. Steensma DP, Sloan JA, Dakhil SR, et al. Phase III, randomized study of the effects of parenteral iron, oral iron, or no iron supplementation on the erythropoietic re sponse to darbepoetin alfa for patients with chemotherapy-associated anemia. J Clin Oncol 2011;29:97–105.
15. Auerbach M, Ballard H, Glaspy J. Clinical update: intravenous iron for anaemia. Lancet 2007;369:1502–1504.
16. Bokemeyer C, Aapro MS, Courdi A, et al. EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer: 2006 update. Eur J Cancer 2007;43:258– 270.
17. Auerbach M, Ballard H, Trout JR, et al. Intravenous iron optimizes the response to recombinant human erythropoietin in cancer patients with chemotherapy-related anemia: a multicenter, open-label, randomized trial. J Clin Oncol 2004;22:1301–1307.
18. Aronoff GR, Bennett WM, Blumenthal S, et al. Iron sucrose in hemodialysis patients: safety of replacement and maintenance regimens. Kidney Int 2004;66:1193–1198.
19. Faich G, Strobos J. Sodium ferric gluconate complex in sucrose: safer intravenous iron therapy than iron dextrans. Am J Kidney Dis 1999;33:464–470.
20. Henry DH, Dahl NV, Auerbach M, Tchekmedyian S, Laufman LR. Intravenous ferric gluconate significantly improves response to epoetin alfa versus oral iron or no iron in anemic patients with cancer receiving chemotherapy. Oncologist 2007;12:231–242.
21. Hedenus M, Birgegård G, Näsman P, et al. Addition of intravenous iron to epoetin beta increases hemoglobin response and decreases epoetin dose requirement in anemic patients with lymphoproliferative malignancies: a randomized multicenter study. Leukemia 2007;21:627–632.
22. Bastit L, Vandebroek A, Altintas S, et al. Randomized, multicenter, controlled trial comparing the efficacy and safety of darbepoetin alpha administered every 3 weeks with or without intravenous iron in patients with chemotherapy-induced anemia. J Clin Oncol 2008;26:1611–1618.
23. Pedrazzoli P, Farris A, Del Prete S, et al. Randomized trial of intravenous iron supplementation in patients with chemotherapy- related anemia without iron deficiency treated with darbepoetin alpha. J Clin Oncol 2008;26:1619–1625.
24. Chandler G, Harchowal J, Macdougall IC. Intravenous iron sucrose: establishing a safe dose. Am J Kidney Dis 2001;38:988–991.
25. Lerchenmueller C, Husseini F, Gaede B, Mossman T, Suto T, Vanderbroek A. Intravenous (IV) iron supplementation in patients with chemotherapy-induced anemia (CIA) receiving darbepoetin alfa every 3 weeks (q3w): iron parameters in a randomized controlled trial. Blood 2006;108:1552.
26. Pinter T, Mossman T, Suto T, Vansteenkiste J. Effects of intravenous iron supplementation on responses to every-3-week darbepoetin alfa by baseline hemoglobin in patients with chemotherapy-induced anemia. J Clin Oncol 2007;25(18S):9106.
27. Glaspy J, Jadeja JS, Justice G, et al. A dose-finding and safety study of novel erythropoiesis stimulating protein (NESP) for the treatment of anaemia in patients receiving multicycle chemotherapy. Br J Cancer 2001;84(suppl 1):17–23.
28. Littlewood TJ, Bajetta E, Nortier JW, Vercammen E, Rapoport B; Epoetin Alfa Study Group. Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: results of a randomized, double-blind, placebocontrolled trial. J Clin Oncol 2001;19:2865– 2874.
29. Henry DH. Supplemental iron: a key to optimizing the response of cancer-related anemia to rHuEPO? Oncologist 1998;3:275–278. 30. Ganz T. Hepcidin—a regulator of intestinal iron absorption and iron recycling by macrophages. Best Pract ClinHaematol 2005;18:171–182.
31. Ganz T. Hepcidin—a peptide hormone at the interface of innate immunity and iron metabolism. Curr Top Microbiol Immunol 2006;306:183–198.
32. Viatte L, Nicolas G, Lou DQ, et al. Chronic hepcidin induction causes hyposideremia and alters the pattern of cellular iron accumulation in hemochromatotic mice. Blood 2006;107:2952–2958.
33. Weinstein DA, Roy CN, Fleming MD, Loda MF, Wolfsdorf JI, Andrews NC. Inappropriate expression of hepcidin is associated with iron refractory anemia: implications for the anemia of chronic disease. Blood 2002;100:3776–3781.
34. Wrighting DM, Andrews NC. Interleukin- 6 induces hepcidin expression through STAT3. Blood 2006;108:3204–3209.
35. Wysowski DK, Swartz L, Borders- Hemphill BV, Goulding MR, Dormitzer C. Use of parenteral iron products and serious anaphylactic-type reactions. Am J Hematol 2010;85:650–654.
36. Bailie GR, Clark JA, Lane CE, Lane PL. Hypersensitivity reactions and deaths associated with intravenous iron preparations. Nephrol Dialysis Transplant 2005;20:1443– 1449.
37. Macdougall IC, Roche A. Administration of intravenous iron sucrose as a 2-minute push to CKD patients: a prospective evaluation of 2,297 injections. Am J Kidney Dis 2005;46:283–289.
38. Fishbane S, Ungureanu VD, Maesaka JK, Kaupke CJ, Lim V, Wish J. The safety of intravenous iron dextran in hemodialysis patients. Am J Kidney Dis 1996;28:529–534.
Lowell B. Anthony, MD,1 Nashat Y. Gabrail, MD,2 Hassan Ghazal, MD,3, Donald V. Woytowitz, MD,4 Marshall S. Flam, MD,5 Anibal Drelichman, MD,6, David M. Loesch, MD,7, Demi A. Niforos, MS,8, and Antoinette Mangione, MD, PharmD9; for the Iron Sucrose Study Group*
1 Stanley S. Scott Cancer Center, Louisiana State University Health Sciences Center, New Orleans, LA; 2 Nashat Cancer Center, Canton, OH; 3 Kentucky Cancer Clinic, Hazard, KY; 4 Florida Cancer Specialists, Fort Myers, FL; 5 Hematology/Oncology Group of Fresno, Fresno, CA; 6 Newland Medical Associates, Southfield, MI; 7 Oncology/Hematology Associates, Indianapolis, IN; 8 AAI Pharma, Inc., Natick, MA; and 9 Luitpold Pharmaceuticals/American Regent, Inc., Norristown, PA
Manuscript received January 2, 2011; accepted June 16, 2011.
This work was presented at the 43rd Annual Meeting of the American Society of Clinical Oncology; June 1–5, 2007 in Chicago, IL, and was supported by Luitpold Pharmaceuticals/American Regent, Inc., Shirley, NY.
Correspondence to: Lowell B. Anthony, MD, LSUHSC New Orleans, Ochsner Kenner Medical Center, 200 West Esplanade, Kenner, LA 70065; e-mail: [email protected].
Conflicts of interest: Ms. Niforos was a fulltime salaried employee of AAI Pharma, Inc., contracted to perform all biostatistical services for the clinical trial. Dr. Mangione was a fulltime salaried employee of the trial sponsor, Luitpold Pharmaceuticals/American Regent, Inc. Drs. Anthony, Gabrail, Ghazal, Woytowitz, Flam, Drelichman, and Loesch have nothing to disclose.
Mild-to-moderate anemia occurs in up to 75% of cancer patients undergoing either single- or multimodality therapy and may contribute to an increased morbidity and reduced quality of life (QOL).1–4 This form of anemia resembles anemia of chronic disease, with a blunted erythropoietin response and inadequate erythropoietin production.5 Increasing hemoglobin (Hgb) concentrations and reducing red blood cell (RBC) transfusions while improving QOL and tolerance to cancer therapies are the treatment-related goals.
Intravenous (IV) iron is commonly administered with ESAs in CKD-associated anemia.12,13 Most studies regarding IV iron replacement in cancer and/or chemotherapy-induced anemia (CCIA) are positive, with one exception: Steensma et al14 reported no benefit in adding IV ferric gluconate to an ESA in a phase III randomized trial in which an oral placebo and iron were used as comparators. Practice guidelines are inconsistent, as the National Comprehensive Cancer Network (NCCN) recommends the IV route when iron is prescribed,6 and the American Society of Hematology/ American Society of Clinical Oncology considers the evidence insufficient to support routine IV iron use.15,16 Auerbach et al17 demonstrated that IV iron dextran results in a greater Hgb level increase than oral iron in ESAtreated patients. Approved formulations of IV iron in the United States include iron dextran, iron sucrose, and ferric gluconate, with the majority of published data with iron dextran.15,18,19 However, the iron dextrans have black-box warnings, and test doses are recommended. Henry et al20 reported that IV ferric gluconate significantly increased Hgb response when compared with oral iron or no iron and was well tolerated in CCIA.
Early work with IV iron sucrose includes a trial evaluating 67 lymphoma patients randomized between ESA or ESA with IV iron sucrose.21 Despite adequate bone marrow iron stores, the Hgb response was greater (91% vs 54%) and the time to reach a Hgb level > 2 g/dL was less (6 vs 12 weeks) in the IV iron-treated group.21 Another trial randomized 398 CCIA patients between fixed IV iron doses (mean weekly dose, 64.8 mg) with ESA versus standard practice (2% received IV iron).22 IV iron resulted in a trend toward a higher ferritin level, but transferrin saturation (TSAT) remained similar between the two groups.22 A study in patients with noniron-deficient anemic solid tumors receiving chemotherapy also demonstrated an increase in hemoglobin levels statistically favoring the darbepoetin alfa (Aranesp)/iron group.23 As additional information is needed, this study was performed to determine whether IV iron sucrose combined with ESA increases Hgb levels in CCIA patients who have been previously treated with an ESA.
Patients and methods
Patient eligibility
his was an open-label, phase III, randomized, institutional review board-approved, multicenter study at 56 US centers. After signing informed consent, patients ≥ 18 years of age with a histologic diagnosis of cancer (acute leukemia or myeloproliferative syndrome excluded) receiving ongoing or planned chemotherapy, with a Hgb level ≤ 10.0 g/dL, body weight > 50 kg, and a Karnofsky performance status of ≥ 60%, were eligible. Patients were excluded if they had iron depletion, active infection, myelophthisic bone marrow (except for hematologic malignancy), hypoplastic bone marrow, uncontrolled hypertension, bleeding, or planned surgery. To ensure a stable baseline Hgb value, no IV iron within 2 months of consent or RBC transfusions within 3 weeks of randomization were allowed.
Treatment
After 8 weeks of fixed ESA doses in stage 1, patients were classified as either ESA responders (≥ 1 g/dL Hgb level increase from baseline) or nonresponders (< 1 g/dL Hgb level increase from baseline), with each group separately randomized centrally using block randomization to receive either IV iron sucrose or no iron treatment (Figure 1). At the time of randomization (beginning of stage 2), patients were stratified according to malignancy type (solid tumor vs hematologic) and Hgb level (< 12 g/dL vs ≥ 12 g/dL for ESA responders; < 9.5 g/dL vs ≥ 9.5 g/dL for ESA nonresponders).
The calculated dose of the study drug (iron sucrose [Venofer]; 7 mg/kg up to 500 mg maximum) was added to 500 mL of normal or half-normal saline and administered IV over 4 hours.24 Patients randomized to receive iron sucrose were scheduled to receive up to three infusions at 1- to 3-week intervals during the first 9 weeks of stage 2, with the first dose administered as soon as possible after randomization. The last dose of ESA was given on or before week 12 of stage 2.
Outcome measures
The primary endpoint for efficacy was the change from baseline (end of stage 1) to the maximum Hgb level achieved during stage 2 in patients who responded to ESA. Major secondary endpoints included changes in Hgb levels when iron sucrose was added to ESA nonresponders as well as the percentage of all randomized patients with Hgb level increases > 1 g/dL, > 2 g/dL, and > 3 g/dL; changes in Hgb levels and iron indices from baseline at each visit; and changes in the 13-item Functional Assessment of Chronic Illness Therapy (FACIT) fatigue scale. Hgb levels were obtained weekly, and iron indices were measured every 3 weeks. The FACIT fatigue scale was measured during stage 1 at consent, weeks 4, and 8 and during stage 2 weeks 3, 6, 9, and at the end of the study.
Adverse events were recorded hourly during iron sucrose administration and from the day of randomization through study completion or 30 days following the last dose of study drug, whichever was later. Investigators provided the date of onset, severity, relationship, date of resolution, action taken, and adverse event outcome. Adverse drug events were events considered by the investigator to be possibly, probably, or definitely related to the study drug.
Statistical method
The sample size was based on the hypothesis that iron-treated ESA responders (group A) would have a 1.0 g/dL or higher mean increase in Hgb levels than would ESA responders who did not receive iron (group B). The standard deviation (SD) of the difference was assumed to be ≤ 1.5 g/dL. Targeting a 1.0 g/dL change in Hgb level to be significant, 49 patients/ group were required (alpha = 0.05; beta = 0.10). Assuming that the ESA response rate in stage I was at least 40% and that the stage I and stage 2 dropout rates were no more than 10% and 25%, respectively, 325 patients were the targeted number for stage I enrollment, with adjustments made by monitoring the stage I response rate.
The intent-to-treat (ITT) population included patients randomized into stage 2 based on actual treatment. The evaluable population included ITT patients who completed at least 10 weeks of stage 2 or who had interventions (RBC transfusions or nonstudy iron) prior to week 10.
Continuous variables were assessed using analysis of covariance and t-tests. Ordinal responses were analyzed with the Fisher’s exact test and Cochran-Mantel-Haenszel statistics. Changes from baseline to each visit for all FACIT scores were assessed for treatment groups with the unpaired two-sample t-test.
Results
Patient disposition and demographics
Of the 375 patients enrolled during the run-in stage 1 period (between July 2003 and October 2005), 132 patients discontinued treatment (the most common reasons were a required intervention [50], withdrawn consent [23], and adverse events [17]). Fourteen patients completed stage 1 but did not enter stage 2. Figure 2 shows the numbers of patients who were randomly assigned to the two treatment groups and were evaluated for safety and efficacy as well as reasons for study discontinuation. Table 1 shows the patient numbers assigned to the various treatment groups (A to D) based on ESA response in stage I and the study population; it also demonstrates the similar baseline demographic characteristics between the treatment groups. At baseline (ie, prior to randomization), there were no statistically significant differences in Hgb level, TSAT, and ferritin level between the ESA responders (A vs B) and nonresponders (C vs D).
Efficacy of iron sucrose
Mean maximum improvement in Hgb levels (Table 2). Among ESA responders (groups A and B), a statistically significantly greater mean maximum Hgb level increase was observed among patients who received iron sucrose (group A) than among those who did not (group B), achieving the primary endpoint (ITT, P = 0.004; evaluable, P = 0.008). A statistically significant greater increase in the mean maximum Hgb level was observed following iron sucrose (groups A and C) when compared with no iron treatment (groups B and D), regardless of prior ESA response. In the ESA nonresponder group, a significant increase (P = 0.027) in the mean maximum Hgb level was observed between those who received iron sucrose (group C) and those who did not (group D) in the ITT population; a statistical difference was not seen in the evaluable population (P = 0.082).
With regard to tumor subtypes, breast cancer and other tumor types, but not lung cancer, were associated with statistically significant increases in maximum Hgb levels following iron sucrose, regardless of prior ESA response.
Absolute increases in Hgb levels (Table 2). A greater proportion of patients assigned to IV iron sucrose achieved a ≥ 2 g/dL and ≥ 3 g/dL increase in Hgb level during the study than did those who did not receive iron. These differences were statistically significant for all the groups except for the evaluable ≥ 3 g/dL nonresponder group. The only statistically significant difference in the proportion achieving a ≥ 1 g/dL Hgb level increase occurred in the ESA nonresponder groups. In addition, baseline hematologic characteristics and iron indices did not predict the efficacy of IV iron treatment (as defined by a > 1 g/dL or > 2 g/dL increase in Hgb level). In the IV iron sucrose-treated group, there was no statistical difference in these baseline characteristics in the patients who demonstrated a > 1 g/dL (data not shown) or a > 2 g/dL treatment response to IV iron.
Changes from baseline in Hgb and ferritin levels and in TSAT. Figure 3 summarizes the Hgb level, ferritin level, and TSAT responses by study visit after IV iron sucrose compared with no iron in the ITT population. Between treatment groups, statistically significant differences (P < 0.05) were present by weeks 7, 3, and 13 for Hgb level, ferritin level, and TSAT, respectively. At the end of the study, week 13, the mean Hgb level increase from baseline was 2.3 g/dL versus 1.2 g/dL (P < 0.002), the mean ferritin level increase from baseline was 419 ng/mL versus a decrease of 50 ng/mL (P < 0.001), and the mean TSAT increase from baseline was 8.8% versus 0.2% (P < 0.005) in the iron sucrose versus no iron group.
Changes in fatigue levels (FACIT fatigue scale). There was a statistically significant decrease in the level of fatigue at the end of the study compared with at baseline (end of stage 1) in the iron sucrose-treated patients in the ITT but not in the evaluable population (–3.3 iron sucrose/–2.1 no iron, P = 0.022 ITT; –3.0 iron sucrose/–1.7 no iron, P = 0.058 evaluable population). No significant decrease in the level of fatigue was experienced by the patients who received no iron. There were no statistically significant differences between the groups in changes from baseline at each visit..
Safety of iron sucrose
Extent of exposure. In the ITT population, the mean per patient total dose of iron sucrose administered was 1,123 (SD, 402) mg in group A (responders) and 1,113 (SD, 387) mg in group C (nonresponders).
Adverse drug events (ADEs). All safety analyses were performed using the ITT population. Serious ADEs were experienced by three patients in the iron sucrose group (chest pain, hypersensitivity, and hypotension, one patient each) and by no patients in the ESA-only group. One ESA-only group patient (arthralgia) and four iron sucrose patients (hypersensitivity; abdominal pain; arthralgia and muscle cramps; myalgia, nausea, and vomiting) were prematurely discontinued from the study drug due to the occurrence of an ADE.
At least one ADE was experienced by 37.4% of the patients in the iron sucrose group and 0.8% in the control group. The most common (³ 5%) ADEs were nausea (8.1%), dysgeusia (8.1%), back pain (6.1%), arthralgia (6.1%), muscle cramp (6.1%), and peripheral edema (5.1%). Within the ESA-only group, the only ADE reported was hypertension (one subject, 0.8%).
Eleven grade 3 (National Institutes of Health/National Cancer Institute– Common Terminology Criteria, version 2.0) ADEs occurred in iron sucrose-treated patients and included nausea (2.0%), hypotension (2.0%), abdominal pain (1.0%), chest pain (1.0%), hypersensitivity (1.0%), arthralgia (1.0%), dizziness (1.0%), dyspnea (1.0%), and hypertension (1.0%). A serious grade 3 hypotensive event occurred in a 49-year-old woman weighing 50 kg who experienced dizziness, nausea, vomiting, and transient hypotension (110/60 mm Hg to 70/40 mm Hg) after her first iron sucrose dose of 375 mg. Ninety minutes later, following IV steroids, iron sucrose was restarted and the hypotension recurred. The patient received two subsequent lower iron sucrose doses (200 mg over 4 hours), with no further adverse reactions.
Deaths and thrombotic events. These events are summarized in Table 3. None of these events was judged by the investigators to be related to the study drug.
Laboratory results. Statistically greater mean increases in ferritin levels, TSAT, Hgb levels, hematocrit, mean corpuscular hemoglobin, mean corpuscular volume, and monocytes oc curred in the iron sucrose-treated group. There were no significant differences between treatment groups in clinical chemistry safety laboratory results.
Discussion
This study is the first to evaluate IV iron in CCIA patients who have received prior ESA therapy. IV iron sucrose administered with ESAs significantly increased Hgb levels in CCIA patients. Prior ESA response did not predict Hgb level response to iron sucrose, as benefit was demonstrated in both ESA responders and nonresponders. Baseline hematologic/ iron indices also did not predict IV iron responsiveness, as these characteristics were similar in IV iron responders and nonresponders. Improvement in QOL, as measured by fatigue levels at study completion, was also observed after IV iron but not in the no iron group. IV iron studies are commonly open-label because of the difficulty in blinding iron’s viscous dark-colored solution.
This study design limits the significance of QOL measurements in IV iron studies, where primary endpoints are typically objective measurements. Even though transfusion rates were lower in the IV iron groups (5.1% in groups A and C [A = 1.7%; C = 10%]) than in the no iron groups (10.4% in groups B and D [B = 2.6%; D = 22.9%]), this difference was not statistically significant (Fisher’s exact test, P = 0.215). Our findings support the prior observations that IV iron replacement in combination with ESAs effectively increases Hgb levels and is safe.17,20,21,25,26
Combining IV iron with ESA increases the Hgb level response and may either shorten the time to response and/or decrease the ESA requirement. Approximately 30%–50% of patients are nonresponders after 12–24 weeks of ESA therapy.8,9,17,27,28 Iron deficiency may be a major factor accounting for ESA resistance. Decreased ESA responsiveness in the dialysis population can be corrected by providing adequate iron supplementation. 11,18 Also, ESA nonresponders may become responders with IV iron replacement while continuing the ESA. ESA treatment in responders can produce a functional iron deficiency, because the ESA produces a rapid initiation of erythropoiesis. Inducing functional iron deficiency with ESA therapy implies that the iron supply to the erythron may be the rate-limiting step in erythropoiesis, and the IV iron dose may be important.25 As ESA responders and nonresponders experienced improvement in Hgb levels with IV iron therapy in this trial, IV iron supplementation may be required to achieve and/or maintain a response to ESA therapy.
Iron available for erythropoiesis is derived from the balance between dietary sources and that in the usable pool within the reticuloendothelial system.29 ESA therapy can result in RBC production that exceeds the rate of iron mobilization, even with adequate iron stores. Inflammatory cytokines may also hinder the release of stored iron from macrophages by inducing hepcidin and thus further contribute to an inadequate rate of RBC production.30–34
Of note, baseline ferritin levels were higher in the ESA nonresponders (groups C and D) than in the ESA responders (groups A and B), although these differences were not statistically significant. This finding may be consistent with elevated inflammatory cytokines impairing the availability of iron, leading to a failed ESA response. ESA resistance is multifactorial, with these factors contributing to the rapid depletion of the usable iron pool, thus blunting the ESA response. Identifying factors that allow for maximizing ESA therapy in CCIA patients may result in greater ESA efficiency. The IV route of iron replacement is superior to oral administration and accounts for one of these variables.17,21,25,26
Safely administering IV iron is an important factor that influences the choice of iron preparations. In the United States, the only IV iron indicated for iron deficiency anemia is iron dextran. The risk of allergic reactions and the need for test doses may account for practitioners limiting the use of iron dextran, despite a compelling medical need for rapid, reliable, and safe replenishment of body iron in populations such as those with CKD35–37 and CCIA. The non–dextran- containing IV irons (iron sucrose, ferric gluconate) are currently only FDA approved for CKD indications at doses of 100–200 mg over 2–5 minutes or up to 400 mg over 2.5 hours for iron sucrose and only 125 mg over 10 minutes for ferric gluconate. 18,19
This study supports other findings that IV iron sucrose is generally well tolerated at doses of 7 mg/kg, up to a maximum of 500 mg over 4 hours, in CCIA. Caution should be exercised, however, especially in patients with a lower body weight. This concern is supported by a study of iron sucrose in nondialysis CKD, where hypotension occurred in two patients < 65 kg after 500 mg doses were administered over 4 hours.38
Conclusion
This study’s primary objective was to determine whether prior response to ESA treatment would influence response to IV iron, not to detect differences between functional and absolute iron deficiency. Our findings support that administration of IV iron while continuing ESA treatment may correct functional, as well as absolute, iron deficiency in CCIA. Baseline iron indices did not predict responsiveness to iron sucrose. Without additional data identifying predictors of ESA responsiveness in CCIA, a more proactive approach that includes IV iron may be warranted, as in CKDrelated anemia. As a better understanding of functional iron deficiency evolves, it is becoming apparent that IV iron is important to optimize the response to ESAs for CCIA. Additional studies are needed to understand the mechanisms responsible for functional iron deficiency in CCIA and to assist in identifying the optimal IV iron administration schedule.
Acknowledgments: The authors wish to thank the study coordinators; the patients at each of the participating centers; and Drs. Perry Rigby and Robert Means, for reviewing the manuscript.
*Additional members of the Iron Sucrose Study Group include Ali Ben-Jacob, MD, Cache Valley Cancer Treatment and Research Clinic, Inc., Logan, UT; Amol Rakkar, MD, Hope Center, Terre Haute, IN; Philip Chatham, MD, Granada Hills, CA; Ahmed Maqbool, MD, Welborn Clinic, Research Center, Evansville, IN; Timothy Pluard, MD, Washington University, Medical Oncology, St. Peters, MO; Nafisa Burhani, MD, Joliet Oncology- Hematology Associates, LTD, Joliet, IL; David Henry, MD, Pennsylvania Hematology and Oncology Associates, Philadelphia, PA; David Watkins, MD, Allison Cancer Center, Midland, TX; Howard Ozer, MD, University of Oklahoma Health Science Center-Hematology Oncology Section, Oklahoma City, OK; Leo Orr, MD, Leo E. Orr, Inc., Los Angeles, CA; Billy Clowney, MD, Santee Hematology Oncology, Sumter, SC, Rene Rothestein-Rubin, MD, Rittenhouse Hematology/ Oncology, Philadelphia, PA; Peter Eisenberg, MD, California Cancer Care, Greenbrae, CA; Rosalba Rodriguez, MD, Chula Vista, CA; Kumar Kapisthalam, MD, United Professional Center, Pasco Hernando Oncology, New Port Richey, FL; Jennifer Caskey, MD, Wheat Ridge, CO; Sayed E. Ahmend, MD, Sebring, FL; Patricia Braly, MD, Hematology and Oncology Specialties, New Orleans, LA; Donald Flemming, MD, Medical Center of Vincennes, The Bierhaus Center, Vincennes, IN; William Tester, MD, Albert Einstein Cancer Center, Philadelphia, PA; William Solomon, MD, SUNY Downstate Medical Center, Brooklyn, NY; Mark Hancock, MD, Mile Hile Oncology, Denver, CO; Youssef Hanna, MD, Huron Medical Center, Port Huron, MI; Scot Sorensen, MD, Prairie View Clinic, Lincoln, NE; and Mark Yoffe, MD, Raleigh, NC.
References
1. Birgegård G, Gascón P, Ludwig H. Evaluation of anaemia in patients with multiple myeloma and lymphoma: findings of the European Cancer Anaemia Survey. Eur J Haematol 2006;77:378–386.
2. Groopman JE, Itri LM. Chemotherapyinduced anemia in adults: incidence and treatment. J Natl Cancer Inst 1999;91:1616–1634.
3. Ludwig H, Van Belle S, Barrett-Lee P, et al. The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey defining the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 2004;40:2293–2306.
4. Williams B, Tannous R, Gupta S, et al. Chemotherapy-induced anemia in geriatric patients with non-Hodgkin’s lymphoma. Consultant Pharmacist 2002;17:131–140.
5. Miller CB, Jones RJ, Piantadosi S, Abeloff MD, Spivak JL. Decreased erythropoietin response in patients with the anemia of cancer. N Engl J Med 1990;322:1689–1692.
6. NCCN Clinical Practice Guidelines in Oncology: Cancer- and Chemotherapy-Induced Anemia, Version 2.2011. National Comprehensive Cancer Network Web site. http:// www.nccn.org/professionals/physician_gls/ pdf/anemia.pdf. Accessed June 16, 2011.
7. Rizzo JD, Brouwers M, Hurley P, et al. American Society of Hematology/American Society of Clinical Oncology clinical practice guideline update on the use of epoetin and darbepoetin in adult patients with cancer. Blood 2010;116:4045–4059.
8. Demetri GD, Kris M, Wade J, Degos L, Cella D. Quality-of-life benefit in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 1998;16:3412–3425.
9. Gabrilove JL, Cleeland CS, Livingston RB, Sarokhan B, Winer E, Einhorn LH. Clinical evaluation of once-weekly dosing of epoetin alfa in chemotherapy patients: improvements in hemoglobin and quality of life are similar to three-times-weekly dosing. J Clin Oncol 2001;19:2875–2882.
10. Glaspy J, Bukowski R, Steinberg D, Taylor C, Tchekmedyian S, Vadhan-Raj S. Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. Procrit Study Group. J Clin Oncol 1997;15:1218–1234.
11. Drüeke TB, Bárány P, Cazzola M, et al. Management of iron deficiency in renal anemia: guidelines for the optimal therapeutic approach in erythropoietin-treated patients. Clin Nephrol 1997;48:1–8.
12. Fishbane S, Frei GL, Maesaka J. Reduction in recombinant human erythropoietin doses by the use of chronic intravenous iron supplementation. Am J Kidney Dis 1995;26:41–46.
13. Van Wyck DB, Roppolo M, Martinez CO, et al. A randomized, controlled trial comparing IV iron sucrose to oral iron in anemic patients with nondialysis-dependent CKD. Kidney Int 2005;68:2846–2856.
14. Steensma DP, Sloan JA, Dakhil SR, et al. Phase III, randomized study of the effects of parenteral iron, oral iron, or no iron supplementation on the erythropoietic re sponse to darbepoetin alfa for patients with chemotherapy-associated anemia. J Clin Oncol 2011;29:97–105.
15. Auerbach M, Ballard H, Glaspy J. Clinical update: intravenous iron for anaemia. Lancet 2007;369:1502–1504.
16. Bokemeyer C, Aapro MS, Courdi A, et al. EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer: 2006 update. Eur J Cancer 2007;43:258– 270.
17. Auerbach M, Ballard H, Trout JR, et al. Intravenous iron optimizes the response to recombinant human erythropoietin in cancer patients with chemotherapy-related anemia: a multicenter, open-label, randomized trial. J Clin Oncol 2004;22:1301–1307.
18. Aronoff GR, Bennett WM, Blumenthal S, et al. Iron sucrose in hemodialysis patients: safety of replacement and maintenance regimens. Kidney Int 2004;66:1193–1198.
19. Faich G, Strobos J. Sodium ferric gluconate complex in sucrose: safer intravenous iron therapy than iron dextrans. Am J Kidney Dis 1999;33:464–470.
20. Henry DH, Dahl NV, Auerbach M, Tchekmedyian S, Laufman LR. Intravenous ferric gluconate significantly improves response to epoetin alfa versus oral iron or no iron in anemic patients with cancer receiving chemotherapy. Oncologist 2007;12:231–242.
21. Hedenus M, Birgegård G, Näsman P, et al. Addition of intravenous iron to epoetin beta increases hemoglobin response and decreases epoetin dose requirement in anemic patients with lymphoproliferative malignancies: a randomized multicenter study. Leukemia 2007;21:627–632.
22. Bastit L, Vandebroek A, Altintas S, et al. Randomized, multicenter, controlled trial comparing the efficacy and safety of darbepoetin alpha administered every 3 weeks with or without intravenous iron in patients with chemotherapy-induced anemia. J Clin Oncol 2008;26:1611–1618.
23. Pedrazzoli P, Farris A, Del Prete S, et al. Randomized trial of intravenous iron supplementation in patients with chemotherapy- related anemia without iron deficiency treated with darbepoetin alpha. J Clin Oncol 2008;26:1619–1625.
24. Chandler G, Harchowal J, Macdougall IC. Intravenous iron sucrose: establishing a safe dose. Am J Kidney Dis 2001;38:988–991.
25. Lerchenmueller C, Husseini F, Gaede B, Mossman T, Suto T, Vanderbroek A. Intravenous (IV) iron supplementation in patients with chemotherapy-induced anemia (CIA) receiving darbepoetin alfa every 3 weeks (q3w): iron parameters in a randomized controlled trial. Blood 2006;108:1552.
26. Pinter T, Mossman T, Suto T, Vansteenkiste J. Effects of intravenous iron supplementation on responses to every-3-week darbepoetin alfa by baseline hemoglobin in patients with chemotherapy-induced anemia. J Clin Oncol 2007;25(18S):9106.
27. Glaspy J, Jadeja JS, Justice G, et al. A dose-finding and safety study of novel erythropoiesis stimulating protein (NESP) for the treatment of anaemia in patients receiving multicycle chemotherapy. Br J Cancer 2001;84(suppl 1):17–23.
28. Littlewood TJ, Bajetta E, Nortier JW, Vercammen E, Rapoport B; Epoetin Alfa Study Group. Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: results of a randomized, double-blind, placebocontrolled trial. J Clin Oncol 2001;19:2865– 2874.
29. Henry DH. Supplemental iron: a key to optimizing the response of cancer-related anemia to rHuEPO? Oncologist 1998;3:275–278. 30. Ganz T. Hepcidin—a regulator of intestinal iron absorption and iron recycling by macrophages. Best Pract ClinHaematol 2005;18:171–182.
31. Ganz T. Hepcidin—a peptide hormone at the interface of innate immunity and iron metabolism. Curr Top Microbiol Immunol 2006;306:183–198.
32. Viatte L, Nicolas G, Lou DQ, et al. Chronic hepcidin induction causes hyposideremia and alters the pattern of cellular iron accumulation in hemochromatotic mice. Blood 2006;107:2952–2958.
33. Weinstein DA, Roy CN, Fleming MD, Loda MF, Wolfsdorf JI, Andrews NC. Inappropriate expression of hepcidin is associated with iron refractory anemia: implications for the anemia of chronic disease. Blood 2002;100:3776–3781.
34. Wrighting DM, Andrews NC. Interleukin- 6 induces hepcidin expression through STAT3. Blood 2006;108:3204–3209.
35. Wysowski DK, Swartz L, Borders- Hemphill BV, Goulding MR, Dormitzer C. Use of parenteral iron products and serious anaphylactic-type reactions. Am J Hematol 2010;85:650–654.
36. Bailie GR, Clark JA, Lane CE, Lane PL. Hypersensitivity reactions and deaths associated with intravenous iron preparations. Nephrol Dialysis Transplant 2005;20:1443– 1449.
37. Macdougall IC, Roche A. Administration of intravenous iron sucrose as a 2-minute push to CKD patients: a prospective evaluation of 2,297 injections. Am J Kidney Dis 2005;46:283–289.
38. Fishbane S, Ungureanu VD, Maesaka JK, Kaupke CJ, Lim V, Wish J. The safety of intravenous iron dextran in hemodialysis patients. Am J Kidney Dis 1996;28:529–534.
Lowell B. Anthony, MD,1 Nashat Y. Gabrail, MD,2 Hassan Ghazal, MD,3, Donald V. Woytowitz, MD,4 Marshall S. Flam, MD,5 Anibal Drelichman, MD,6, David M. Loesch, MD,7, Demi A. Niforos, MS,8, and Antoinette Mangione, MD, PharmD9; for the Iron Sucrose Study Group*
1 Stanley S. Scott Cancer Center, Louisiana State University Health Sciences Center, New Orleans, LA; 2 Nashat Cancer Center, Canton, OH; 3 Kentucky Cancer Clinic, Hazard, KY; 4 Florida Cancer Specialists, Fort Myers, FL; 5 Hematology/Oncology Group of Fresno, Fresno, CA; 6 Newland Medical Associates, Southfield, MI; 7 Oncology/Hematology Associates, Indianapolis, IN; 8 AAI Pharma, Inc., Natick, MA; and 9 Luitpold Pharmaceuticals/American Regent, Inc., Norristown, PA
Manuscript received January 2, 2011; accepted June 16, 2011.
This work was presented at the 43rd Annual Meeting of the American Society of Clinical Oncology; June 1–5, 2007 in Chicago, IL, and was supported by Luitpold Pharmaceuticals/American Regent, Inc., Shirley, NY.
Correspondence to: Lowell B. Anthony, MD, LSUHSC New Orleans, Ochsner Kenner Medical Center, 200 West Esplanade, Kenner, LA 70065; e-mail: [email protected].
Conflicts of interest: Ms. Niforos was a fulltime salaried employee of AAI Pharma, Inc., contracted to perform all biostatistical services for the clinical trial. Dr. Mangione was a fulltime salaried employee of the trial sponsor, Luitpold Pharmaceuticals/American Regent, Inc. Drs. Anthony, Gabrail, Ghazal, Woytowitz, Flam, Drelichman, and Loesch have nothing to disclose.
Mild-to-moderate anemia occurs in up to 75% of cancer patients undergoing either single- or multimodality therapy and may contribute to an increased morbidity and reduced quality of life (QOL).1–4 This form of anemia resembles anemia of chronic disease, with a blunted erythropoietin response and inadequate erythropoietin production.5 Increasing hemoglobin (Hgb) concentrations and reducing red blood cell (RBC) transfusions while improving QOL and tolerance to cancer therapies are the treatment-related goals.
Intravenous (IV) iron is commonly administered with ESAs in CKD-associated anemia.12,13 Most studies regarding IV iron replacement in cancer and/or chemotherapy-induced anemia (CCIA) are positive, with one exception: Steensma et al14 reported no benefit in adding IV ferric gluconate to an ESA in a phase III randomized trial in which an oral placebo and iron were used as comparators. Practice guidelines are inconsistent, as the National Comprehensive Cancer Network (NCCN) recommends the IV route when iron is prescribed,6 and the American Society of Hematology/ American Society of Clinical Oncology considers the evidence insufficient to support routine IV iron use.15,16 Auerbach et al17 demonstrated that IV iron dextran results in a greater Hgb level increase than oral iron in ESAtreated patients. Approved formulations of IV iron in the United States include iron dextran, iron sucrose, and ferric gluconate, with the majority of published data with iron dextran.15,18,19 However, the iron dextrans have black-box warnings, and test doses are recommended. Henry et al20 reported that IV ferric gluconate significantly increased Hgb response when compared with oral iron or no iron and was well tolerated in CCIA.
Early work with IV iron sucrose includes a trial evaluating 67 lymphoma patients randomized between ESA or ESA with IV iron sucrose.21 Despite adequate bone marrow iron stores, the Hgb response was greater (91% vs 54%) and the time to reach a Hgb level > 2 g/dL was less (6 vs 12 weeks) in the IV iron-treated group.21 Another trial randomized 398 CCIA patients between fixed IV iron doses (mean weekly dose, 64.8 mg) with ESA versus standard practice (2% received IV iron).22 IV iron resulted in a trend toward a higher ferritin level, but transferrin saturation (TSAT) remained similar between the two groups.22 A study in patients with noniron-deficient anemic solid tumors receiving chemotherapy also demonstrated an increase in hemoglobin levels statistically favoring the darbepoetin alfa (Aranesp)/iron group.23 As additional information is needed, this study was performed to determine whether IV iron sucrose combined with ESA increases Hgb levels in CCIA patients who have been previously treated with an ESA.
Patients and methods
Patient eligibility
his was an open-label, phase III, randomized, institutional review board-approved, multicenter study at 56 US centers. After signing informed consent, patients ≥ 18 years of age with a histologic diagnosis of cancer (acute leukemia or myeloproliferative syndrome excluded) receiving ongoing or planned chemotherapy, with a Hgb level ≤ 10.0 g/dL, body weight > 50 kg, and a Karnofsky performance status of ≥ 60%, were eligible. Patients were excluded if they had iron depletion, active infection, myelophthisic bone marrow (except for hematologic malignancy), hypoplastic bone marrow, uncontrolled hypertension, bleeding, or planned surgery. To ensure a stable baseline Hgb value, no IV iron within 2 months of consent or RBC transfusions within 3 weeks of randomization were allowed.
Treatment
After 8 weeks of fixed ESA doses in stage 1, patients were classified as either ESA responders (≥ 1 g/dL Hgb level increase from baseline) or nonresponders (< 1 g/dL Hgb level increase from baseline), with each group separately randomized centrally using block randomization to receive either IV iron sucrose or no iron treatment (Figure 1). At the time of randomization (beginning of stage 2), patients were stratified according to malignancy type (solid tumor vs hematologic) and Hgb level (< 12 g/dL vs ≥ 12 g/dL for ESA responders; < 9.5 g/dL vs ≥ 9.5 g/dL for ESA nonresponders).
The calculated dose of the study drug (iron sucrose [Venofer]; 7 mg/kg up to 500 mg maximum) was added to 500 mL of normal or half-normal saline and administered IV over 4 hours.24 Patients randomized to receive iron sucrose were scheduled to receive up to three infusions at 1- to 3-week intervals during the first 9 weeks of stage 2, with the first dose administered as soon as possible after randomization. The last dose of ESA was given on or before week 12 of stage 2.
Outcome measures
The primary endpoint for efficacy was the change from baseline (end of stage 1) to the maximum Hgb level achieved during stage 2 in patients who responded to ESA. Major secondary endpoints included changes in Hgb levels when iron sucrose was added to ESA nonresponders as well as the percentage of all randomized patients with Hgb level increases > 1 g/dL, > 2 g/dL, and > 3 g/dL; changes in Hgb levels and iron indices from baseline at each visit; and changes in the 13-item Functional Assessment of Chronic Illness Therapy (FACIT) fatigue scale. Hgb levels were obtained weekly, and iron indices were measured every 3 weeks. The FACIT fatigue scale was measured during stage 1 at consent, weeks 4, and 8 and during stage 2 weeks 3, 6, 9, and at the end of the study.
Adverse events were recorded hourly during iron sucrose administration and from the day of randomization through study completion or 30 days following the last dose of study drug, whichever was later. Investigators provided the date of onset, severity, relationship, date of resolution, action taken, and adverse event outcome. Adverse drug events were events considered by the investigator to be possibly, probably, or definitely related to the study drug.
Statistical method
The sample size was based on the hypothesis that iron-treated ESA responders (group A) would have a 1.0 g/dL or higher mean increase in Hgb levels than would ESA responders who did not receive iron (group B). The standard deviation (SD) of the difference was assumed to be ≤ 1.5 g/dL. Targeting a 1.0 g/dL change in Hgb level to be significant, 49 patients/ group were required (alpha = 0.05; beta = 0.10). Assuming that the ESA response rate in stage I was at least 40% and that the stage I and stage 2 dropout rates were no more than 10% and 25%, respectively, 325 patients were the targeted number for stage I enrollment, with adjustments made by monitoring the stage I response rate.
The intent-to-treat (ITT) population included patients randomized into stage 2 based on actual treatment. The evaluable population included ITT patients who completed at least 10 weeks of stage 2 or who had interventions (RBC transfusions or nonstudy iron) prior to week 10.
Continuous variables were assessed using analysis of covariance and t-tests. Ordinal responses were analyzed with the Fisher’s exact test and Cochran-Mantel-Haenszel statistics. Changes from baseline to each visit for all FACIT scores were assessed for treatment groups with the unpaired two-sample t-test.
Results
Patient disposition and demographics
Of the 375 patients enrolled during the run-in stage 1 period (between July 2003 and October 2005), 132 patients discontinued treatment (the most common reasons were a required intervention [50], withdrawn consent [23], and adverse events [17]). Fourteen patients completed stage 1 but did not enter stage 2. Figure 2 shows the numbers of patients who were randomly assigned to the two treatment groups and were evaluated for safety and efficacy as well as reasons for study discontinuation. Table 1 shows the patient numbers assigned to the various treatment groups (A to D) based on ESA response in stage I and the study population; it also demonstrates the similar baseline demographic characteristics between the treatment groups. At baseline (ie, prior to randomization), there were no statistically significant differences in Hgb level, TSAT, and ferritin level between the ESA responders (A vs B) and nonresponders (C vs D).
Efficacy of iron sucrose
Mean maximum improvement in Hgb levels (Table 2). Among ESA responders (groups A and B), a statistically significantly greater mean maximum Hgb level increase was observed among patients who received iron sucrose (group A) than among those who did not (group B), achieving the primary endpoint (ITT, P = 0.004; evaluable, P = 0.008). A statistically significant greater increase in the mean maximum Hgb level was observed following iron sucrose (groups A and C) when compared with no iron treatment (groups B and D), regardless of prior ESA response. In the ESA nonresponder group, a significant increase (P = 0.027) in the mean maximum Hgb level was observed between those who received iron sucrose (group C) and those who did not (group D) in the ITT population; a statistical difference was not seen in the evaluable population (P = 0.082).
With regard to tumor subtypes, breast cancer and other tumor types, but not lung cancer, were associated with statistically significant increases in maximum Hgb levels following iron sucrose, regardless of prior ESA response.
Absolute increases in Hgb levels (Table 2). A greater proportion of patients assigned to IV iron sucrose achieved a ≥ 2 g/dL and ≥ 3 g/dL increase in Hgb level during the study than did those who did not receive iron. These differences were statistically significant for all the groups except for the evaluable ≥ 3 g/dL nonresponder group. The only statistically significant difference in the proportion achieving a ≥ 1 g/dL Hgb level increase occurred in the ESA nonresponder groups. In addition, baseline hematologic characteristics and iron indices did not predict the efficacy of IV iron treatment (as defined by a > 1 g/dL or > 2 g/dL increase in Hgb level). In the IV iron sucrose-treated group, there was no statistical difference in these baseline characteristics in the patients who demonstrated a > 1 g/dL (data not shown) or a > 2 g/dL treatment response to IV iron.
Changes from baseline in Hgb and ferritin levels and in TSAT. Figure 3 summarizes the Hgb level, ferritin level, and TSAT responses by study visit after IV iron sucrose compared with no iron in the ITT population. Between treatment groups, statistically significant differences (P < 0.05) were present by weeks 7, 3, and 13 for Hgb level, ferritin level, and TSAT, respectively. At the end of the study, week 13, the mean Hgb level increase from baseline was 2.3 g/dL versus 1.2 g/dL (P < 0.002), the mean ferritin level increase from baseline was 419 ng/mL versus a decrease of 50 ng/mL (P < 0.001), and the mean TSAT increase from baseline was 8.8% versus 0.2% (P < 0.005) in the iron sucrose versus no iron group.
Changes in fatigue levels (FACIT fatigue scale). There was a statistically significant decrease in the level of fatigue at the end of the study compared with at baseline (end of stage 1) in the iron sucrose-treated patients in the ITT but not in the evaluable population (–3.3 iron sucrose/–2.1 no iron, P = 0.022 ITT; –3.0 iron sucrose/–1.7 no iron, P = 0.058 evaluable population). No significant decrease in the level of fatigue was experienced by the patients who received no iron. There were no statistically significant differences between the groups in changes from baseline at each visit..
Safety of iron sucrose
Extent of exposure. In the ITT population, the mean per patient total dose of iron sucrose administered was 1,123 (SD, 402) mg in group A (responders) and 1,113 (SD, 387) mg in group C (nonresponders).
Adverse drug events (ADEs). All safety analyses were performed using the ITT population. Serious ADEs were experienced by three patients in the iron sucrose group (chest pain, hypersensitivity, and hypotension, one patient each) and by no patients in the ESA-only group. One ESA-only group patient (arthralgia) and four iron sucrose patients (hypersensitivity; abdominal pain; arthralgia and muscle cramps; myalgia, nausea, and vomiting) were prematurely discontinued from the study drug due to the occurrence of an ADE.
At least one ADE was experienced by 37.4% of the patients in the iron sucrose group and 0.8% in the control group. The most common (³ 5%) ADEs were nausea (8.1%), dysgeusia (8.1%), back pain (6.1%), arthralgia (6.1%), muscle cramp (6.1%), and peripheral edema (5.1%). Within the ESA-only group, the only ADE reported was hypertension (one subject, 0.8%).
Eleven grade 3 (National Institutes of Health/National Cancer Institute– Common Terminology Criteria, version 2.0) ADEs occurred in iron sucrose-treated patients and included nausea (2.0%), hypotension (2.0%), abdominal pain (1.0%), chest pain (1.0%), hypersensitivity (1.0%), arthralgia (1.0%), dizziness (1.0%), dyspnea (1.0%), and hypertension (1.0%). A serious grade 3 hypotensive event occurred in a 49-year-old woman weighing 50 kg who experienced dizziness, nausea, vomiting, and transient hypotension (110/60 mm Hg to 70/40 mm Hg) after her first iron sucrose dose of 375 mg. Ninety minutes later, following IV steroids, iron sucrose was restarted and the hypotension recurred. The patient received two subsequent lower iron sucrose doses (200 mg over 4 hours), with no further adverse reactions.
Deaths and thrombotic events. These events are summarized in Table 3. None of these events was judged by the investigators to be related to the study drug.
Laboratory results. Statistically greater mean increases in ferritin levels, TSAT, Hgb levels, hematocrit, mean corpuscular hemoglobin, mean corpuscular volume, and monocytes oc curred in the iron sucrose-treated group. There were no significant differences between treatment groups in clinical chemistry safety laboratory results.
Discussion
This study is the first to evaluate IV iron in CCIA patients who have received prior ESA therapy. IV iron sucrose administered with ESAs significantly increased Hgb levels in CCIA patients. Prior ESA response did not predict Hgb level response to iron sucrose, as benefit was demonstrated in both ESA responders and nonresponders. Baseline hematologic/ iron indices also did not predict IV iron responsiveness, as these characteristics were similar in IV iron responders and nonresponders. Improvement in QOL, as measured by fatigue levels at study completion, was also observed after IV iron but not in the no iron group. IV iron studies are commonly open-label because of the difficulty in blinding iron’s viscous dark-colored solution.
This study design limits the significance of QOL measurements in IV iron studies, where primary endpoints are typically objective measurements. Even though transfusion rates were lower in the IV iron groups (5.1% in groups A and C [A = 1.7%; C = 10%]) than in the no iron groups (10.4% in groups B and D [B = 2.6%; D = 22.9%]), this difference was not statistically significant (Fisher’s exact test, P = 0.215). Our findings support the prior observations that IV iron replacement in combination with ESAs effectively increases Hgb levels and is safe.17,20,21,25,26
Combining IV iron with ESA increases the Hgb level response and may either shorten the time to response and/or decrease the ESA requirement. Approximately 30%–50% of patients are nonresponders after 12–24 weeks of ESA therapy.8,9,17,27,28 Iron deficiency may be a major factor accounting for ESA resistance. Decreased ESA responsiveness in the dialysis population can be corrected by providing adequate iron supplementation. 11,18 Also, ESA nonresponders may become responders with IV iron replacement while continuing the ESA. ESA treatment in responders can produce a functional iron deficiency, because the ESA produces a rapid initiation of erythropoiesis. Inducing functional iron deficiency with ESA therapy implies that the iron supply to the erythron may be the rate-limiting step in erythropoiesis, and the IV iron dose may be important.25 As ESA responders and nonresponders experienced improvement in Hgb levels with IV iron therapy in this trial, IV iron supplementation may be required to achieve and/or maintain a response to ESA therapy.
Iron available for erythropoiesis is derived from the balance between dietary sources and that in the usable pool within the reticuloendothelial system.29 ESA therapy can result in RBC production that exceeds the rate of iron mobilization, even with adequate iron stores. Inflammatory cytokines may also hinder the release of stored iron from macrophages by inducing hepcidin and thus further contribute to an inadequate rate of RBC production.30–34
Of note, baseline ferritin levels were higher in the ESA nonresponders (groups C and D) than in the ESA responders (groups A and B), although these differences were not statistically significant. This finding may be consistent with elevated inflammatory cytokines impairing the availability of iron, leading to a failed ESA response. ESA resistance is multifactorial, with these factors contributing to the rapid depletion of the usable iron pool, thus blunting the ESA response. Identifying factors that allow for maximizing ESA therapy in CCIA patients may result in greater ESA efficiency. The IV route of iron replacement is superior to oral administration and accounts for one of these variables.17,21,25,26
Safely administering IV iron is an important factor that influences the choice of iron preparations. In the United States, the only IV iron indicated for iron deficiency anemia is iron dextran. The risk of allergic reactions and the need for test doses may account for practitioners limiting the use of iron dextran, despite a compelling medical need for rapid, reliable, and safe replenishment of body iron in populations such as those with CKD35–37 and CCIA. The non–dextran- containing IV irons (iron sucrose, ferric gluconate) are currently only FDA approved for CKD indications at doses of 100–200 mg over 2–5 minutes or up to 400 mg over 2.5 hours for iron sucrose and only 125 mg over 10 minutes for ferric gluconate. 18,19
This study supports other findings that IV iron sucrose is generally well tolerated at doses of 7 mg/kg, up to a maximum of 500 mg over 4 hours, in CCIA. Caution should be exercised, however, especially in patients with a lower body weight. This concern is supported by a study of iron sucrose in nondialysis CKD, where hypotension occurred in two patients < 65 kg after 500 mg doses were administered over 4 hours.38
Conclusion
This study’s primary objective was to determine whether prior response to ESA treatment would influence response to IV iron, not to detect differences between functional and absolute iron deficiency. Our findings support that administration of IV iron while continuing ESA treatment may correct functional, as well as absolute, iron deficiency in CCIA. Baseline iron indices did not predict responsiveness to iron sucrose. Without additional data identifying predictors of ESA responsiveness in CCIA, a more proactive approach that includes IV iron may be warranted, as in CKDrelated anemia. As a better understanding of functional iron deficiency evolves, it is becoming apparent that IV iron is important to optimize the response to ESAs for CCIA. Additional studies are needed to understand the mechanisms responsible for functional iron deficiency in CCIA and to assist in identifying the optimal IV iron administration schedule.
Acknowledgments: The authors wish to thank the study coordinators; the patients at each of the participating centers; and Drs. Perry Rigby and Robert Means, for reviewing the manuscript.
*Additional members of the Iron Sucrose Study Group include Ali Ben-Jacob, MD, Cache Valley Cancer Treatment and Research Clinic, Inc., Logan, UT; Amol Rakkar, MD, Hope Center, Terre Haute, IN; Philip Chatham, MD, Granada Hills, CA; Ahmed Maqbool, MD, Welborn Clinic, Research Center, Evansville, IN; Timothy Pluard, MD, Washington University, Medical Oncology, St. Peters, MO; Nafisa Burhani, MD, Joliet Oncology- Hematology Associates, LTD, Joliet, IL; David Henry, MD, Pennsylvania Hematology and Oncology Associates, Philadelphia, PA; David Watkins, MD, Allison Cancer Center, Midland, TX; Howard Ozer, MD, University of Oklahoma Health Science Center-Hematology Oncology Section, Oklahoma City, OK; Leo Orr, MD, Leo E. Orr, Inc., Los Angeles, CA; Billy Clowney, MD, Santee Hematology Oncology, Sumter, SC, Rene Rothestein-Rubin, MD, Rittenhouse Hematology/ Oncology, Philadelphia, PA; Peter Eisenberg, MD, California Cancer Care, Greenbrae, CA; Rosalba Rodriguez, MD, Chula Vista, CA; Kumar Kapisthalam, MD, United Professional Center, Pasco Hernando Oncology, New Port Richey, FL; Jennifer Caskey, MD, Wheat Ridge, CO; Sayed E. Ahmend, MD, Sebring, FL; Patricia Braly, MD, Hematology and Oncology Specialties, New Orleans, LA; Donald Flemming, MD, Medical Center of Vincennes, The Bierhaus Center, Vincennes, IN; William Tester, MD, Albert Einstein Cancer Center, Philadelphia, PA; William Solomon, MD, SUNY Downstate Medical Center, Brooklyn, NY; Mark Hancock, MD, Mile Hile Oncology, Denver, CO; Youssef Hanna, MD, Huron Medical Center, Port Huron, MI; Scot Sorensen, MD, Prairie View Clinic, Lincoln, NE; and Mark Yoffe, MD, Raleigh, NC.
References
1. Birgegård G, Gascón P, Ludwig H. Evaluation of anaemia in patients with multiple myeloma and lymphoma: findings of the European Cancer Anaemia Survey. Eur J Haematol 2006;77:378–386.
2. Groopman JE, Itri LM. Chemotherapyinduced anemia in adults: incidence and treatment. J Natl Cancer Inst 1999;91:1616–1634.
3. Ludwig H, Van Belle S, Barrett-Lee P, et al. The European Cancer Anaemia Survey (ECAS): a large, multinational, prospective survey defining the prevalence, incidence, and treatment of anaemia in cancer patients. Eur J Cancer 2004;40:2293–2306.
4. Williams B, Tannous R, Gupta S, et al. Chemotherapy-induced anemia in geriatric patients with non-Hodgkin’s lymphoma. Consultant Pharmacist 2002;17:131–140.
5. Miller CB, Jones RJ, Piantadosi S, Abeloff MD, Spivak JL. Decreased erythropoietin response in patients with the anemia of cancer. N Engl J Med 1990;322:1689–1692.
6. NCCN Clinical Practice Guidelines in Oncology: Cancer- and Chemotherapy-Induced Anemia, Version 2.2011. National Comprehensive Cancer Network Web site. http:// www.nccn.org/professionals/physician_gls/ pdf/anemia.pdf. Accessed June 16, 2011.
7. Rizzo JD, Brouwers M, Hurley P, et al. American Society of Hematology/American Society of Clinical Oncology clinical practice guideline update on the use of epoetin and darbepoetin in adult patients with cancer. Blood 2010;116:4045–4059.
8. Demetri GD, Kris M, Wade J, Degos L, Cella D. Quality-of-life benefit in chemotherapy patients treated with epoetin alfa is independent of disease response or tumor type: results from a prospective community oncology study. Procrit Study Group. J Clin Oncol 1998;16:3412–3425.
9. Gabrilove JL, Cleeland CS, Livingston RB, Sarokhan B, Winer E, Einhorn LH. Clinical evaluation of once-weekly dosing of epoetin alfa in chemotherapy patients: improvements in hemoglobin and quality of life are similar to three-times-weekly dosing. J Clin Oncol 2001;19:2875–2882.
10. Glaspy J, Bukowski R, Steinberg D, Taylor C, Tchekmedyian S, Vadhan-Raj S. Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice. Procrit Study Group. J Clin Oncol 1997;15:1218–1234.
11. Drüeke TB, Bárány P, Cazzola M, et al. Management of iron deficiency in renal anemia: guidelines for the optimal therapeutic approach in erythropoietin-treated patients. Clin Nephrol 1997;48:1–8.
12. Fishbane S, Frei GL, Maesaka J. Reduction in recombinant human erythropoietin doses by the use of chronic intravenous iron supplementation. Am J Kidney Dis 1995;26:41–46.
13. Van Wyck DB, Roppolo M, Martinez CO, et al. A randomized, controlled trial comparing IV iron sucrose to oral iron in anemic patients with nondialysis-dependent CKD. Kidney Int 2005;68:2846–2856.
14. Steensma DP, Sloan JA, Dakhil SR, et al. Phase III, randomized study of the effects of parenteral iron, oral iron, or no iron supplementation on the erythropoietic re sponse to darbepoetin alfa for patients with chemotherapy-associated anemia. J Clin Oncol 2011;29:97–105.
15. Auerbach M, Ballard H, Glaspy J. Clinical update: intravenous iron for anaemia. Lancet 2007;369:1502–1504.
16. Bokemeyer C, Aapro MS, Courdi A, et al. EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer: 2006 update. Eur J Cancer 2007;43:258– 270.
17. Auerbach M, Ballard H, Trout JR, et al. Intravenous iron optimizes the response to recombinant human erythropoietin in cancer patients with chemotherapy-related anemia: a multicenter, open-label, randomized trial. J Clin Oncol 2004;22:1301–1307.
18. Aronoff GR, Bennett WM, Blumenthal S, et al. Iron sucrose in hemodialysis patients: safety of replacement and maintenance regimens. Kidney Int 2004;66:1193–1198.
19. Faich G, Strobos J. Sodium ferric gluconate complex in sucrose: safer intravenous iron therapy than iron dextrans. Am J Kidney Dis 1999;33:464–470.
20. Henry DH, Dahl NV, Auerbach M, Tchekmedyian S, Laufman LR. Intravenous ferric gluconate significantly improves response to epoetin alfa versus oral iron or no iron in anemic patients with cancer receiving chemotherapy. Oncologist 2007;12:231–242.
21. Hedenus M, Birgegård G, Näsman P, et al. Addition of intravenous iron to epoetin beta increases hemoglobin response and decreases epoetin dose requirement in anemic patients with lymphoproliferative malignancies: a randomized multicenter study. Leukemia 2007;21:627–632.
22. Bastit L, Vandebroek A, Altintas S, et al. Randomized, multicenter, controlled trial comparing the efficacy and safety of darbepoetin alpha administered every 3 weeks with or without intravenous iron in patients with chemotherapy-induced anemia. J Clin Oncol 2008;26:1611–1618.
23. Pedrazzoli P, Farris A, Del Prete S, et al. Randomized trial of intravenous iron supplementation in patients with chemotherapy- related anemia without iron deficiency treated with darbepoetin alpha. J Clin Oncol 2008;26:1619–1625.
24. Chandler G, Harchowal J, Macdougall IC. Intravenous iron sucrose: establishing a safe dose. Am J Kidney Dis 2001;38:988–991.
25. Lerchenmueller C, Husseini F, Gaede B, Mossman T, Suto T, Vanderbroek A. Intravenous (IV) iron supplementation in patients with chemotherapy-induced anemia (CIA) receiving darbepoetin alfa every 3 weeks (q3w): iron parameters in a randomized controlled trial. Blood 2006;108:1552.
26. Pinter T, Mossman T, Suto T, Vansteenkiste J. Effects of intravenous iron supplementation on responses to every-3-week darbepoetin alfa by baseline hemoglobin in patients with chemotherapy-induced anemia. J Clin Oncol 2007;25(18S):9106.
27. Glaspy J, Jadeja JS, Justice G, et al. A dose-finding and safety study of novel erythropoiesis stimulating protein (NESP) for the treatment of anaemia in patients receiving multicycle chemotherapy. Br J Cancer 2001;84(suppl 1):17–23.
28. Littlewood TJ, Bajetta E, Nortier JW, Vercammen E, Rapoport B; Epoetin Alfa Study Group. Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: results of a randomized, double-blind, placebocontrolled trial. J Clin Oncol 2001;19:2865– 2874.
29. Henry DH. Supplemental iron: a key to optimizing the response of cancer-related anemia to rHuEPO? Oncologist 1998;3:275–278. 30. Ganz T. Hepcidin—a regulator of intestinal iron absorption and iron recycling by macrophages. Best Pract ClinHaematol 2005;18:171–182.
31. Ganz T. Hepcidin—a peptide hormone at the interface of innate immunity and iron metabolism. Curr Top Microbiol Immunol 2006;306:183–198.
32. Viatte L, Nicolas G, Lou DQ, et al. Chronic hepcidin induction causes hyposideremia and alters the pattern of cellular iron accumulation in hemochromatotic mice. Blood 2006;107:2952–2958.
33. Weinstein DA, Roy CN, Fleming MD, Loda MF, Wolfsdorf JI, Andrews NC. Inappropriate expression of hepcidin is associated with iron refractory anemia: implications for the anemia of chronic disease. Blood 2002;100:3776–3781.
34. Wrighting DM, Andrews NC. Interleukin- 6 induces hepcidin expression through STAT3. Blood 2006;108:3204–3209.
35. Wysowski DK, Swartz L, Borders- Hemphill BV, Goulding MR, Dormitzer C. Use of parenteral iron products and serious anaphylactic-type reactions. Am J Hematol 2010;85:650–654.
36. Bailie GR, Clark JA, Lane CE, Lane PL. Hypersensitivity reactions and deaths associated with intravenous iron preparations. Nephrol Dialysis Transplant 2005;20:1443– 1449.
37. Macdougall IC, Roche A. Administration of intravenous iron sucrose as a 2-minute push to CKD patients: a prospective evaluation of 2,297 injections. Am J Kidney Dis 2005;46:283–289.
38. Fishbane S, Ungureanu VD, Maesaka JK, Kaupke CJ, Lim V, Wish J. The safety of intravenous iron dextran in hemodialysis patients. Am J Kidney Dis 1996;28:529–534.
Direct-to-consumer genetic testing: helpful, harmful, or pure entertainment?
The genomic revolution has the potential to change the practice of medicine and healthcare. Advances in technology are driving down the cost of genetic testing, the public is more aware of and interested in such testing and its uses, and genetic testing companies are now offering direct-to-consumer (DTC) testing via kits sold on the Internet. Some argue that these changes make genetic testing more accessible to the general public and medical community at large and that these changes are critical to the full integration of genomic technology into society. However, many of the DTC tests offered are of questionable accuracy and utility and combine tests that provide mundane information about genetic traits with those for serious genetic conditions with life-threatening implications for the patient and the entire family. We will examine both sides of the issues surrounding DTC genetic testing, focusing on important considerations for clinicians using case vignettes to illustrate key issues.
Is DTC testing worthwhile for your patients?
Point: The testing and result interpretations provided are not necessarily clinically accurate or helpful.
Numerous studies have shown that average patients and clinicians are not well versed on even the most basic elements of genetic testing and result interpretation. 1 It is even less likely that consumers and clinicians will understand the subtle differences between DTC companies and tests—particularly genomic profiling based on single nucleotide polymorphisms (SNPs)—which are offered by most DTC companies. The difference between SNPbased testing and clinical single-gene testing with regard to predictive value is a confusing concept that typical consumers and clinicians without adequate training may not understand. For most of the SNP-based genomic profiles offered DTC, there is a lack of consensus among the scientific and medical communities regarding test interpretation.
The positive and negative predictive values, along with the discriminatory accuracy of these tests, are in sharp contrast to those associated with testing for mutations in high-penetrance Mendelian genes. Discriminatory accuracy refers to the ability of a test to determine who will or will not develop the disease or condition and is a balance between sensitivity and specificity. For example, those who test positive for a mutation in the Huntington's disease gene are certain to develop the disease, whereas those who test negative are unlikely to do so. Similarly, in cancer genetics, those with a mutation in the APC gene (causative of familial adenomatous polyposis) will develop colon polyps and colon cancer without appropriate intervention.
With these Mendelian conditions, the relative risk of disease (penetrance) is large (often 10-fold or greater), and the proportion of disease attribestimated by these companies fail to account for known family history or other risk factors. An investigation in July 2010 by the US General Accounting Office concluded that such tests were “misleading and of little or no practical use.”14 The risk profiles provided by such companies have been described as having “no predictive value” and may falsely alarm or reassure consumers.15
Counterpoint: Individuals have the right to explore their own genetic makeup through these emerging tests and may deem testing worthwhile. Some individuals may learn important information they would otherwise not have learned.
One argument to be made is that genetic technology exists and individuals have the right to explore the testing available, regardless of its accuracy and utility. Many DTC sites do state that their testing is intended for entertainment purposes only, although this statement is often in the fine print (see next point). It is possible that DTC testing has made the public more aware of genetic and genomic testing, therefore encouraging the medical profession to keep up with this technology as it pertains to one's area of expertise. Although some may see this as an additional burden on the primary care provider, it can also be viewed as an opportunity to increase awareness and education on the part of both the consumer and the professional. Most current genomic technology was not developed when most of today’s physicians went to medical school, and even most new graduates are not well versed in genetic testing. DTC genetic testing could encourage the medical community to increase its understanding of this new emerging technology.
Although most of the DTC testing offered is of questionable utility, some panels do include important genetic information, like the three common Jewish BRCA mutations. It may be considered ethically questionable to offer testing for hereditary breast cancer and ovarian cancer without full informed consent and adequate result disclosure. However, one can also argue that individuals will gain potentially lifesaving information they might never have uncovered without such testing.
Proponents of DTC genomic profiling suggest that the clinical usefulness of SNP-based testing will improve based on emerging data that analyze the interaction between genetic polymorphisms and environmental factors in predicting disease risk. For example, preliminary studies evaluating SNPs associated with colorectal cancer have found that relative risks are increased when combined with environmental factors, such as high intake of processed meats.16,17 Despite these early findings, it remains to be seen whether the complexity of gene-environment and gene-gene interactions can be elucidated to the extent that will be required to yield highly predictive DTC tests.
Is DTC advertising accurate, fair, and helpful?
Point: DTC advertising overemphasizes the benefits of testing and does not adequately describe the limitations and risks.
Unlike pharmaceutical advertising, which is tightly regulated and requires companies to discuss associated risks, oversight of genetic testing advertisements is limited.18 This can lead to skewing of information that overemphasizes the benefits of genetic testing and does not adequately define the risks and limitations.19,20 One recent study demonstrated that when individuals are provided with accurate information regarding the actual risks and limitations of online DTC genetic testing, they have reduced intentions to obtain such tests and less positive views of Internet-based testing. 19 Furthermore, an investigation of DTC advertising revealed “egregious examples” of deceptive marketing, such as claims that consumers’ DNA could be used to create personalized supplements to cure disease or that results of such DNA tests could be used to predict the athletic potential of their children.14
Many DTC companies have a clause indicating that their testing is intended for research, educational, or entertainment purposes only, and not for diagnostic or medical use. However, this clause is embedded deep into the Web site and surrounded by text like “disease risk prediction,” “predicted response to drugs,” and a long list of serious medical conditions that appear to be anything but entertainment- based. Together with mundane traits, such as bitter taste perception, freckling, earwax type, and hair thickness, consumers are tested for susceptibility to amyotrophic lateral sclerosis, bipolar disease, heroin addiction, and cancer predisposition. This strange combination of offerings makes it difficult to determine whether this testing is truly entertainmentbased or a serious predictive testing panel requiring careful interpretation by a medical professional.
Complicating these issues even further, several DTC testing companies do not fully explain which SNPs or gene mutations are evaluated in their specific disease risk profiles. Therefore, the average woman may think she has had comprehensive sequencing for BRCA1 and BRCA2 because her “breast cancer risk” has been assessed, when in fact she has been tested for a panel of low-risk SNPs and only the three common Jewish BRCA mutations. For a non-Jewish patient, these results would be virtually meaningless in terms of assessing hereditary risk.
Even well-versed genetics professionals may not be able to decipher the implications of tests offered by some companies. Recently, a cancer genetics professional reported that she reviewed a multigene-disease panel offered by one DTC genetic testing company that reportedly included testing for MYH-associated polyposis. When she discussed this test with the company, she discovered that it provided analysis for only one of the two common MYH mutations in Caucasians (unpublished, personal communication). These important omissions could lead both patients and clinicians to think that comprehensive testing has been performed when, in fact, it has not.
Counterpoint: DTC advertising has increased physician and patient awareness of genetic testing and technology.
As consumers become more sophisticated in their medical knowledge base, many are turning to the Internet for healthcare information before, or instead of, addressing their concerns with a medical professional. Although medical information on the Internet is of varied accuracy, obtaining medical information online is a trend that is here to stay, and genetic testing is no exception. The DTC genetic testing companies have exploited this thirst for medical information and have developed educational information on basic “Genetics 101” concepts for the lay public. For example, 23andMe, a DTC genetic testing company, describes genomic testing through an animated video clip using cartoon characters on its Web site. This new educational delivery system not only helps the genetic testing company sell their product but also helps the consumer learn more about genomic technology. In response to DTC genetic testing, many medical groups have launched educational campaigns to help bring clinicians up-to-date on the genetic testing their patients may be pursuing and bringing to their office.
Many professional medical organizations have addressed this gap in knowledge and have continually offered additional training regarding regarding the implications of genomic and genetic technologies in their subspeciality. Oncology has been at the forefront of this movement, offering training to oncologists through the American Society of Clinical Oncology to expand the knowledge base on genetic and genomic testing.
Is DTC genetic testing without the guidance of a qualified professional doing more harm than good?
Point: For consumers to have DTC testing without the guidance of a qualified professional to interpret complex information is not in their best interest.
Many DTC genetic testing companies allow consumers to order their own tests and provide results, risk calculations, and interpretations to patients without any contact with a knowledgeable clinician. As illustrated in the BRCA vignette below, the information provided to patients with their results is often not only inaccurate and irresponsible but may be harmful or deadly to the patient and at-risk family members.
An undercover investigation of DTC testing companies revealed poor to nonexistent medical advice from supposed consult experts, in addition to possible privacy concerns.14 Although some DTC genetic testing companies offer genetic counseling as part of their services, the consumer needs to be aware that medical advice from an individual with a financial interest in the product or company may be biased.21,22 Much as patients would not contact a pharmaceutical company to ask whether they should take a drug it supplies, patients should be wary of receiving advice regarding genetic testing from a company that stands to profit directly from their decision.
The complex information provided by these DTC genetic tests for disease risk is at best difficult to interpret, and there is concern that patients and physicians may make important healthcare decisions based on this information without a clear understanding of the test limitations. 18,23 A 2008 survey of physicians found that among those aware of personal genomic testing, 15% had a patient who had brought in test results for discussion, and more than 50% indicated that such tests would influence their decisions regarding patient care.24 Similarly, McGuire et al found that a majority (78%) of consumers report that they would take results of DTC genetic testing to their healthcare providers.25 These findings lend credence to the concerns expressed by several authors that DTC genetic testing may lead to overuse of the healthcare system, including unnecessary clinician visits and follow-up evaluations or tests.18,22,25–28
Many professional societies have issued DTC testing opinion statements that address the concern for misinterpretation of genetic test results without an appropriately trained professional and the need for federal regulation to protect consumers.18,29–31 Very recently, the American Medical Association issued a letter to the US Food and Drug Administration urging the agency to recommend that genetic testing be carried out under the supervision of a qualified healthcare professional.32
Counterpoint: DTC is a new service delivery model and provides information to consumers who may otherwise not have genetic testing.
Some DTC testing companies have employed genetic counselors on staff to be proactive in addressing patients’ concerns and questions about their results. And, although bias is a concern, certified genetic counselors working for these companies can provide basic information to consumers and will hopefully refer them to outside professionals to verify test results in a CLIA (Clinical Laboratory Improvement Amendments)-approved laboratory, provide personalized result interpretations based on the patients’ personal and family histories, and provide long-term patient and family follow-up. Some patients who undergo DTC testing may not have sought clinic-based genetic counseling and testing. Thus, this new model of genetic testing may allow more consumers and their families to access and benefit from genetic information.
Conclusion
DTC genetic testing is now widely available, with companies offering testing for a broad variety of traits and diseases, including several cancers. The disease risk information provided is typically based on SNP panels without demonstrated clinical validity or utility and may also include testing for a limited number of high-risk gene mutations. It can be difficult for even a knowledgeable genetics expert to separate out these data based on the limited information provided by these companies.
Although proponents of DTC genetic testing cite increased access, privacy, patient autonomy, and improved genetics knowledge, there is increasing concern among the scientific, medical, ethical, and legal communities that such testing is misleading and potentially harmful to patients as well as society and thus should be subject to increased regulation.5,15,18,22,28,30,33 As such regulation will take time, patients who inquire about DTC genetic testing for cancer risk should be cautioned that the information they receive may be inaccurate and misleading. Insofar as this new model of genetic testing has the potential to reach many consumers who would not otherwise have accessed testing, potential benefit and minimization of harm can be facilitated by involvement of a qualified genetics professional to allow for appropriate test interpretation in the context of medical and family histories.
References
1. Myers MF, Chang M-H, Jorgensen C, et al. Genetic testing for susceptibility to breast and ovarian cancer: evaluating the impact of a direct-to-consumer marketing campaign on physicians’ knowledge and practices. Genet Med 2006;8:361–370.
2. Janssens CJW, Gwinn M, Bradley LA, et al. A critical appraisal of the scientific basis of commercial genomic profiles used to assess health risks and personalized health interventions. Am J Hum Genet 2008;82:593–599.
3. Magnus D, Cho M, Cook-Deegan B. Direct-to-consumer genetic tests: beyond medical regulation? Genome Med 2009;1:17.
4. Eng C, Sharp R. Bioethical and clinical dilemmas of direct-to-consumer personal genomic testing: the problem of misattributed equivalence. Sci Transl Med 2010;2(17):17cm5.
5. Khoury MJ, McBride CM, Schully SD, et al. Centers for Disease Control and Prevention. The scientific foundation for personal genomics: recommendations from a National Institutes of Health-Centers for Disease Control and Prevention multidisciplinary workshop. Genet Med 2009;11:559–567.
6. Wacholder S, Haretge P, Prentice R. Performance of common genetic variants in breast cancer risk models. N Engl J Med 2010;363:986–993.
7. Chen S, Parmigiani G. Meta-analysis of BRCA1 and BRCA2 penetrance. J Clin Oncol 2007;25:1329–1333.
8. Mealiffe M, Stokowski R, Thees B, Prentice R, Pettinger M, Hinds D. Assessment of clinical validity of a breast cancer risk model combining genetic and clinical information. J Natl Cancer Inst 2010;102:1618–1627.
9. deCODE Genetics Web site. http://www. decode.com. Accessed March 13, 2011.
10. 23andME Web site. http://www.23andMe. com. Accessed March 13, 2011.
11. Navigenics Web site. http://www.navigenics. com. Accessed March 13, 2011.
12. Proactive Genomics Web site. http:// www.proactivegenomics.com. Accessed March 13, 2011.
13. Ng PC, Murray SS, Levy S, Venter JC. An agenda for personalized medicine. Nature 2009;461:724–726.
14. GAO report: Direct-to-consumer genetic tests: misleading test results are further complicated by deceptive marketing and other questionable practices. GAO-10-847T July 22, 2010. http://www.gao.gov/products/GAO-10- 847T. Accessed March 13, 2011.
15. Udesky L. The ethics of direct-to-consumer genetic testing. Lancet 2010;376:1377– 1378.
16. Berndt S, Platz E, Fallin M. Mismatch repair polymorphism and the risk of colorectal cancer. Int J Cancer 2007;120:1548–1554.
17. da Silva TD, Felipe AV, de Lima JM, Oshima CT, Forones NM. N-Acetyltransferase 2 genetic polymorphisms and risk of colorectal cancer. World J Gastroenterol 2011;17:760– 765.
18. Hogarth S, Javitt G, Melzer D. The current landscape for direct-to-consumer genetic testing: legal, ethical, and policy issues. Annu Rev Genomics Hum Genet 2008;9:161–182.
19. Gray S, Olopade OI. Direct-to-consumer marketing of genetics tests for cancer: buyer beware. J Clin Oncol 2003;21:3191– 3193.
20. Gollust SE, Chandros Hull S, Wilfond BS. Limitations of direct-to-consumer advertising for clinical genetic testing. JAMA 2002;288:1762–1767.
21. Wolfberg AJ. Genes on the Web: directto- consumer marketing of genetic testing. N Engl J Med 2006;355:543–545.
22. European Society of Human Genetics. Statement of the ESHG on direct-to-consumer genetic testing for health-related purposes. Eur J Hum Genet 2010;18:1271–1273.
23. Williams-Jones B. “Be ready against cancer, now”: direct-to-consumer advertising for genetic testing. New Genet Soc 2006;25:89–107.
24. Kolar K, Liu T, St Pierre J, Khoury MJ. Heath care provider and consumer awareness, perceptions, and use of direct-to-consumer personal genomic tests. Genet Med 2009;11:595.
25. McGuire AL, Diaz C, Wang T, et al. Social networkers’ attitudes toward direct-toconsumer personal genome testing. Am J Bioethics 2009;9:3–10.
26. McGuire AL, Burke W. An unwelcome side effect of direct-to consumer personal genome testing: raiding the medical commons. JAMA 2008;300:2669–2671.
27. Caufield T, Ries NM, Ray PN, Shuman C, Wilson B. Direct-to-consumer genetic testing: good, bad or benign? Clin Genetic 2010;77:101–105.
28. Mykitiuk R. Caveat emptor: direct-toconsumer supply and advertising of genetic testing. Clin Invest Med 2004;27:23–32.
29. American College of Medical Genetics (ACMG) statement on direct-to-consumer genetic testing: April 7, 2008. http://www. acmg.net/AM/Template.cfm?Section=Policy_ Statements&Template=/CM/ContentDisplay. cfm&ContentID=2975, Accessed March 13, 2011.
30. Hudson K, Javitt G, Burke W. ASHG statement on direct-to-consumer genetic testing in the United States. Obstet Gynecol 2007;110:1392–1395.
31. NSGC statement on direct to consumer genetic testing. Adopted 2007. http://www. nsgc.org/Media/PositionStatements/tabid/ 330/Default.aspx#DTC. Accessed March 13, 2011.
32. AMA Press Release, February 23, 2011. AMA to FDA: genetic testing should be conducted by qualified health professionals. http:// www.ama-assn.org/ama/pub/news/news/genetic- testing-qualified-professionals.page. Accessed March 8, 2011.
33. Annes JP, Giovanni MA, Murray MF. Risk of presymptomatic direct-to-consumer genetic testing. N Engl J Med 2010;363:1100–1101.
The genomic revolution has the potential to change the practice of medicine and healthcare. Advances in technology are driving down the cost of genetic testing, the public is more aware of and interested in such testing and its uses, and genetic testing companies are now offering direct-to-consumer (DTC) testing via kits sold on the Internet. Some argue that these changes make genetic testing more accessible to the general public and medical community at large and that these changes are critical to the full integration of genomic technology into society. However, many of the DTC tests offered are of questionable accuracy and utility and combine tests that provide mundane information about genetic traits with those for serious genetic conditions with life-threatening implications for the patient and the entire family. We will examine both sides of the issues surrounding DTC genetic testing, focusing on important considerations for clinicians using case vignettes to illustrate key issues.
Is DTC testing worthwhile for your patients?
Point: The testing and result interpretations provided are not necessarily clinically accurate or helpful.
Numerous studies have shown that average patients and clinicians are not well versed on even the most basic elements of genetic testing and result interpretation. 1 It is even less likely that consumers and clinicians will understand the subtle differences between DTC companies and tests—particularly genomic profiling based on single nucleotide polymorphisms (SNPs)—which are offered by most DTC companies. The difference between SNPbased testing and clinical single-gene testing with regard to predictive value is a confusing concept that typical consumers and clinicians without adequate training may not understand. For most of the SNP-based genomic profiles offered DTC, there is a lack of consensus among the scientific and medical communities regarding test interpretation.
The positive and negative predictive values, along with the discriminatory accuracy of these tests, are in sharp contrast to those associated with testing for mutations in high-penetrance Mendelian genes. Discriminatory accuracy refers to the ability of a test to determine who will or will not develop the disease or condition and is a balance between sensitivity and specificity. For example, those who test positive for a mutation in the Huntington's disease gene are certain to develop the disease, whereas those who test negative are unlikely to do so. Similarly, in cancer genetics, those with a mutation in the APC gene (causative of familial adenomatous polyposis) will develop colon polyps and colon cancer without appropriate intervention.
With these Mendelian conditions, the relative risk of disease (penetrance) is large (often 10-fold or greater), and the proportion of disease attribestimated by these companies fail to account for known family history or other risk factors. An investigation in July 2010 by the US General Accounting Office concluded that such tests were “misleading and of little or no practical use.”14 The risk profiles provided by such companies have been described as having “no predictive value” and may falsely alarm or reassure consumers.15
Counterpoint: Individuals have the right to explore their own genetic makeup through these emerging tests and may deem testing worthwhile. Some individuals may learn important information they would otherwise not have learned.
One argument to be made is that genetic technology exists and individuals have the right to explore the testing available, regardless of its accuracy and utility. Many DTC sites do state that their testing is intended for entertainment purposes only, although this statement is often in the fine print (see next point). It is possible that DTC testing has made the public more aware of genetic and genomic testing, therefore encouraging the medical profession to keep up with this technology as it pertains to one's area of expertise. Although some may see this as an additional burden on the primary care provider, it can also be viewed as an opportunity to increase awareness and education on the part of both the consumer and the professional. Most current genomic technology was not developed when most of today’s physicians went to medical school, and even most new graduates are not well versed in genetic testing. DTC genetic testing could encourage the medical community to increase its understanding of this new emerging technology.
Although most of the DTC testing offered is of questionable utility, some panels do include important genetic information, like the three common Jewish BRCA mutations. It may be considered ethically questionable to offer testing for hereditary breast cancer and ovarian cancer without full informed consent and adequate result disclosure. However, one can also argue that individuals will gain potentially lifesaving information they might never have uncovered without such testing.
Proponents of DTC genomic profiling suggest that the clinical usefulness of SNP-based testing will improve based on emerging data that analyze the interaction between genetic polymorphisms and environmental factors in predicting disease risk. For example, preliminary studies evaluating SNPs associated with colorectal cancer have found that relative risks are increased when combined with environmental factors, such as high intake of processed meats.16,17 Despite these early findings, it remains to be seen whether the complexity of gene-environment and gene-gene interactions can be elucidated to the extent that will be required to yield highly predictive DTC tests.
Is DTC advertising accurate, fair, and helpful?
Point: DTC advertising overemphasizes the benefits of testing and does not adequately describe the limitations and risks.
Unlike pharmaceutical advertising, which is tightly regulated and requires companies to discuss associated risks, oversight of genetic testing advertisements is limited.18 This can lead to skewing of information that overemphasizes the benefits of genetic testing and does not adequately define the risks and limitations.19,20 One recent study demonstrated that when individuals are provided with accurate information regarding the actual risks and limitations of online DTC genetic testing, they have reduced intentions to obtain such tests and less positive views of Internet-based testing. 19 Furthermore, an investigation of DTC advertising revealed “egregious examples” of deceptive marketing, such as claims that consumers’ DNA could be used to create personalized supplements to cure disease or that results of such DNA tests could be used to predict the athletic potential of their children.14
Many DTC companies have a clause indicating that their testing is intended for research, educational, or entertainment purposes only, and not for diagnostic or medical use. However, this clause is embedded deep into the Web site and surrounded by text like “disease risk prediction,” “predicted response to drugs,” and a long list of serious medical conditions that appear to be anything but entertainment- based. Together with mundane traits, such as bitter taste perception, freckling, earwax type, and hair thickness, consumers are tested for susceptibility to amyotrophic lateral sclerosis, bipolar disease, heroin addiction, and cancer predisposition. This strange combination of offerings makes it difficult to determine whether this testing is truly entertainmentbased or a serious predictive testing panel requiring careful interpretation by a medical professional.
Complicating these issues even further, several DTC testing companies do not fully explain which SNPs or gene mutations are evaluated in their specific disease risk profiles. Therefore, the average woman may think she has had comprehensive sequencing for BRCA1 and BRCA2 because her “breast cancer risk” has been assessed, when in fact she has been tested for a panel of low-risk SNPs and only the three common Jewish BRCA mutations. For a non-Jewish patient, these results would be virtually meaningless in terms of assessing hereditary risk.
Even well-versed genetics professionals may not be able to decipher the implications of tests offered by some companies. Recently, a cancer genetics professional reported that she reviewed a multigene-disease panel offered by one DTC genetic testing company that reportedly included testing for MYH-associated polyposis. When she discussed this test with the company, she discovered that it provided analysis for only one of the two common MYH mutations in Caucasians (unpublished, personal communication). These important omissions could lead both patients and clinicians to think that comprehensive testing has been performed when, in fact, it has not.
Counterpoint: DTC advertising has increased physician and patient awareness of genetic testing and technology.
As consumers become more sophisticated in their medical knowledge base, many are turning to the Internet for healthcare information before, or instead of, addressing their concerns with a medical professional. Although medical information on the Internet is of varied accuracy, obtaining medical information online is a trend that is here to stay, and genetic testing is no exception. The DTC genetic testing companies have exploited this thirst for medical information and have developed educational information on basic “Genetics 101” concepts for the lay public. For example, 23andMe, a DTC genetic testing company, describes genomic testing through an animated video clip using cartoon characters on its Web site. This new educational delivery system not only helps the genetic testing company sell their product but also helps the consumer learn more about genomic technology. In response to DTC genetic testing, many medical groups have launched educational campaigns to help bring clinicians up-to-date on the genetic testing their patients may be pursuing and bringing to their office.
Many professional medical organizations have addressed this gap in knowledge and have continually offered additional training regarding regarding the implications of genomic and genetic technologies in their subspeciality. Oncology has been at the forefront of this movement, offering training to oncologists through the American Society of Clinical Oncology to expand the knowledge base on genetic and genomic testing.
Is DTC genetic testing without the guidance of a qualified professional doing more harm than good?
Point: For consumers to have DTC testing without the guidance of a qualified professional to interpret complex information is not in their best interest.
Many DTC genetic testing companies allow consumers to order their own tests and provide results, risk calculations, and interpretations to patients without any contact with a knowledgeable clinician. As illustrated in the BRCA vignette below, the information provided to patients with their results is often not only inaccurate and irresponsible but may be harmful or deadly to the patient and at-risk family members.
An undercover investigation of DTC testing companies revealed poor to nonexistent medical advice from supposed consult experts, in addition to possible privacy concerns.14 Although some DTC genetic testing companies offer genetic counseling as part of their services, the consumer needs to be aware that medical advice from an individual with a financial interest in the product or company may be biased.21,22 Much as patients would not contact a pharmaceutical company to ask whether they should take a drug it supplies, patients should be wary of receiving advice regarding genetic testing from a company that stands to profit directly from their decision.
The complex information provided by these DTC genetic tests for disease risk is at best difficult to interpret, and there is concern that patients and physicians may make important healthcare decisions based on this information without a clear understanding of the test limitations. 18,23 A 2008 survey of physicians found that among those aware of personal genomic testing, 15% had a patient who had brought in test results for discussion, and more than 50% indicated that such tests would influence their decisions regarding patient care.24 Similarly, McGuire et al found that a majority (78%) of consumers report that they would take results of DTC genetic testing to their healthcare providers.25 These findings lend credence to the concerns expressed by several authors that DTC genetic testing may lead to overuse of the healthcare system, including unnecessary clinician visits and follow-up evaluations or tests.18,22,25–28
Many professional societies have issued DTC testing opinion statements that address the concern for misinterpretation of genetic test results without an appropriately trained professional and the need for federal regulation to protect consumers.18,29–31 Very recently, the American Medical Association issued a letter to the US Food and Drug Administration urging the agency to recommend that genetic testing be carried out under the supervision of a qualified healthcare professional.32
Counterpoint: DTC is a new service delivery model and provides information to consumers who may otherwise not have genetic testing.
Some DTC testing companies have employed genetic counselors on staff to be proactive in addressing patients’ concerns and questions about their results. And, although bias is a concern, certified genetic counselors working for these companies can provide basic information to consumers and will hopefully refer them to outside professionals to verify test results in a CLIA (Clinical Laboratory Improvement Amendments)-approved laboratory, provide personalized result interpretations based on the patients’ personal and family histories, and provide long-term patient and family follow-up. Some patients who undergo DTC testing may not have sought clinic-based genetic counseling and testing. Thus, this new model of genetic testing may allow more consumers and their families to access and benefit from genetic information.
Conclusion
DTC genetic testing is now widely available, with companies offering testing for a broad variety of traits and diseases, including several cancers. The disease risk information provided is typically based on SNP panels without demonstrated clinical validity or utility and may also include testing for a limited number of high-risk gene mutations. It can be difficult for even a knowledgeable genetics expert to separate out these data based on the limited information provided by these companies.
Although proponents of DTC genetic testing cite increased access, privacy, patient autonomy, and improved genetics knowledge, there is increasing concern among the scientific, medical, ethical, and legal communities that such testing is misleading and potentially harmful to patients as well as society and thus should be subject to increased regulation.5,15,18,22,28,30,33 As such regulation will take time, patients who inquire about DTC genetic testing for cancer risk should be cautioned that the information they receive may be inaccurate and misleading. Insofar as this new model of genetic testing has the potential to reach many consumers who would not otherwise have accessed testing, potential benefit and minimization of harm can be facilitated by involvement of a qualified genetics professional to allow for appropriate test interpretation in the context of medical and family histories.
References
1. Myers MF, Chang M-H, Jorgensen C, et al. Genetic testing for susceptibility to breast and ovarian cancer: evaluating the impact of a direct-to-consumer marketing campaign on physicians’ knowledge and practices. Genet Med 2006;8:361–370.
2. Janssens CJW, Gwinn M, Bradley LA, et al. A critical appraisal of the scientific basis of commercial genomic profiles used to assess health risks and personalized health interventions. Am J Hum Genet 2008;82:593–599.
3. Magnus D, Cho M, Cook-Deegan B. Direct-to-consumer genetic tests: beyond medical regulation? Genome Med 2009;1:17.
4. Eng C, Sharp R. Bioethical and clinical dilemmas of direct-to-consumer personal genomic testing: the problem of misattributed equivalence. Sci Transl Med 2010;2(17):17cm5.
5. Khoury MJ, McBride CM, Schully SD, et al. Centers for Disease Control and Prevention. The scientific foundation for personal genomics: recommendations from a National Institutes of Health-Centers for Disease Control and Prevention multidisciplinary workshop. Genet Med 2009;11:559–567.
6. Wacholder S, Haretge P, Prentice R. Performance of common genetic variants in breast cancer risk models. N Engl J Med 2010;363:986–993.
7. Chen S, Parmigiani G. Meta-analysis of BRCA1 and BRCA2 penetrance. J Clin Oncol 2007;25:1329–1333.
8. Mealiffe M, Stokowski R, Thees B, Prentice R, Pettinger M, Hinds D. Assessment of clinical validity of a breast cancer risk model combining genetic and clinical information. J Natl Cancer Inst 2010;102:1618–1627.
9. deCODE Genetics Web site. http://www. decode.com. Accessed March 13, 2011.
10. 23andME Web site. http://www.23andMe. com. Accessed March 13, 2011.
11. Navigenics Web site. http://www.navigenics. com. Accessed March 13, 2011.
12. Proactive Genomics Web site. http:// www.proactivegenomics.com. Accessed March 13, 2011.
13. Ng PC, Murray SS, Levy S, Venter JC. An agenda for personalized medicine. Nature 2009;461:724–726.
14. GAO report: Direct-to-consumer genetic tests: misleading test results are further complicated by deceptive marketing and other questionable practices. GAO-10-847T July 22, 2010. http://www.gao.gov/products/GAO-10- 847T. Accessed March 13, 2011.
15. Udesky L. The ethics of direct-to-consumer genetic testing. Lancet 2010;376:1377– 1378.
16. Berndt S, Platz E, Fallin M. Mismatch repair polymorphism and the risk of colorectal cancer. Int J Cancer 2007;120:1548–1554.
17. da Silva TD, Felipe AV, de Lima JM, Oshima CT, Forones NM. N-Acetyltransferase 2 genetic polymorphisms and risk of colorectal cancer. World J Gastroenterol 2011;17:760– 765.
18. Hogarth S, Javitt G, Melzer D. The current landscape for direct-to-consumer genetic testing: legal, ethical, and policy issues. Annu Rev Genomics Hum Genet 2008;9:161–182.
19. Gray S, Olopade OI. Direct-to-consumer marketing of genetics tests for cancer: buyer beware. J Clin Oncol 2003;21:3191– 3193.
20. Gollust SE, Chandros Hull S, Wilfond BS. Limitations of direct-to-consumer advertising for clinical genetic testing. JAMA 2002;288:1762–1767.
21. Wolfberg AJ. Genes on the Web: directto- consumer marketing of genetic testing. N Engl J Med 2006;355:543–545.
22. European Society of Human Genetics. Statement of the ESHG on direct-to-consumer genetic testing for health-related purposes. Eur J Hum Genet 2010;18:1271–1273.
23. Williams-Jones B. “Be ready against cancer, now”: direct-to-consumer advertising for genetic testing. New Genet Soc 2006;25:89–107.
24. Kolar K, Liu T, St Pierre J, Khoury MJ. Heath care provider and consumer awareness, perceptions, and use of direct-to-consumer personal genomic tests. Genet Med 2009;11:595.
25. McGuire AL, Diaz C, Wang T, et al. Social networkers’ attitudes toward direct-toconsumer personal genome testing. Am J Bioethics 2009;9:3–10.
26. McGuire AL, Burke W. An unwelcome side effect of direct-to consumer personal genome testing: raiding the medical commons. JAMA 2008;300:2669–2671.
27. Caufield T, Ries NM, Ray PN, Shuman C, Wilson B. Direct-to-consumer genetic testing: good, bad or benign? Clin Genetic 2010;77:101–105.
28. Mykitiuk R. Caveat emptor: direct-toconsumer supply and advertising of genetic testing. Clin Invest Med 2004;27:23–32.
29. American College of Medical Genetics (ACMG) statement on direct-to-consumer genetic testing: April 7, 2008. http://www. acmg.net/AM/Template.cfm?Section=Policy_ Statements&Template=/CM/ContentDisplay. cfm&ContentID=2975, Accessed March 13, 2011.
30. Hudson K, Javitt G, Burke W. ASHG statement on direct-to-consumer genetic testing in the United States. Obstet Gynecol 2007;110:1392–1395.
31. NSGC statement on direct to consumer genetic testing. Adopted 2007. http://www. nsgc.org/Media/PositionStatements/tabid/ 330/Default.aspx#DTC. Accessed March 13, 2011.
32. AMA Press Release, February 23, 2011. AMA to FDA: genetic testing should be conducted by qualified health professionals. http:// www.ama-assn.org/ama/pub/news/news/genetic- testing-qualified-professionals.page. Accessed March 8, 2011.
33. Annes JP, Giovanni MA, Murray MF. Risk of presymptomatic direct-to-consumer genetic testing. N Engl J Med 2010;363:1100–1101.
The genomic revolution has the potential to change the practice of medicine and healthcare. Advances in technology are driving down the cost of genetic testing, the public is more aware of and interested in such testing and its uses, and genetic testing companies are now offering direct-to-consumer (DTC) testing via kits sold on the Internet. Some argue that these changes make genetic testing more accessible to the general public and medical community at large and that these changes are critical to the full integration of genomic technology into society. However, many of the DTC tests offered are of questionable accuracy and utility and combine tests that provide mundane information about genetic traits with those for serious genetic conditions with life-threatening implications for the patient and the entire family. We will examine both sides of the issues surrounding DTC genetic testing, focusing on important considerations for clinicians using case vignettes to illustrate key issues.
Is DTC testing worthwhile for your patients?
Point: The testing and result interpretations provided are not necessarily clinically accurate or helpful.
Numerous studies have shown that average patients and clinicians are not well versed on even the most basic elements of genetic testing and result interpretation. 1 It is even less likely that consumers and clinicians will understand the subtle differences between DTC companies and tests—particularly genomic profiling based on single nucleotide polymorphisms (SNPs)—which are offered by most DTC companies. The difference between SNPbased testing and clinical single-gene testing with regard to predictive value is a confusing concept that typical consumers and clinicians without adequate training may not understand. For most of the SNP-based genomic profiles offered DTC, there is a lack of consensus among the scientific and medical communities regarding test interpretation.
The positive and negative predictive values, along with the discriminatory accuracy of these tests, are in sharp contrast to those associated with testing for mutations in high-penetrance Mendelian genes. Discriminatory accuracy refers to the ability of a test to determine who will or will not develop the disease or condition and is a balance between sensitivity and specificity. For example, those who test positive for a mutation in the Huntington's disease gene are certain to develop the disease, whereas those who test negative are unlikely to do so. Similarly, in cancer genetics, those with a mutation in the APC gene (causative of familial adenomatous polyposis) will develop colon polyps and colon cancer without appropriate intervention.
With these Mendelian conditions, the relative risk of disease (penetrance) is large (often 10-fold or greater), and the proportion of disease attribestimated by these companies fail to account for known family history or other risk factors. An investigation in July 2010 by the US General Accounting Office concluded that such tests were “misleading and of little or no practical use.”14 The risk profiles provided by such companies have been described as having “no predictive value” and may falsely alarm or reassure consumers.15
Counterpoint: Individuals have the right to explore their own genetic makeup through these emerging tests and may deem testing worthwhile. Some individuals may learn important information they would otherwise not have learned.
One argument to be made is that genetic technology exists and individuals have the right to explore the testing available, regardless of its accuracy and utility. Many DTC sites do state that their testing is intended for entertainment purposes only, although this statement is often in the fine print (see next point). It is possible that DTC testing has made the public more aware of genetic and genomic testing, therefore encouraging the medical profession to keep up with this technology as it pertains to one's area of expertise. Although some may see this as an additional burden on the primary care provider, it can also be viewed as an opportunity to increase awareness and education on the part of both the consumer and the professional. Most current genomic technology was not developed when most of today’s physicians went to medical school, and even most new graduates are not well versed in genetic testing. DTC genetic testing could encourage the medical community to increase its understanding of this new emerging technology.
Although most of the DTC testing offered is of questionable utility, some panels do include important genetic information, like the three common Jewish BRCA mutations. It may be considered ethically questionable to offer testing for hereditary breast cancer and ovarian cancer without full informed consent and adequate result disclosure. However, one can also argue that individuals will gain potentially lifesaving information they might never have uncovered without such testing.
Proponents of DTC genomic profiling suggest that the clinical usefulness of SNP-based testing will improve based on emerging data that analyze the interaction between genetic polymorphisms and environmental factors in predicting disease risk. For example, preliminary studies evaluating SNPs associated with colorectal cancer have found that relative risks are increased when combined with environmental factors, such as high intake of processed meats.16,17 Despite these early findings, it remains to be seen whether the complexity of gene-environment and gene-gene interactions can be elucidated to the extent that will be required to yield highly predictive DTC tests.
Is DTC advertising accurate, fair, and helpful?
Point: DTC advertising overemphasizes the benefits of testing and does not adequately describe the limitations and risks.
Unlike pharmaceutical advertising, which is tightly regulated and requires companies to discuss associated risks, oversight of genetic testing advertisements is limited.18 This can lead to skewing of information that overemphasizes the benefits of genetic testing and does not adequately define the risks and limitations.19,20 One recent study demonstrated that when individuals are provided with accurate information regarding the actual risks and limitations of online DTC genetic testing, they have reduced intentions to obtain such tests and less positive views of Internet-based testing. 19 Furthermore, an investigation of DTC advertising revealed “egregious examples” of deceptive marketing, such as claims that consumers’ DNA could be used to create personalized supplements to cure disease or that results of such DNA tests could be used to predict the athletic potential of their children.14
Many DTC companies have a clause indicating that their testing is intended for research, educational, or entertainment purposes only, and not for diagnostic or medical use. However, this clause is embedded deep into the Web site and surrounded by text like “disease risk prediction,” “predicted response to drugs,” and a long list of serious medical conditions that appear to be anything but entertainment- based. Together with mundane traits, such as bitter taste perception, freckling, earwax type, and hair thickness, consumers are tested for susceptibility to amyotrophic lateral sclerosis, bipolar disease, heroin addiction, and cancer predisposition. This strange combination of offerings makes it difficult to determine whether this testing is truly entertainmentbased or a serious predictive testing panel requiring careful interpretation by a medical professional.
Complicating these issues even further, several DTC testing companies do not fully explain which SNPs or gene mutations are evaluated in their specific disease risk profiles. Therefore, the average woman may think she has had comprehensive sequencing for BRCA1 and BRCA2 because her “breast cancer risk” has been assessed, when in fact she has been tested for a panel of low-risk SNPs and only the three common Jewish BRCA mutations. For a non-Jewish patient, these results would be virtually meaningless in terms of assessing hereditary risk.
Even well-versed genetics professionals may not be able to decipher the implications of tests offered by some companies. Recently, a cancer genetics professional reported that she reviewed a multigene-disease panel offered by one DTC genetic testing company that reportedly included testing for MYH-associated polyposis. When she discussed this test with the company, she discovered that it provided analysis for only one of the two common MYH mutations in Caucasians (unpublished, personal communication). These important omissions could lead both patients and clinicians to think that comprehensive testing has been performed when, in fact, it has not.
Counterpoint: DTC advertising has increased physician and patient awareness of genetic testing and technology.
As consumers become more sophisticated in their medical knowledge base, many are turning to the Internet for healthcare information before, or instead of, addressing their concerns with a medical professional. Although medical information on the Internet is of varied accuracy, obtaining medical information online is a trend that is here to stay, and genetic testing is no exception. The DTC genetic testing companies have exploited this thirst for medical information and have developed educational information on basic “Genetics 101” concepts for the lay public. For example, 23andMe, a DTC genetic testing company, describes genomic testing through an animated video clip using cartoon characters on its Web site. This new educational delivery system not only helps the genetic testing company sell their product but also helps the consumer learn more about genomic technology. In response to DTC genetic testing, many medical groups have launched educational campaigns to help bring clinicians up-to-date on the genetic testing their patients may be pursuing and bringing to their office.
Many professional medical organizations have addressed this gap in knowledge and have continually offered additional training regarding regarding the implications of genomic and genetic technologies in their subspeciality. Oncology has been at the forefront of this movement, offering training to oncologists through the American Society of Clinical Oncology to expand the knowledge base on genetic and genomic testing.
Is DTC genetic testing without the guidance of a qualified professional doing more harm than good?
Point: For consumers to have DTC testing without the guidance of a qualified professional to interpret complex information is not in their best interest.
Many DTC genetic testing companies allow consumers to order their own tests and provide results, risk calculations, and interpretations to patients without any contact with a knowledgeable clinician. As illustrated in the BRCA vignette below, the information provided to patients with their results is often not only inaccurate and irresponsible but may be harmful or deadly to the patient and at-risk family members.
An undercover investigation of DTC testing companies revealed poor to nonexistent medical advice from supposed consult experts, in addition to possible privacy concerns.14 Although some DTC genetic testing companies offer genetic counseling as part of their services, the consumer needs to be aware that medical advice from an individual with a financial interest in the product or company may be biased.21,22 Much as patients would not contact a pharmaceutical company to ask whether they should take a drug it supplies, patients should be wary of receiving advice regarding genetic testing from a company that stands to profit directly from their decision.
The complex information provided by these DTC genetic tests for disease risk is at best difficult to interpret, and there is concern that patients and physicians may make important healthcare decisions based on this information without a clear understanding of the test limitations. 18,23 A 2008 survey of physicians found that among those aware of personal genomic testing, 15% had a patient who had brought in test results for discussion, and more than 50% indicated that such tests would influence their decisions regarding patient care.24 Similarly, McGuire et al found that a majority (78%) of consumers report that they would take results of DTC genetic testing to their healthcare providers.25 These findings lend credence to the concerns expressed by several authors that DTC genetic testing may lead to overuse of the healthcare system, including unnecessary clinician visits and follow-up evaluations or tests.18,22,25–28
Many professional societies have issued DTC testing opinion statements that address the concern for misinterpretation of genetic test results without an appropriately trained professional and the need for federal regulation to protect consumers.18,29–31 Very recently, the American Medical Association issued a letter to the US Food and Drug Administration urging the agency to recommend that genetic testing be carried out under the supervision of a qualified healthcare professional.32
Counterpoint: DTC is a new service delivery model and provides information to consumers who may otherwise not have genetic testing.
Some DTC testing companies have employed genetic counselors on staff to be proactive in addressing patients’ concerns and questions about their results. And, although bias is a concern, certified genetic counselors working for these companies can provide basic information to consumers and will hopefully refer them to outside professionals to verify test results in a CLIA (Clinical Laboratory Improvement Amendments)-approved laboratory, provide personalized result interpretations based on the patients’ personal and family histories, and provide long-term patient and family follow-up. Some patients who undergo DTC testing may not have sought clinic-based genetic counseling and testing. Thus, this new model of genetic testing may allow more consumers and their families to access and benefit from genetic information.
Conclusion
DTC genetic testing is now widely available, with companies offering testing for a broad variety of traits and diseases, including several cancers. The disease risk information provided is typically based on SNP panels without demonstrated clinical validity or utility and may also include testing for a limited number of high-risk gene mutations. It can be difficult for even a knowledgeable genetics expert to separate out these data based on the limited information provided by these companies.
Although proponents of DTC genetic testing cite increased access, privacy, patient autonomy, and improved genetics knowledge, there is increasing concern among the scientific, medical, ethical, and legal communities that such testing is misleading and potentially harmful to patients as well as society and thus should be subject to increased regulation.5,15,18,22,28,30,33 As such regulation will take time, patients who inquire about DTC genetic testing for cancer risk should be cautioned that the information they receive may be inaccurate and misleading. Insofar as this new model of genetic testing has the potential to reach many consumers who would not otherwise have accessed testing, potential benefit and minimization of harm can be facilitated by involvement of a qualified genetics professional to allow for appropriate test interpretation in the context of medical and family histories.
References
1. Myers MF, Chang M-H, Jorgensen C, et al. Genetic testing for susceptibility to breast and ovarian cancer: evaluating the impact of a direct-to-consumer marketing campaign on physicians’ knowledge and practices. Genet Med 2006;8:361–370.
2. Janssens CJW, Gwinn M, Bradley LA, et al. A critical appraisal of the scientific basis of commercial genomic profiles used to assess health risks and personalized health interventions. Am J Hum Genet 2008;82:593–599.
3. Magnus D, Cho M, Cook-Deegan B. Direct-to-consumer genetic tests: beyond medical regulation? Genome Med 2009;1:17.
4. Eng C, Sharp R. Bioethical and clinical dilemmas of direct-to-consumer personal genomic testing: the problem of misattributed equivalence. Sci Transl Med 2010;2(17):17cm5.
5. Khoury MJ, McBride CM, Schully SD, et al. Centers for Disease Control and Prevention. The scientific foundation for personal genomics: recommendations from a National Institutes of Health-Centers for Disease Control and Prevention multidisciplinary workshop. Genet Med 2009;11:559–567.
6. Wacholder S, Haretge P, Prentice R. Performance of common genetic variants in breast cancer risk models. N Engl J Med 2010;363:986–993.
7. Chen S, Parmigiani G. Meta-analysis of BRCA1 and BRCA2 penetrance. J Clin Oncol 2007;25:1329–1333.
8. Mealiffe M, Stokowski R, Thees B, Prentice R, Pettinger M, Hinds D. Assessment of clinical validity of a breast cancer risk model combining genetic and clinical information. J Natl Cancer Inst 2010;102:1618–1627.
9. deCODE Genetics Web site. http://www. decode.com. Accessed March 13, 2011.
10. 23andME Web site. http://www.23andMe. com. Accessed March 13, 2011.
11. Navigenics Web site. http://www.navigenics. com. Accessed March 13, 2011.
12. Proactive Genomics Web site. http:// www.proactivegenomics.com. Accessed March 13, 2011.
13. Ng PC, Murray SS, Levy S, Venter JC. An agenda for personalized medicine. Nature 2009;461:724–726.
14. GAO report: Direct-to-consumer genetic tests: misleading test results are further complicated by deceptive marketing and other questionable practices. GAO-10-847T July 22, 2010. http://www.gao.gov/products/GAO-10- 847T. Accessed March 13, 2011.
15. Udesky L. The ethics of direct-to-consumer genetic testing. Lancet 2010;376:1377– 1378.
16. Berndt S, Platz E, Fallin M. Mismatch repair polymorphism and the risk of colorectal cancer. Int J Cancer 2007;120:1548–1554.
17. da Silva TD, Felipe AV, de Lima JM, Oshima CT, Forones NM. N-Acetyltransferase 2 genetic polymorphisms and risk of colorectal cancer. World J Gastroenterol 2011;17:760– 765.
18. Hogarth S, Javitt G, Melzer D. The current landscape for direct-to-consumer genetic testing: legal, ethical, and policy issues. Annu Rev Genomics Hum Genet 2008;9:161–182.
19. Gray S, Olopade OI. Direct-to-consumer marketing of genetics tests for cancer: buyer beware. J Clin Oncol 2003;21:3191– 3193.
20. Gollust SE, Chandros Hull S, Wilfond BS. Limitations of direct-to-consumer advertising for clinical genetic testing. JAMA 2002;288:1762–1767.
21. Wolfberg AJ. Genes on the Web: directto- consumer marketing of genetic testing. N Engl J Med 2006;355:543–545.
22. European Society of Human Genetics. Statement of the ESHG on direct-to-consumer genetic testing for health-related purposes. Eur J Hum Genet 2010;18:1271–1273.
23. Williams-Jones B. “Be ready against cancer, now”: direct-to-consumer advertising for genetic testing. New Genet Soc 2006;25:89–107.
24. Kolar K, Liu T, St Pierre J, Khoury MJ. Heath care provider and consumer awareness, perceptions, and use of direct-to-consumer personal genomic tests. Genet Med 2009;11:595.
25. McGuire AL, Diaz C, Wang T, et al. Social networkers’ attitudes toward direct-toconsumer personal genome testing. Am J Bioethics 2009;9:3–10.
26. McGuire AL, Burke W. An unwelcome side effect of direct-to consumer personal genome testing: raiding the medical commons. JAMA 2008;300:2669–2671.
27. Caufield T, Ries NM, Ray PN, Shuman C, Wilson B. Direct-to-consumer genetic testing: good, bad or benign? Clin Genetic 2010;77:101–105.
28. Mykitiuk R. Caveat emptor: direct-toconsumer supply and advertising of genetic testing. Clin Invest Med 2004;27:23–32.
29. American College of Medical Genetics (ACMG) statement on direct-to-consumer genetic testing: April 7, 2008. http://www. acmg.net/AM/Template.cfm?Section=Policy_ Statements&Template=/CM/ContentDisplay. cfm&ContentID=2975, Accessed March 13, 2011.
30. Hudson K, Javitt G, Burke W. ASHG statement on direct-to-consumer genetic testing in the United States. Obstet Gynecol 2007;110:1392–1395.
31. NSGC statement on direct to consumer genetic testing. Adopted 2007. http://www. nsgc.org/Media/PositionStatements/tabid/ 330/Default.aspx#DTC. Accessed March 13, 2011.
32. AMA Press Release, February 23, 2011. AMA to FDA: genetic testing should be conducted by qualified health professionals. http:// www.ama-assn.org/ama/pub/news/news/genetic- testing-qualified-professionals.page. Accessed March 8, 2011.
33. Annes JP, Giovanni MA, Murray MF. Risk of presymptomatic direct-to-consumer genetic testing. N Engl J Med 2010;363:1100–1101.
FROM THE FEDERAL REGISTER
Pharmaceutical patient assistance programs in the outpatient pharmacy of a large tertiary cancer center
Oral anticancer and supportive care agents administered to cancer patients are costly and are associated with large copayment requirements or are often not fully reimbursed by private health insurers or Medicare.1 To facilitate access to oral medications, pharmaceutical manufacturers have developed patient assistance programs (PAPs) that provide selected oral medications at no or reduced cost to financially eligible patients. Eligibility criteria, application processes, and program administration for PAPs differ by manufacturer and by product, which can ultimately present logistical barriers.2–4 A systematic review of PAPs found improvements in disease indicator outcomes for patients with common chronic diseases who access these programs.5 However, knowledge about the use of PAPs among cancer patients is limited.6
The University of Texas MD Anderson Cancer Center (MDACC), the largest tertiary care cancer center in the country, has developed a systematic approach to administering a large number of PAPs. In 1996, the MDACC established an institutional program staffed by hospital pharmacy personnel, who navigate cancer patients through PAPs in inpatient and outpatient settings. This program removes the operational and administrative barriers often experienced by patients in smaller clinical settings.
Cancer patients eligible for PAPs at MDACC include those who are uninsured, those who are underinsured, those whose pharmacy benefit limits have been reached, and those whose private health or government insurance has denied coverage of certain oral medications. For example, the Texas Medicaid program limits its low-income beneficiaries to three prescriptions per month, which may lead some of them, particularly those with cancer, to require additional medication assistance through PAPs. As of April 2008, this institutional program established formal relationships with 29 pharmaceutical companies that provide 104 therapeutic or supportive care agents through PAPs to eligible cancer patients in the MDACC outpatient pharmacy.
Methods
Data source
Approval for this study was obtained from the MDACC Institutional Review Board. We conducted a retrospective, secondary analysis of noninvestigational prescription medications from the outpatient pharmacy at MDACC. Data from July 1, 2006, to December 31, 2007, were extracted from computerized pharmacy, medical, and cancer registry databases at MDACC. Prescriptions had to include both patient medical record and social security numbers to validate the patient’s identity as well as the date of pickup to validate that the medication had been dispensed during the study period. When the date of pickup was missing but billing was documented, the date the medication was dispensed was used as the pickup date. All data were de-identified prior to analysis.
PAPs
Prescriptions for oral medications were available to financially eligible individuals via two types of PAPs at MDACC: individual enrollment (60 distinct medications) and bulk drug replacement (44 distinct medications). Individual enrollment required that an eligible patient apply directly to a pharmaceutical company’s PAP for the medication (s) needed. Once approved, the requested medication was mailed directly to the patient or dispensed in the MDACC pharmacy. Given the purpose of this study, we were only interested in those PAP prescription medications dispensed at the outpatient pharmacy.
Bulk replacement PAPs provide available prescription medications in bulk quantities on a monthly (in some cases quarterly) basis to MDACC’s pharmacy to replace medications dispensed to patients who were classified as “indigent” by MDACC-established criteria. Financially indigent patients included those who were Texas residents, uninsured or insured by Medicaid, and not responsible for charges billed to MDACC. All eligible patients could apply for the 60 medications available through individual PAP enrollment, but only indigent patients qualified for the 44 medications available through bulk drug replacement to MDACC.
Patient classifications
Prescription data were extracted from a pharmacy administrative dispensing database; a systematic process was developed to identify case patients (based on financial eligibility) and control patients (similar to case patients with respect to treatments received but were nonusers of PAP programs). Only patients who were potentially eligible for PAPs were included in the study. The case selection was based on MDACC’s determination of a patient’s ability to pay, referred to as credit rating, at the time of a patient’s registration at the institution. Regardless of health insurance status, patients who had a low credit rating (responsible for 0%– 50% of their charges) were classified as being potentially eligible for PAPs. Patients with low credit ratings also included those who were indigent. The control selection identified a set of insured patients, including those with high credit ratings (responsible for 100% of their charges), who had been referred for special financial assistance to obtain specific medications through PAPs.
To be included in the study, patients identified based on a low credit rating had to receive at least 1 of the 104 medications through a PAP to be classified as a PAP user; these patients could receive other medications through traditional payment. PAP nonusers had to receive at least 1 of the 104 medications associated with PAPs through traditional payment or other third-party source, not through a PAP. Patients who had been referred for special assistance had to receive one or more of the PAP medications initially requested from a PAP.
For a drug to be verified as a PAP prescription medication, the pharmacy record could not have documentation of third-party payer or patient payment for that medication. The only exception made for payer and patient payment was for prescription medications provided by one particular pharmaceutical company, which required a $10 copayment for its PAP medications. Once PAP and non-PAP prescriptions were verified, they were aggregated by a unique patient identifier to yield prescriptionuse data for individual patients who were categorized as PAP users versus PAP nonusers.
Patient characteristics
Data on patient gender, race/ethnicity, age, insurance status, and primary cancer site were extracted. Race/ ethnicity was categorized as white, black, Hispanic, or Asian/other. Age was calculated as of July 1, 2006, from the patient’s birth date. Insurance status was based on the patient’s insurance status at the time of registration at MDACC and categorized as follows: no insurance (include self-payers and patients referred from the county public hospital), Medicare, Medicaid, or any of a variety of private/commercial insurances. Private insurances were combined into one category. Information on each patient’s primary cancer site was categorized as blood, breast, genitourinary, head and neck, or other (primarily brain, central nervous system, and an unknown primary site).
The gender and insurance variables had some missing data. When there were conflicting data for a particular patient’s gender, we coded gender as missing. When the insurance type was missing, data on the patient’s insurance status at the time of registration at MDACC were retrieved from MDACC’s financial department.
Prescription medication fills
Data on the prescription medication name (generic or brand) and institutional billing charges per fill were extracted from pharmacy records. Prescriptions were aggregated by generic and brand names, regardless of strength, dosage form, or method of administration, to identify the 20 most frequently dispensed medications overall and for the treatment of cancer. We then used Rxlist.com (www.rxlist.com), an online medication reference program, to identify each medication’s clinical indication(s). For example, the brand name medication Zofran would be aggregated with its generic, ondansetron, and would be considered as one medication indicated for nausea and vomiting.
We extracted patient billing charge per medication fill in dollars by the date of pickup in the outpatient pharmacy. Patient billing charge included patient copayments and did not include any payments from the patient’s payer or health plan. If the billing charge was missing for a medication fill, we applied a comparable charge from a prescription medication of the same name, dosage, quantity, date of pickup, and patient insurance status. When quantity, date, or patient insurance status differed, the lowest available charge was used. All charges were adjusted to the year 2008 using the US Bureau of Labor’s Annual Producer Price Index for pharmaceutical preparation and manufacturing.7
Data analysis
For patient-level analyses, a PAP user was a patient who received at least one medication through a PAP during the study period. We used descriptive statistics to compare patient characteristics of PAP users versus PAP nonusers. Next, we conducted separate unadjusted binary logit regression analyses (interpreted with odds ratios [ORs] and 95% confidence intervals [CIs]) to estimate the differences in the probability of being a PAP user for each of the patient characteristics. All patient characteristics that were statistically significant at P < 0.20 for the unadjusted analyses8 were included in the final multivariable model. The a priori level of significance was set at P < 0.05 for the multivariable model.
For other analyses conducted at the prescription level, a PAP medication was a medication verified as being provided through a PAP. We used descriptive statistics to compare the 20 most frequently dispensed prescription medications (overall and for anticancer agents specifically) by PAP status and clinical indication. Analyses were conducted in Microsoft Excel and STATA Version 11.9
Results
Study patients and prescription medications
During the 18-month observation period, a monthly mean of 1,550 patients received a monthly total of 19,000 noninvestigational medications in the outpatient pharmacy. Of these patients, 7.5% (n = 1,929) met study eligibility criteria for PAPs and received 1 of the 104 medications provided through PAPs. Thus, there were 979 PAP users and 950 PAP nonusers in the final study population. In total, the study population received 23.3% (n = 77,592) of all outpatient medications administered during this period, of which anticancer agents represented 4% (n = 3,105; Table 1).
Comparison of patient characteristics
In comparison to PAP nonusers, PAP users were, on average, younger (48 vs 52 years), indigent (73% vs 19%), white (50% vs 43%), and covered by Medicaid or were uninsured (75% versus 20%). PAP users also had more prescriptions fills (median = 30 vs 20) during the study period at the institution. Univariate analyses showed that all patient characteristics, except gender, significantly predicted PAP use. Given the strong correlation of indigent and insurance status to PAP use, we conducted post hoc analyses to assess the potential for multicollinearity between the two patient characteristics. The variance inflation factor (VIF = 4.57) did not indicate multicollinearity concerns.
In the adjusted model, patients who were indigent (OR = 16.95; 95% CI: 6.845, 41.960), uninsured (OR = 4.60; 95% CI: 2.118, 9.970), and under the age of 65 years (OR = 2.31; 95% CI: 1.517, 3.509) were 2- to 17- fold more likely than others to be PAP users. Black patients were 31% (P = 0.020) less likely to access PAPs than were white patients.
Overall prescription medication fills The top-20 prescription medication fills from the MDACC outpatient pharmacy differed by PAP user group and PAP status. For PAP users, 88% of the most common medications obtained from PAPs were supportive care agents, including treatments of bacterial infections (n =887 fills; 49/month), antiemetics (n = 492 fills; 27/month), and gastroesophageal reflux disease (n = 492 fills; 27/month). Conversely, treatments for neutropenia and anticoagulation represented nearly half ($1.8 million) of the total charges avoided through PAPs to PAP users ($3.9 million). The most common medications not obtained from PAPs were for treatment of pain (PAP users = 292 fills/month, nonusers = 218 fills/month), versus only 13 fills/ month for pain medications from PAPs. Medications indicated to treat pain and nausea/vomiting accounted for the largest proportion of charges for medications not filled by PAPs for both PAP users and nonusers.
Anticancer agent prescription fills
For both PAP users and nonusers, the top-20 anticancer oral agent fills represented 93% (n = 2,892 of 3,105) of all anticancer oral fills (Table 6), with 16% (n = 454) of these oral fills being provided through PAPs. Among PAP users, anticancer agents from PAPs accounted for 40% of their total charges and 35% of the total number of agents. Temozolomide (Temodar; mean charge/fill = $3,346) represented the highest amount of total charges ($220,857) from PAPs, whereas imatinib (Gleevec; mean charge/fill = $5,372) and dasatinib (Sprycel; mean charge/fill = $5,221) accounted for the highest average charges per fill. Anastrozole (Arimidex; n = 178 fills; 10/month), capecitabine (Xeloda; n = 91 fills; 5/month), and temozolomide (n = 66 fills; 4/month) accounted for 70% of agents from PAPs. PAP users who were given bicalutamide received 100% of those agents from PAPs. Five of the seven oral anticancer agents with no fills from PAPs had initial US Food and Drug Administration approval years before 2000.
Discussion
At MDACC, PAPs are designed to help cancer patients overcome financial barriers to accessing oral supportive and anticancer agents. Over an 18-month observation period, less than 5% of the cancer patients at MDACC who received prescription medications from the outpatient pharmacy were enrolled in a PAP— and these PAPs provided 13% of their medication fills, representing an annualized $3.6 million in pharmaceutical expenditures. In interpreting our findings, several factors should be considered.
Oral anticancer agents accounted for 4% of all prescription medication fills during the study period. Comparatively, an analysis of the 2007 National Ambulatory Medical Care Survey showed that less than 1% of cancer patients were prescribed at least one oral anticancer agent.10 This finding indicates that both nationally and at MDACC, chemotherapy continues to be largely provided parenterally, as there is more of a financial benefit from intravenous therapies that are often reimbursed by insurers as well as PAPs.
In the outpatient pharmacy at MDACC, PAPs provided nearly onethird of oral anticancer fills for PAP users—totaling a mean of $500,000 per month in expenditures. However, three agents, anastrozole (for breast cancer), capecitabine (for breast and GI cancers, primarily), and temozolomide (for brain tumors) accounted for 75% of all of the anticancer agents provided by PAPs. We also found that pharmaceutical companies provided expensive newer, targeted, anticancer agents (primarily dasatinib and imatinib, the two agents with the greatest pharmaceutical per-person expenditures by the PAP program) through PAPs.
Although PAPs filled a strong and focused need for a small number of oral chemotherapy agents for some individuals with breast, GI, and brain cancers, they did not provide much benefit for a wide range of supportive care agents, particularly those that are schedule C and are used to treatcancer pain. Pain is the most prevalent symptom reported by cancer patients, 11 but there were few schedule C pain medications among the most common medications provided through PAPs. These substances are generally not provided by PAPs because of legal and substance abuse concerns.12 However, these medications were commonly prescribed to PAP users and PAP nonusers alike, outside of the PAP program. It would be important to evaluate the comparative success in treating pain among cancer patients at MDACC who receive a limited array of pain medications from PAPs (usually agents that are not substance-controlled by the Drug Enforcement Administration) versus treatment of pain experienced by patients whose medications are not reimbursed by PAPs.
We found that being younger than 65 years old, being indigent, and having no health insurance were the strongest predictors of using a PAP. This finding was expected, given that US adults younger than age 65 are ineligible for outpatient prescription medication coverage through Medicare Part D. However, contrary to expectations, about 45% of PAP users had either private or governmentsupplied health insurance. Because it is not uncommon for cancer patients to endure economic hardship (including bankruptcy) when trying to finance their care,13 healthcare professionals could recommend PAPs and other relevant assistance programs to all of their cancer patients.
With the expansion of health insurance through the Patient Protection and Affordable Health Care Act of 2010, it is hoped that the need for cancer patients to enroll in PAPs will be diminished; yet, given the reality of the high cost of anticancer agents, reimbursement policies for these agents, and tiered formularies among insurers leading to high outof- pocket costs for patients, the need for PAPs is likely to remain. PAPs can be a viable option for some patients, but healthcare professionals should be aware that there are a number of concerns about these programs, including their complex and burdensome application process and often limited variety of available drugs.14
This study is not without its limitations. First, we may have underestimated our sample of PAP patients due to the fact that MDACC did not electronically or systematically track the use of PAPs within its pharmacy database at the time of the study. The institution is in the process of developing such a system.
Second, the data used in this study cannot be assumed to reflect a “closed pharmacy” setting because some patients, particularly those who have health insurance with prescription medication coverage, may have received some of their medications from outside pharmacies.
Third, because insurance status is not necessarily a static characteristic, insurance status in this study was classified based on that at the time of registration with MDACC’s financial department, and no account was taken of changes that might have occurred.
Last, our results are not necessarily generalizable to all cancer populations, time periods, or settings. Cancer patients treated in academic centers such as MDACC may differ from those who are treated in community settings. In particular, fewer than 10% of patients at MDACC qualified for indigent financial assistance in 2007,15 which is likely to have impacted the number of patients who were potentially eligible for PAPs. It is also likely that had our study been conducted prior to the implementation of Medicare Part D, our sample of PAP patients would have been older. Nevertheless, our results may be generalizable to cancer patients receiving care in other academic cancer centers.
Conclusion This study builds upon a previous description of implementing PAPs in a comprehensive cancer center16 as well as contributes to our limited knowledge of the use of PAPs among cancer patients.6 Future studies should prospectively examine cancer patients’ experiences and satisfaction with PAPs from the process of applying to the point of receiving requested therapies and evaluate the effect of PAPs on cancer outcomes in various care settings. Multidisciplinary teams, including pharmacists and clinicians, should establish and recommend valid and relevant clinical endpoints for researchers to use in effectiveness studies of PAPs and cancer patients, particularly as they relate to oral anticancer agent use. Given that these oral agents represent more than 25% of cancer therapies in development,17 future studies of PAPs are ideal for evaluating concerns of accessibility, affordability, and compliance related to these agents.
MDACC is a unique resource for observers of PAPs, as it is the largest cancer center in the United States. However, few cancer patients at MDACC were eligible for and accessed PAPs in the outpatient pharmacy. Although smaller cancer centers may not be able to devote the same degree of financial and personnel resources to their patients as does MDACC, these centers could seek to build relationships with specific pharmaceutical companies that provide PAPs for the oral anticancer and supportive care therapies most commonly prescribed and administered at their centers. Scarce resources could also be utilized in other ways, such as by developing public-private risk pools for establishment of indigent care funds.
Acknowledgments: The authors thank Chun Feng, Jason Lau, and Oliver Max for their special assistance; Dr. Phoenix Do for her study design recommendations; and Karyn Popham for her editorial support. They especially thank Rebecca Arbuckle, RPh, for her support of this project. At the time of the study, Dr. Felder was supported by a Predoctoral Fellowship from The University of Texas School of Public Health Cancer Education and Career Development Program, funded by National Cancer Institute/NIH Grant R25-CA-57712-17.
References
1. Hede K. Increase in oral cancer drugs raises thorny issues for oncology practices. J Natl Cancer Inst 2009;101:1534–1536.
2. Chisholm MA, DiPiro JT. Pharmaceutical manufacturer assistance programs. Arch Intern Med 2002;162:780–784.
3. Duke KS, Raube K, Lipton HL. Patientassistance programs: assessment of and use by safety-net clinics. Am J Health Syst Pharm 2005;62:726–731.
4. Pisu M, Richman J, Allison JJ, Williams OD, Kiefe CI. Pharmaceuticals companies’ medication assistance programs: potentially useful but too burdensome to use? South Med J 2009;102:139–144.
5. Felder TM, Palmer NR, Lal LS, Mullen PD. What is the evidence for pharmaceutical patient assistance programs? a systematic review. J Health Care Poor Underserved 2011;22:24–49.
6. Meropol NJ, Schrag D, Smith TJ, et al. American Society of Clinical Oncology guidance statement: the cost of cancer care. J Clin Oncol 2009;27:3868–3874.
7. United States Department of Labor, Bureau of Labor Statistics: Producer Price Index Industry Data—Pharmaceutical Preparation & Manufacturing; 2010 [updated April 26, 2010]. http://www.bls.gov/ppi/data.htm. Accessed June 22, 2011.
8. Hosmer DW, Lemeshow S. Applied Logistic
Logistic
Regression. New York: Wiley; 2000. 9. StataCorp LP. STATA statistical software. 2009; Release 11.
10. Arora S. Use of oral chemotherapeutic medications in non-traditional ambulatory settings; 2009. http://digarchive.library.vcu. edu/dspace/bitstream/10156/2711/1/Thesis_ MPH_sameer.pdf. Accessed June 22, 2011.
11. Cherny NI. The management of cancer pain. CA Cancer J Clin 2000;50:70–116.
12. Williams K. Accessing patient assistance programs to meet clients’ medication needs. J Am Acad Nurse Pract 2000;12:233– 235.
13. National Survey of Households Affected by Cancer: Kaiser Family Foundation; 2006 [updated November 2006]. http://kff.org/kaiserpolls/ upload/7591.pdf. Accessed June 22, 2011.
14. Choudhry NK, Lee JL, Agnew-Blais J, Corcoran C, Shrank WH. Drug company- sponsored patient assistance programs: a viable safety net? Health Aff (Millwood) 2009;28:827–834.
15. Ackerman T. M.D. Anderson submits its records on charitable care: cancer center hopes to quell Iowa senator’s investigation. Houston Chronicle. October 9, 2008. http:// www.chron.com/disp/story.mpl/metropolitan/ 6050254.html. Accessed June 22, 2011.
16. Johnson PE. Patient assistance programs and patient advocacy foundations: alternatives for obtaining prescription medications when insurance fails. Am J Health Syst Pharm 2006;63(21 suppl 7):S13–S17.
17. Weingart SA, Brown E, Bach PB, et al. National Comprehensive Cancer Network task force report: oral chemotherapy. JNCCN 2008;6(suppl 3):S1–S25.
Oral anticancer and supportive care agents administered to cancer patients are costly and are associated with large copayment requirements or are often not fully reimbursed by private health insurers or Medicare.1 To facilitate access to oral medications, pharmaceutical manufacturers have developed patient assistance programs (PAPs) that provide selected oral medications at no or reduced cost to financially eligible patients. Eligibility criteria, application processes, and program administration for PAPs differ by manufacturer and by product, which can ultimately present logistical barriers.2–4 A systematic review of PAPs found improvements in disease indicator outcomes for patients with common chronic diseases who access these programs.5 However, knowledge about the use of PAPs among cancer patients is limited.6
The University of Texas MD Anderson Cancer Center (MDACC), the largest tertiary care cancer center in the country, has developed a systematic approach to administering a large number of PAPs. In 1996, the MDACC established an institutional program staffed by hospital pharmacy personnel, who navigate cancer patients through PAPs in inpatient and outpatient settings. This program removes the operational and administrative barriers often experienced by patients in smaller clinical settings.
Cancer patients eligible for PAPs at MDACC include those who are uninsured, those who are underinsured, those whose pharmacy benefit limits have been reached, and those whose private health or government insurance has denied coverage of certain oral medications. For example, the Texas Medicaid program limits its low-income beneficiaries to three prescriptions per month, which may lead some of them, particularly those with cancer, to require additional medication assistance through PAPs. As of April 2008, this institutional program established formal relationships with 29 pharmaceutical companies that provide 104 therapeutic or supportive care agents through PAPs to eligible cancer patients in the MDACC outpatient pharmacy.
Methods
Data source
Approval for this study was obtained from the MDACC Institutional Review Board. We conducted a retrospective, secondary analysis of noninvestigational prescription medications from the outpatient pharmacy at MDACC. Data from July 1, 2006, to December 31, 2007, were extracted from computerized pharmacy, medical, and cancer registry databases at MDACC. Prescriptions had to include both patient medical record and social security numbers to validate the patient’s identity as well as the date of pickup to validate that the medication had been dispensed during the study period. When the date of pickup was missing but billing was documented, the date the medication was dispensed was used as the pickup date. All data were de-identified prior to analysis.
PAPs
Prescriptions for oral medications were available to financially eligible individuals via two types of PAPs at MDACC: individual enrollment (60 distinct medications) and bulk drug replacement (44 distinct medications). Individual enrollment required that an eligible patient apply directly to a pharmaceutical company’s PAP for the medication (s) needed. Once approved, the requested medication was mailed directly to the patient or dispensed in the MDACC pharmacy. Given the purpose of this study, we were only interested in those PAP prescription medications dispensed at the outpatient pharmacy.
Bulk replacement PAPs provide available prescription medications in bulk quantities on a monthly (in some cases quarterly) basis to MDACC’s pharmacy to replace medications dispensed to patients who were classified as “indigent” by MDACC-established criteria. Financially indigent patients included those who were Texas residents, uninsured or insured by Medicaid, and not responsible for charges billed to MDACC. All eligible patients could apply for the 60 medications available through individual PAP enrollment, but only indigent patients qualified for the 44 medications available through bulk drug replacement to MDACC.
Patient classifications
Prescription data were extracted from a pharmacy administrative dispensing database; a systematic process was developed to identify case patients (based on financial eligibility) and control patients (similar to case patients with respect to treatments received but were nonusers of PAP programs). Only patients who were potentially eligible for PAPs were included in the study. The case selection was based on MDACC’s determination of a patient’s ability to pay, referred to as credit rating, at the time of a patient’s registration at the institution. Regardless of health insurance status, patients who had a low credit rating (responsible for 0%– 50% of their charges) were classified as being potentially eligible for PAPs. Patients with low credit ratings also included those who were indigent. The control selection identified a set of insured patients, including those with high credit ratings (responsible for 100% of their charges), who had been referred for special financial assistance to obtain specific medications through PAPs.
To be included in the study, patients identified based on a low credit rating had to receive at least 1 of the 104 medications through a PAP to be classified as a PAP user; these patients could receive other medications through traditional payment. PAP nonusers had to receive at least 1 of the 104 medications associated with PAPs through traditional payment or other third-party source, not through a PAP. Patients who had been referred for special assistance had to receive one or more of the PAP medications initially requested from a PAP.
For a drug to be verified as a PAP prescription medication, the pharmacy record could not have documentation of third-party payer or patient payment for that medication. The only exception made for payer and patient payment was for prescription medications provided by one particular pharmaceutical company, which required a $10 copayment for its PAP medications. Once PAP and non-PAP prescriptions were verified, they were aggregated by a unique patient identifier to yield prescriptionuse data for individual patients who were categorized as PAP users versus PAP nonusers.
Patient characteristics
Data on patient gender, race/ethnicity, age, insurance status, and primary cancer site were extracted. Race/ ethnicity was categorized as white, black, Hispanic, or Asian/other. Age was calculated as of July 1, 2006, from the patient’s birth date. Insurance status was based on the patient’s insurance status at the time of registration at MDACC and categorized as follows: no insurance (include self-payers and patients referred from the county public hospital), Medicare, Medicaid, or any of a variety of private/commercial insurances. Private insurances were combined into one category. Information on each patient’s primary cancer site was categorized as blood, breast, genitourinary, head and neck, or other (primarily brain, central nervous system, and an unknown primary site).
The gender and insurance variables had some missing data. When there were conflicting data for a particular patient’s gender, we coded gender as missing. When the insurance type was missing, data on the patient’s insurance status at the time of registration at MDACC were retrieved from MDACC’s financial department.
Prescription medication fills
Data on the prescription medication name (generic or brand) and institutional billing charges per fill were extracted from pharmacy records. Prescriptions were aggregated by generic and brand names, regardless of strength, dosage form, or method of administration, to identify the 20 most frequently dispensed medications overall and for the treatment of cancer. We then used Rxlist.com (www.rxlist.com), an online medication reference program, to identify each medication’s clinical indication(s). For example, the brand name medication Zofran would be aggregated with its generic, ondansetron, and would be considered as one medication indicated for nausea and vomiting.
We extracted patient billing charge per medication fill in dollars by the date of pickup in the outpatient pharmacy. Patient billing charge included patient copayments and did not include any payments from the patient’s payer or health plan. If the billing charge was missing for a medication fill, we applied a comparable charge from a prescription medication of the same name, dosage, quantity, date of pickup, and patient insurance status. When quantity, date, or patient insurance status differed, the lowest available charge was used. All charges were adjusted to the year 2008 using the US Bureau of Labor’s Annual Producer Price Index for pharmaceutical preparation and manufacturing.7
Data analysis
For patient-level analyses, a PAP user was a patient who received at least one medication through a PAP during the study period. We used descriptive statistics to compare patient characteristics of PAP users versus PAP nonusers. Next, we conducted separate unadjusted binary logit regression analyses (interpreted with odds ratios [ORs] and 95% confidence intervals [CIs]) to estimate the differences in the probability of being a PAP user for each of the patient characteristics. All patient characteristics that were statistically significant at P < 0.20 for the unadjusted analyses8 were included in the final multivariable model. The a priori level of significance was set at P < 0.05 for the multivariable model.
For other analyses conducted at the prescription level, a PAP medication was a medication verified as being provided through a PAP. We used descriptive statistics to compare the 20 most frequently dispensed prescription medications (overall and for anticancer agents specifically) by PAP status and clinical indication. Analyses were conducted in Microsoft Excel and STATA Version 11.9
Results
Study patients and prescription medications
During the 18-month observation period, a monthly mean of 1,550 patients received a monthly total of 19,000 noninvestigational medications in the outpatient pharmacy. Of these patients, 7.5% (n = 1,929) met study eligibility criteria for PAPs and received 1 of the 104 medications provided through PAPs. Thus, there were 979 PAP users and 950 PAP nonusers in the final study population. In total, the study population received 23.3% (n = 77,592) of all outpatient medications administered during this period, of which anticancer agents represented 4% (n = 3,105; Table 1).
Comparison of patient characteristics
In comparison to PAP nonusers, PAP users were, on average, younger (48 vs 52 years), indigent (73% vs 19%), white (50% vs 43%), and covered by Medicaid or were uninsured (75% versus 20%). PAP users also had more prescriptions fills (median = 30 vs 20) during the study period at the institution. Univariate analyses showed that all patient characteristics, except gender, significantly predicted PAP use. Given the strong correlation of indigent and insurance status to PAP use, we conducted post hoc analyses to assess the potential for multicollinearity between the two patient characteristics. The variance inflation factor (VIF = 4.57) did not indicate multicollinearity concerns.
In the adjusted model, patients who were indigent (OR = 16.95; 95% CI: 6.845, 41.960), uninsured (OR = 4.60; 95% CI: 2.118, 9.970), and under the age of 65 years (OR = 2.31; 95% CI: 1.517, 3.509) were 2- to 17- fold more likely than others to be PAP users. Black patients were 31% (P = 0.020) less likely to access PAPs than were white patients.
Overall prescription medication fills The top-20 prescription medication fills from the MDACC outpatient pharmacy differed by PAP user group and PAP status. For PAP users, 88% of the most common medications obtained from PAPs were supportive care agents, including treatments of bacterial infections (n =887 fills; 49/month), antiemetics (n = 492 fills; 27/month), and gastroesophageal reflux disease (n = 492 fills; 27/month). Conversely, treatments for neutropenia and anticoagulation represented nearly half ($1.8 million) of the total charges avoided through PAPs to PAP users ($3.9 million). The most common medications not obtained from PAPs were for treatment of pain (PAP users = 292 fills/month, nonusers = 218 fills/month), versus only 13 fills/ month for pain medications from PAPs. Medications indicated to treat pain and nausea/vomiting accounted for the largest proportion of charges for medications not filled by PAPs for both PAP users and nonusers.
Anticancer agent prescription fills
For both PAP users and nonusers, the top-20 anticancer oral agent fills represented 93% (n = 2,892 of 3,105) of all anticancer oral fills (Table 6), with 16% (n = 454) of these oral fills being provided through PAPs. Among PAP users, anticancer agents from PAPs accounted for 40% of their total charges and 35% of the total number of agents. Temozolomide (Temodar; mean charge/fill = $3,346) represented the highest amount of total charges ($220,857) from PAPs, whereas imatinib (Gleevec; mean charge/fill = $5,372) and dasatinib (Sprycel; mean charge/fill = $5,221) accounted for the highest average charges per fill. Anastrozole (Arimidex; n = 178 fills; 10/month), capecitabine (Xeloda; n = 91 fills; 5/month), and temozolomide (n = 66 fills; 4/month) accounted for 70% of agents from PAPs. PAP users who were given bicalutamide received 100% of those agents from PAPs. Five of the seven oral anticancer agents with no fills from PAPs had initial US Food and Drug Administration approval years before 2000.
Discussion
At MDACC, PAPs are designed to help cancer patients overcome financial barriers to accessing oral supportive and anticancer agents. Over an 18-month observation period, less than 5% of the cancer patients at MDACC who received prescription medications from the outpatient pharmacy were enrolled in a PAP— and these PAPs provided 13% of their medication fills, representing an annualized $3.6 million in pharmaceutical expenditures. In interpreting our findings, several factors should be considered.
Oral anticancer agents accounted for 4% of all prescription medication fills during the study period. Comparatively, an analysis of the 2007 National Ambulatory Medical Care Survey showed that less than 1% of cancer patients were prescribed at least one oral anticancer agent.10 This finding indicates that both nationally and at MDACC, chemotherapy continues to be largely provided parenterally, as there is more of a financial benefit from intravenous therapies that are often reimbursed by insurers as well as PAPs.
In the outpatient pharmacy at MDACC, PAPs provided nearly onethird of oral anticancer fills for PAP users—totaling a mean of $500,000 per month in expenditures. However, three agents, anastrozole (for breast cancer), capecitabine (for breast and GI cancers, primarily), and temozolomide (for brain tumors) accounted for 75% of all of the anticancer agents provided by PAPs. We also found that pharmaceutical companies provided expensive newer, targeted, anticancer agents (primarily dasatinib and imatinib, the two agents with the greatest pharmaceutical per-person expenditures by the PAP program) through PAPs.
Although PAPs filled a strong and focused need for a small number of oral chemotherapy agents for some individuals with breast, GI, and brain cancers, they did not provide much benefit for a wide range of supportive care agents, particularly those that are schedule C and are used to treatcancer pain. Pain is the most prevalent symptom reported by cancer patients, 11 but there were few schedule C pain medications among the most common medications provided through PAPs. These substances are generally not provided by PAPs because of legal and substance abuse concerns.12 However, these medications were commonly prescribed to PAP users and PAP nonusers alike, outside of the PAP program. It would be important to evaluate the comparative success in treating pain among cancer patients at MDACC who receive a limited array of pain medications from PAPs (usually agents that are not substance-controlled by the Drug Enforcement Administration) versus treatment of pain experienced by patients whose medications are not reimbursed by PAPs.
We found that being younger than 65 years old, being indigent, and having no health insurance were the strongest predictors of using a PAP. This finding was expected, given that US adults younger than age 65 are ineligible for outpatient prescription medication coverage through Medicare Part D. However, contrary to expectations, about 45% of PAP users had either private or governmentsupplied health insurance. Because it is not uncommon for cancer patients to endure economic hardship (including bankruptcy) when trying to finance their care,13 healthcare professionals could recommend PAPs and other relevant assistance programs to all of their cancer patients.
With the expansion of health insurance through the Patient Protection and Affordable Health Care Act of 2010, it is hoped that the need for cancer patients to enroll in PAPs will be diminished; yet, given the reality of the high cost of anticancer agents, reimbursement policies for these agents, and tiered formularies among insurers leading to high outof- pocket costs for patients, the need for PAPs is likely to remain. PAPs can be a viable option for some patients, but healthcare professionals should be aware that there are a number of concerns about these programs, including their complex and burdensome application process and often limited variety of available drugs.14
This study is not without its limitations. First, we may have underestimated our sample of PAP patients due to the fact that MDACC did not electronically or systematically track the use of PAPs within its pharmacy database at the time of the study. The institution is in the process of developing such a system.
Second, the data used in this study cannot be assumed to reflect a “closed pharmacy” setting because some patients, particularly those who have health insurance with prescription medication coverage, may have received some of their medications from outside pharmacies.
Third, because insurance status is not necessarily a static characteristic, insurance status in this study was classified based on that at the time of registration with MDACC’s financial department, and no account was taken of changes that might have occurred.
Last, our results are not necessarily generalizable to all cancer populations, time periods, or settings. Cancer patients treated in academic centers such as MDACC may differ from those who are treated in community settings. In particular, fewer than 10% of patients at MDACC qualified for indigent financial assistance in 2007,15 which is likely to have impacted the number of patients who were potentially eligible for PAPs. It is also likely that had our study been conducted prior to the implementation of Medicare Part D, our sample of PAP patients would have been older. Nevertheless, our results may be generalizable to cancer patients receiving care in other academic cancer centers.
Conclusion This study builds upon a previous description of implementing PAPs in a comprehensive cancer center16 as well as contributes to our limited knowledge of the use of PAPs among cancer patients.6 Future studies should prospectively examine cancer patients’ experiences and satisfaction with PAPs from the process of applying to the point of receiving requested therapies and evaluate the effect of PAPs on cancer outcomes in various care settings. Multidisciplinary teams, including pharmacists and clinicians, should establish and recommend valid and relevant clinical endpoints for researchers to use in effectiveness studies of PAPs and cancer patients, particularly as they relate to oral anticancer agent use. Given that these oral agents represent more than 25% of cancer therapies in development,17 future studies of PAPs are ideal for evaluating concerns of accessibility, affordability, and compliance related to these agents.
MDACC is a unique resource for observers of PAPs, as it is the largest cancer center in the United States. However, few cancer patients at MDACC were eligible for and accessed PAPs in the outpatient pharmacy. Although smaller cancer centers may not be able to devote the same degree of financial and personnel resources to their patients as does MDACC, these centers could seek to build relationships with specific pharmaceutical companies that provide PAPs for the oral anticancer and supportive care therapies most commonly prescribed and administered at their centers. Scarce resources could also be utilized in other ways, such as by developing public-private risk pools for establishment of indigent care funds.
Acknowledgments: The authors thank Chun Feng, Jason Lau, and Oliver Max for their special assistance; Dr. Phoenix Do for her study design recommendations; and Karyn Popham for her editorial support. They especially thank Rebecca Arbuckle, RPh, for her support of this project. At the time of the study, Dr. Felder was supported by a Predoctoral Fellowship from The University of Texas School of Public Health Cancer Education and Career Development Program, funded by National Cancer Institute/NIH Grant R25-CA-57712-17.
References
1. Hede K. Increase in oral cancer drugs raises thorny issues for oncology practices. J Natl Cancer Inst 2009;101:1534–1536.
2. Chisholm MA, DiPiro JT. Pharmaceutical manufacturer assistance programs. Arch Intern Med 2002;162:780–784.
3. Duke KS, Raube K, Lipton HL. Patientassistance programs: assessment of and use by safety-net clinics. Am J Health Syst Pharm 2005;62:726–731.
4. Pisu M, Richman J, Allison JJ, Williams OD, Kiefe CI. Pharmaceuticals companies’ medication assistance programs: potentially useful but too burdensome to use? South Med J 2009;102:139–144.
5. Felder TM, Palmer NR, Lal LS, Mullen PD. What is the evidence for pharmaceutical patient assistance programs? a systematic review. J Health Care Poor Underserved 2011;22:24–49.
6. Meropol NJ, Schrag D, Smith TJ, et al. American Society of Clinical Oncology guidance statement: the cost of cancer care. J Clin Oncol 2009;27:3868–3874.
7. United States Department of Labor, Bureau of Labor Statistics: Producer Price Index Industry Data—Pharmaceutical Preparation & Manufacturing; 2010 [updated April 26, 2010]. http://www.bls.gov/ppi/data.htm. Accessed June 22, 2011.
8. Hosmer DW, Lemeshow S. Applied Logistic
Logistic
Regression. New York: Wiley; 2000. 9. StataCorp LP. STATA statistical software. 2009; Release 11.
10. Arora S. Use of oral chemotherapeutic medications in non-traditional ambulatory settings; 2009. http://digarchive.library.vcu. edu/dspace/bitstream/10156/2711/1/Thesis_ MPH_sameer.pdf. Accessed June 22, 2011.
11. Cherny NI. The management of cancer pain. CA Cancer J Clin 2000;50:70–116.
12. Williams K. Accessing patient assistance programs to meet clients’ medication needs. J Am Acad Nurse Pract 2000;12:233– 235.
13. National Survey of Households Affected by Cancer: Kaiser Family Foundation; 2006 [updated November 2006]. http://kff.org/kaiserpolls/ upload/7591.pdf. Accessed June 22, 2011.
14. Choudhry NK, Lee JL, Agnew-Blais J, Corcoran C, Shrank WH. Drug company- sponsored patient assistance programs: a viable safety net? Health Aff (Millwood) 2009;28:827–834.
15. Ackerman T. M.D. Anderson submits its records on charitable care: cancer center hopes to quell Iowa senator’s investigation. Houston Chronicle. October 9, 2008. http:// www.chron.com/disp/story.mpl/metropolitan/ 6050254.html. Accessed June 22, 2011.
16. Johnson PE. Patient assistance programs and patient advocacy foundations: alternatives for obtaining prescription medications when insurance fails. Am J Health Syst Pharm 2006;63(21 suppl 7):S13–S17.
17. Weingart SA, Brown E, Bach PB, et al. National Comprehensive Cancer Network task force report: oral chemotherapy. JNCCN 2008;6(suppl 3):S1–S25.
Oral anticancer and supportive care agents administered to cancer patients are costly and are associated with large copayment requirements or are often not fully reimbursed by private health insurers or Medicare.1 To facilitate access to oral medications, pharmaceutical manufacturers have developed patient assistance programs (PAPs) that provide selected oral medications at no or reduced cost to financially eligible patients. Eligibility criteria, application processes, and program administration for PAPs differ by manufacturer and by product, which can ultimately present logistical barriers.2–4 A systematic review of PAPs found improvements in disease indicator outcomes for patients with common chronic diseases who access these programs.5 However, knowledge about the use of PAPs among cancer patients is limited.6
The University of Texas MD Anderson Cancer Center (MDACC), the largest tertiary care cancer center in the country, has developed a systematic approach to administering a large number of PAPs. In 1996, the MDACC established an institutional program staffed by hospital pharmacy personnel, who navigate cancer patients through PAPs in inpatient and outpatient settings. This program removes the operational and administrative barriers often experienced by patients in smaller clinical settings.
Cancer patients eligible for PAPs at MDACC include those who are uninsured, those who are underinsured, those whose pharmacy benefit limits have been reached, and those whose private health or government insurance has denied coverage of certain oral medications. For example, the Texas Medicaid program limits its low-income beneficiaries to three prescriptions per month, which may lead some of them, particularly those with cancer, to require additional medication assistance through PAPs. As of April 2008, this institutional program established formal relationships with 29 pharmaceutical companies that provide 104 therapeutic or supportive care agents through PAPs to eligible cancer patients in the MDACC outpatient pharmacy.
Methods
Data source
Approval for this study was obtained from the MDACC Institutional Review Board. We conducted a retrospective, secondary analysis of noninvestigational prescription medications from the outpatient pharmacy at MDACC. Data from July 1, 2006, to December 31, 2007, were extracted from computerized pharmacy, medical, and cancer registry databases at MDACC. Prescriptions had to include both patient medical record and social security numbers to validate the patient’s identity as well as the date of pickup to validate that the medication had been dispensed during the study period. When the date of pickup was missing but billing was documented, the date the medication was dispensed was used as the pickup date. All data were de-identified prior to analysis.
PAPs
Prescriptions for oral medications were available to financially eligible individuals via two types of PAPs at MDACC: individual enrollment (60 distinct medications) and bulk drug replacement (44 distinct medications). Individual enrollment required that an eligible patient apply directly to a pharmaceutical company’s PAP for the medication (s) needed. Once approved, the requested medication was mailed directly to the patient or dispensed in the MDACC pharmacy. Given the purpose of this study, we were only interested in those PAP prescription medications dispensed at the outpatient pharmacy.
Bulk replacement PAPs provide available prescription medications in bulk quantities on a monthly (in some cases quarterly) basis to MDACC’s pharmacy to replace medications dispensed to patients who were classified as “indigent” by MDACC-established criteria. Financially indigent patients included those who were Texas residents, uninsured or insured by Medicaid, and not responsible for charges billed to MDACC. All eligible patients could apply for the 60 medications available through individual PAP enrollment, but only indigent patients qualified for the 44 medications available through bulk drug replacement to MDACC.
Patient classifications
Prescription data were extracted from a pharmacy administrative dispensing database; a systematic process was developed to identify case patients (based on financial eligibility) and control patients (similar to case patients with respect to treatments received but were nonusers of PAP programs). Only patients who were potentially eligible for PAPs were included in the study. The case selection was based on MDACC’s determination of a patient’s ability to pay, referred to as credit rating, at the time of a patient’s registration at the institution. Regardless of health insurance status, patients who had a low credit rating (responsible for 0%– 50% of their charges) were classified as being potentially eligible for PAPs. Patients with low credit ratings also included those who were indigent. The control selection identified a set of insured patients, including those with high credit ratings (responsible for 100% of their charges), who had been referred for special financial assistance to obtain specific medications through PAPs.
To be included in the study, patients identified based on a low credit rating had to receive at least 1 of the 104 medications through a PAP to be classified as a PAP user; these patients could receive other medications through traditional payment. PAP nonusers had to receive at least 1 of the 104 medications associated with PAPs through traditional payment or other third-party source, not through a PAP. Patients who had been referred for special assistance had to receive one or more of the PAP medications initially requested from a PAP.
For a drug to be verified as a PAP prescription medication, the pharmacy record could not have documentation of third-party payer or patient payment for that medication. The only exception made for payer and patient payment was for prescription medications provided by one particular pharmaceutical company, which required a $10 copayment for its PAP medications. Once PAP and non-PAP prescriptions were verified, they were aggregated by a unique patient identifier to yield prescriptionuse data for individual patients who were categorized as PAP users versus PAP nonusers.
Patient characteristics
Data on patient gender, race/ethnicity, age, insurance status, and primary cancer site were extracted. Race/ ethnicity was categorized as white, black, Hispanic, or Asian/other. Age was calculated as of July 1, 2006, from the patient’s birth date. Insurance status was based on the patient’s insurance status at the time of registration at MDACC and categorized as follows: no insurance (include self-payers and patients referred from the county public hospital), Medicare, Medicaid, or any of a variety of private/commercial insurances. Private insurances were combined into one category. Information on each patient’s primary cancer site was categorized as blood, breast, genitourinary, head and neck, or other (primarily brain, central nervous system, and an unknown primary site).
The gender and insurance variables had some missing data. When there were conflicting data for a particular patient’s gender, we coded gender as missing. When the insurance type was missing, data on the patient’s insurance status at the time of registration at MDACC were retrieved from MDACC’s financial department.
Prescription medication fills
Data on the prescription medication name (generic or brand) and institutional billing charges per fill were extracted from pharmacy records. Prescriptions were aggregated by generic and brand names, regardless of strength, dosage form, or method of administration, to identify the 20 most frequently dispensed medications overall and for the treatment of cancer. We then used Rxlist.com (www.rxlist.com), an online medication reference program, to identify each medication’s clinical indication(s). For example, the brand name medication Zofran would be aggregated with its generic, ondansetron, and would be considered as one medication indicated for nausea and vomiting.
We extracted patient billing charge per medication fill in dollars by the date of pickup in the outpatient pharmacy. Patient billing charge included patient copayments and did not include any payments from the patient’s payer or health plan. If the billing charge was missing for a medication fill, we applied a comparable charge from a prescription medication of the same name, dosage, quantity, date of pickup, and patient insurance status. When quantity, date, or patient insurance status differed, the lowest available charge was used. All charges were adjusted to the year 2008 using the US Bureau of Labor’s Annual Producer Price Index for pharmaceutical preparation and manufacturing.7
Data analysis
For patient-level analyses, a PAP user was a patient who received at least one medication through a PAP during the study period. We used descriptive statistics to compare patient characteristics of PAP users versus PAP nonusers. Next, we conducted separate unadjusted binary logit regression analyses (interpreted with odds ratios [ORs] and 95% confidence intervals [CIs]) to estimate the differences in the probability of being a PAP user for each of the patient characteristics. All patient characteristics that were statistically significant at P < 0.20 for the unadjusted analyses8 were included in the final multivariable model. The a priori level of significance was set at P < 0.05 for the multivariable model.
For other analyses conducted at the prescription level, a PAP medication was a medication verified as being provided through a PAP. We used descriptive statistics to compare the 20 most frequently dispensed prescription medications (overall and for anticancer agents specifically) by PAP status and clinical indication. Analyses were conducted in Microsoft Excel and STATA Version 11.9
Results
Study patients and prescription medications
During the 18-month observation period, a monthly mean of 1,550 patients received a monthly total of 19,000 noninvestigational medications in the outpatient pharmacy. Of these patients, 7.5% (n = 1,929) met study eligibility criteria for PAPs and received 1 of the 104 medications provided through PAPs. Thus, there were 979 PAP users and 950 PAP nonusers in the final study population. In total, the study population received 23.3% (n = 77,592) of all outpatient medications administered during this period, of which anticancer agents represented 4% (n = 3,105; Table 1).
Comparison of patient characteristics
In comparison to PAP nonusers, PAP users were, on average, younger (48 vs 52 years), indigent (73% vs 19%), white (50% vs 43%), and covered by Medicaid or were uninsured (75% versus 20%). PAP users also had more prescriptions fills (median = 30 vs 20) during the study period at the institution. Univariate analyses showed that all patient characteristics, except gender, significantly predicted PAP use. Given the strong correlation of indigent and insurance status to PAP use, we conducted post hoc analyses to assess the potential for multicollinearity between the two patient characteristics. The variance inflation factor (VIF = 4.57) did not indicate multicollinearity concerns.
In the adjusted model, patients who were indigent (OR = 16.95; 95% CI: 6.845, 41.960), uninsured (OR = 4.60; 95% CI: 2.118, 9.970), and under the age of 65 years (OR = 2.31; 95% CI: 1.517, 3.509) were 2- to 17- fold more likely than others to be PAP users. Black patients were 31% (P = 0.020) less likely to access PAPs than were white patients.
Overall prescription medication fills The top-20 prescription medication fills from the MDACC outpatient pharmacy differed by PAP user group and PAP status. For PAP users, 88% of the most common medications obtained from PAPs were supportive care agents, including treatments of bacterial infections (n =887 fills; 49/month), antiemetics (n = 492 fills; 27/month), and gastroesophageal reflux disease (n = 492 fills; 27/month). Conversely, treatments for neutropenia and anticoagulation represented nearly half ($1.8 million) of the total charges avoided through PAPs to PAP users ($3.9 million). The most common medications not obtained from PAPs were for treatment of pain (PAP users = 292 fills/month, nonusers = 218 fills/month), versus only 13 fills/ month for pain medications from PAPs. Medications indicated to treat pain and nausea/vomiting accounted for the largest proportion of charges for medications not filled by PAPs for both PAP users and nonusers.
Anticancer agent prescription fills
For both PAP users and nonusers, the top-20 anticancer oral agent fills represented 93% (n = 2,892 of 3,105) of all anticancer oral fills (Table 6), with 16% (n = 454) of these oral fills being provided through PAPs. Among PAP users, anticancer agents from PAPs accounted for 40% of their total charges and 35% of the total number of agents. Temozolomide (Temodar; mean charge/fill = $3,346) represented the highest amount of total charges ($220,857) from PAPs, whereas imatinib (Gleevec; mean charge/fill = $5,372) and dasatinib (Sprycel; mean charge/fill = $5,221) accounted for the highest average charges per fill. Anastrozole (Arimidex; n = 178 fills; 10/month), capecitabine (Xeloda; n = 91 fills; 5/month), and temozolomide (n = 66 fills; 4/month) accounted for 70% of agents from PAPs. PAP users who were given bicalutamide received 100% of those agents from PAPs. Five of the seven oral anticancer agents with no fills from PAPs had initial US Food and Drug Administration approval years before 2000.
Discussion
At MDACC, PAPs are designed to help cancer patients overcome financial barriers to accessing oral supportive and anticancer agents. Over an 18-month observation period, less than 5% of the cancer patients at MDACC who received prescription medications from the outpatient pharmacy were enrolled in a PAP— and these PAPs provided 13% of their medication fills, representing an annualized $3.6 million in pharmaceutical expenditures. In interpreting our findings, several factors should be considered.
Oral anticancer agents accounted for 4% of all prescription medication fills during the study period. Comparatively, an analysis of the 2007 National Ambulatory Medical Care Survey showed that less than 1% of cancer patients were prescribed at least one oral anticancer agent.10 This finding indicates that both nationally and at MDACC, chemotherapy continues to be largely provided parenterally, as there is more of a financial benefit from intravenous therapies that are often reimbursed by insurers as well as PAPs.
In the outpatient pharmacy at MDACC, PAPs provided nearly onethird of oral anticancer fills for PAP users—totaling a mean of $500,000 per month in expenditures. However, three agents, anastrozole (for breast cancer), capecitabine (for breast and GI cancers, primarily), and temozolomide (for brain tumors) accounted for 75% of all of the anticancer agents provided by PAPs. We also found that pharmaceutical companies provided expensive newer, targeted, anticancer agents (primarily dasatinib and imatinib, the two agents with the greatest pharmaceutical per-person expenditures by the PAP program) through PAPs.
Although PAPs filled a strong and focused need for a small number of oral chemotherapy agents for some individuals with breast, GI, and brain cancers, they did not provide much benefit for a wide range of supportive care agents, particularly those that are schedule C and are used to treatcancer pain. Pain is the most prevalent symptom reported by cancer patients, 11 but there were few schedule C pain medications among the most common medications provided through PAPs. These substances are generally not provided by PAPs because of legal and substance abuse concerns.12 However, these medications were commonly prescribed to PAP users and PAP nonusers alike, outside of the PAP program. It would be important to evaluate the comparative success in treating pain among cancer patients at MDACC who receive a limited array of pain medications from PAPs (usually agents that are not substance-controlled by the Drug Enforcement Administration) versus treatment of pain experienced by patients whose medications are not reimbursed by PAPs.
We found that being younger than 65 years old, being indigent, and having no health insurance were the strongest predictors of using a PAP. This finding was expected, given that US adults younger than age 65 are ineligible for outpatient prescription medication coverage through Medicare Part D. However, contrary to expectations, about 45% of PAP users had either private or governmentsupplied health insurance. Because it is not uncommon for cancer patients to endure economic hardship (including bankruptcy) when trying to finance their care,13 healthcare professionals could recommend PAPs and other relevant assistance programs to all of their cancer patients.
With the expansion of health insurance through the Patient Protection and Affordable Health Care Act of 2010, it is hoped that the need for cancer patients to enroll in PAPs will be diminished; yet, given the reality of the high cost of anticancer agents, reimbursement policies for these agents, and tiered formularies among insurers leading to high outof- pocket costs for patients, the need for PAPs is likely to remain. PAPs can be a viable option for some patients, but healthcare professionals should be aware that there are a number of concerns about these programs, including their complex and burdensome application process and often limited variety of available drugs.14
This study is not without its limitations. First, we may have underestimated our sample of PAP patients due to the fact that MDACC did not electronically or systematically track the use of PAPs within its pharmacy database at the time of the study. The institution is in the process of developing such a system.
Second, the data used in this study cannot be assumed to reflect a “closed pharmacy” setting because some patients, particularly those who have health insurance with prescription medication coverage, may have received some of their medications from outside pharmacies.
Third, because insurance status is not necessarily a static characteristic, insurance status in this study was classified based on that at the time of registration with MDACC’s financial department, and no account was taken of changes that might have occurred.
Last, our results are not necessarily generalizable to all cancer populations, time periods, or settings. Cancer patients treated in academic centers such as MDACC may differ from those who are treated in community settings. In particular, fewer than 10% of patients at MDACC qualified for indigent financial assistance in 2007,15 which is likely to have impacted the number of patients who were potentially eligible for PAPs. It is also likely that had our study been conducted prior to the implementation of Medicare Part D, our sample of PAP patients would have been older. Nevertheless, our results may be generalizable to cancer patients receiving care in other academic cancer centers.
Conclusion This study builds upon a previous description of implementing PAPs in a comprehensive cancer center16 as well as contributes to our limited knowledge of the use of PAPs among cancer patients.6 Future studies should prospectively examine cancer patients’ experiences and satisfaction with PAPs from the process of applying to the point of receiving requested therapies and evaluate the effect of PAPs on cancer outcomes in various care settings. Multidisciplinary teams, including pharmacists and clinicians, should establish and recommend valid and relevant clinical endpoints for researchers to use in effectiveness studies of PAPs and cancer patients, particularly as they relate to oral anticancer agent use. Given that these oral agents represent more than 25% of cancer therapies in development,17 future studies of PAPs are ideal for evaluating concerns of accessibility, affordability, and compliance related to these agents.
MDACC is a unique resource for observers of PAPs, as it is the largest cancer center in the United States. However, few cancer patients at MDACC were eligible for and accessed PAPs in the outpatient pharmacy. Although smaller cancer centers may not be able to devote the same degree of financial and personnel resources to their patients as does MDACC, these centers could seek to build relationships with specific pharmaceutical companies that provide PAPs for the oral anticancer and supportive care therapies most commonly prescribed and administered at their centers. Scarce resources could also be utilized in other ways, such as by developing public-private risk pools for establishment of indigent care funds.
Acknowledgments: The authors thank Chun Feng, Jason Lau, and Oliver Max for their special assistance; Dr. Phoenix Do for her study design recommendations; and Karyn Popham for her editorial support. They especially thank Rebecca Arbuckle, RPh, for her support of this project. At the time of the study, Dr. Felder was supported by a Predoctoral Fellowship from The University of Texas School of Public Health Cancer Education and Career Development Program, funded by National Cancer Institute/NIH Grant R25-CA-57712-17.
References
1. Hede K. Increase in oral cancer drugs raises thorny issues for oncology practices. J Natl Cancer Inst 2009;101:1534–1536.
2. Chisholm MA, DiPiro JT. Pharmaceutical manufacturer assistance programs. Arch Intern Med 2002;162:780–784.
3. Duke KS, Raube K, Lipton HL. Patientassistance programs: assessment of and use by safety-net clinics. Am J Health Syst Pharm 2005;62:726–731.
4. Pisu M, Richman J, Allison JJ, Williams OD, Kiefe CI. Pharmaceuticals companies’ medication assistance programs: potentially useful but too burdensome to use? South Med J 2009;102:139–144.
5. Felder TM, Palmer NR, Lal LS, Mullen PD. What is the evidence for pharmaceutical patient assistance programs? a systematic review. J Health Care Poor Underserved 2011;22:24–49.
6. Meropol NJ, Schrag D, Smith TJ, et al. American Society of Clinical Oncology guidance statement: the cost of cancer care. J Clin Oncol 2009;27:3868–3874.
7. United States Department of Labor, Bureau of Labor Statistics: Producer Price Index Industry Data—Pharmaceutical Preparation & Manufacturing; 2010 [updated April 26, 2010]. http://www.bls.gov/ppi/data.htm. Accessed June 22, 2011.
8. Hosmer DW, Lemeshow S. Applied Logistic
Logistic
Regression. New York: Wiley; 2000. 9. StataCorp LP. STATA statistical software. 2009; Release 11.
10. Arora S. Use of oral chemotherapeutic medications in non-traditional ambulatory settings; 2009. http://digarchive.library.vcu. edu/dspace/bitstream/10156/2711/1/Thesis_ MPH_sameer.pdf. Accessed June 22, 2011.
11. Cherny NI. The management of cancer pain. CA Cancer J Clin 2000;50:70–116.
12. Williams K. Accessing patient assistance programs to meet clients’ medication needs. J Am Acad Nurse Pract 2000;12:233– 235.
13. National Survey of Households Affected by Cancer: Kaiser Family Foundation; 2006 [updated November 2006]. http://kff.org/kaiserpolls/ upload/7591.pdf. Accessed June 22, 2011.
14. Choudhry NK, Lee JL, Agnew-Blais J, Corcoran C, Shrank WH. Drug company- sponsored patient assistance programs: a viable safety net? Health Aff (Millwood) 2009;28:827–834.
15. Ackerman T. M.D. Anderson submits its records on charitable care: cancer center hopes to quell Iowa senator’s investigation. Houston Chronicle. October 9, 2008. http:// www.chron.com/disp/story.mpl/metropolitan/ 6050254.html. Accessed June 22, 2011.
16. Johnson PE. Patient assistance programs and patient advocacy foundations: alternatives for obtaining prescription medications when insurance fails. Am J Health Syst Pharm 2006;63(21 suppl 7):S13–S17.
17. Weingart SA, Brown E, Bach PB, et al. National Comprehensive Cancer Network task force report: oral chemotherapy. JNCCN 2008;6(suppl 3):S1–S25.
Medicare ACO cost incentives for potential prescribing shifts in cancer therapies worry stakeholders
The Centers for Medicare & Medicaid Services (CMS) should look for ways to ensure that financial incentives for Medicare accountable care organizations (ACOs) do not promote inappropriate shifts in prescribing from the Part B benefit to Part D, urged clinician and biotechnology groups in comments filed with the agency on June 6, 2011.
The Part B benefit covers drugs administered in a physician’s office, such as infused products. It excludes most oral and other medications that patients obtain at retail or mail-order pharmacies for use at home; these medications instead are covered under Part D. So in theory, the ACO program will encompass Part B expenditures, but not Part D. ACOs could save on the cost of drugs by shifting some prescriptions to Part D.
The comments respond to the proposed rule to establish ACOs that the CMS published on March 31, 2011. ACOs are part of a “shared savings” initiative, mandated by the Affordable Care Act, aimed at driving providers through financial incentives to better manage and coordinate healthcare delivery to improve quality and reduce costs. ACOs are slated to begin operating January 1, 2012. Under the proposed regulation, ACOs would be made up of physician practices, hospitals, and other healthcare providers. Primary care physicians are expected to serve as coordinators of individual patient care. The proposal posal does not specifically discuss potential shifts in prescribing. However, the Biotechnology Industry Organization (BIO) and others have urged the CMS to develop a process to ensure that patient care is not secondary to the pursuit of savings.
The implications of prescribing shifts
The BIO points out that the Congressional Research Service (CRS) identified prescribing shifts as a possible issue in an April 25 report on ACOs and the Medicare shared savings program. Such shifts might occur because the savings calculations used to evaluate ACOs are based on Parts A and B expenditures only, suggests the CRS, an analytic unit employed by Congress. Thus, “there may be instances where there is the appearance of cost savings as a result of providers unduly relying on Part D prescription medicines over other forms of care,” the CRS cautioned. The BIO added, “this issue may present problems from both the perspective of quality patient care as well as for Medicare program expenditures.” A shift to Part D drugs could actually increase patient outof- pocket costs, “which in turn may impact prescription drug adherence and ultimately clinical outcomes,” the comments stated.
The prospect of increasing patient costs under Part D is supported by a CMS-commissioned report released in August 2010. The report looked at the potential impact of consolidating Medicare reimbursement for drug categories with overlapping Parts B and D coverage. It found, on average, that patients would face higher costsharing if categories are consolidated under Part D, but that the Medicare program would save money, because Part D has less generous coverage rules. The study looked at oral anticancer and antiemetic drugs, insulin, vaccinations, inhalants, and rheumatoid arthritis medications.
In separate comments, the American Society of Clinical Oncology (ASCO) also targeted potential prescribing shifts in cancer therapies. “The CMS should address perverse incentives that will arise due to the proposed rule’s exclusion of Part D drugs from the costs by which ACO savings are measured,” ASCO warned. “Safeguards are needed to ensure that cancer patients receive the most appropriate therapy, regardless of whether that therapy is typically covered under Part B or under Part D.” (In addition to chemotherapy administered in a physician’s office, some oral anticancer drugs are also covered under Part B when they are direct substitutions for physician-administered treatments.)
Even the medical device and diagnostics industry registered concern with possible prescribing shifts to Part D drugs. The Advanced Medical Technology Association pointed out that providers might have an artificial incentive to substitute a pain medication, for example, when a surgical procedure covered under Part A or B would be considered standard of care.
Carving out the cost of new technologies
A number of stakeholders flagged the prospect that cost issues might also deter the adoption of cuttingedge drugs or other medical technologies by ACOs. The BIO and the Medical Device Manufacturers Association are among the groups urging CMS to consider ways to carve out new technologies from the benchmark for ACO costs and from annual expenditure performance reviews.
As proposed, the ACO program would be risk-based: an organization would receive a bonus payment if cost and quality benchmarks are met but would also face penalties if they are not. As a result, providers may be reluctant to adopt expensive new technologies, based on the comments.
To ensure beneficiary access to new therapies and to retain incentives for innovation, the stakeholders have suggested that the CMS adapt the Medicare payment mechanisms already used for new technologies in the hospital setting. Medicare’s hospital payment scheme includes mechanisms for add-on (inpatient) or pass-through (outpatient) payments. They provide payment for new and relatively expensive treatments in their first few years on the market in addition to the reimbursement rate set for their therapeutic category. The system is designed to account for the fact that the CMS does not yet know with certainty how much the new treatment will add to hospital charges and costs. And the extra reimbursement is intended to make it more likely that hospitals can afford to “try out” new technologies while the CMS collects the data on cost and use patterns that will ultimately allow it to develop a more permanent payment level.
The BIO has suggested that the carveout should apply regardless of the healthcare setting and should include drugs and biologics provided in the physician’s office. “Such congruity is necessary to ensure that the policy does not create an incentive to perform procedures in the hospital rather than the physician’s office,” according to the comments.
Quality metrics and warfarin
In comments on other aspects of the proposal, the BIO has recommended that the CMS should establish a process to rapidly update the quality measures used to evaluate ACO performance. The proposal outlines a number of quality measures, including whether patients at risk of serious disease or frail elderly patients are taking drugs for managing chronic disease. The agency also will track patient measurements, such as cholesterol levels and hemoglobin A1c values, to monitor the effectiveness of disease management. The quality measures are expected to improve adherence to drugs for chronic disease.
Nevertheless, the BIO takes issue with the quality metric assessing adherence to warfarin therapy in patients with heart failure and paroxysmal or chronic atrial fibrillation; the concern is that it could discourage the use of newer agents. “This [warfarin] measure already is outdated,” the BIO stated. “There are more advanced therapeutic options available for these patients today as well as additional therapies in development. Leaving the measure in place may force ACOs to use a therapy that is no longer the only clinically appropriate, or even recommended, choice.”
The latest newcomer to the category is dabigatran (Pradaxa). Two other oral anticoagulants are in late-stage development: rivaroxaban (Xarelto) and apixaban. The biotech group’s concerns reflect the tension in the anticoagulant market segment among developers of newer agents and payer concerns over whether the advantages offered by such drugs can justify their increased cost.
A more prominent role for specialists
The ACO proposal is designed to emphasize the role of primary care providers. However, the BIO has advised that the CMS also ensure that specialists play an important part in the operation of ACOs, such as during the development of clinical guidelines and processes. Under the proposed rule, ACOs would be required to “develop and implement evidence-based medical practice or clinical guidelines and processes for delivering care consistent with the goals of better care for individuals, better health for populations, and lower growth in expenditures.”
ASCO echoed the BIO recommendation. “Given the prevalence of cancer in the Medicare community … [the] CMS should require ACOs to secure substantial input from specialists practicing in the community.” ASCO also requested that the CMS ensure that oncology practices do not face undue financial or procedural challenges under the ACO program.
“Oncologists already face a significant form of financial risk arising from the need to purchase and administer chemotherapy drugs within an environment in which Medicare and other health insurers often provide ambiguous coverage policies, payment rates that are inadequate to cover acquisition costs and slow preauthorization decisions,” ASCO stated. Furthermore, ASCO noted that Medicare contractors frequently require “excessive documentation” on the effectiveness of off-label drug indications, even when the indication is supported in the peer-reviewed literature.
The Centers for Medicare & Medicaid Services (CMS) should look for ways to ensure that financial incentives for Medicare accountable care organizations (ACOs) do not promote inappropriate shifts in prescribing from the Part B benefit to Part D, urged clinician and biotechnology groups in comments filed with the agency on June 6, 2011.
The Part B benefit covers drugs administered in a physician’s office, such as infused products. It excludes most oral and other medications that patients obtain at retail or mail-order pharmacies for use at home; these medications instead are covered under Part D. So in theory, the ACO program will encompass Part B expenditures, but not Part D. ACOs could save on the cost of drugs by shifting some prescriptions to Part D.
The comments respond to the proposed rule to establish ACOs that the CMS published on March 31, 2011. ACOs are part of a “shared savings” initiative, mandated by the Affordable Care Act, aimed at driving providers through financial incentives to better manage and coordinate healthcare delivery to improve quality and reduce costs. ACOs are slated to begin operating January 1, 2012. Under the proposed regulation, ACOs would be made up of physician practices, hospitals, and other healthcare providers. Primary care physicians are expected to serve as coordinators of individual patient care. The proposal posal does not specifically discuss potential shifts in prescribing. However, the Biotechnology Industry Organization (BIO) and others have urged the CMS to develop a process to ensure that patient care is not secondary to the pursuit of savings.
The implications of prescribing shifts
The BIO points out that the Congressional Research Service (CRS) identified prescribing shifts as a possible issue in an April 25 report on ACOs and the Medicare shared savings program. Such shifts might occur because the savings calculations used to evaluate ACOs are based on Parts A and B expenditures only, suggests the CRS, an analytic unit employed by Congress. Thus, “there may be instances where there is the appearance of cost savings as a result of providers unduly relying on Part D prescription medicines over other forms of care,” the CRS cautioned. The BIO added, “this issue may present problems from both the perspective of quality patient care as well as for Medicare program expenditures.” A shift to Part D drugs could actually increase patient outof- pocket costs, “which in turn may impact prescription drug adherence and ultimately clinical outcomes,” the comments stated.
The prospect of increasing patient costs under Part D is supported by a CMS-commissioned report released in August 2010. The report looked at the potential impact of consolidating Medicare reimbursement for drug categories with overlapping Parts B and D coverage. It found, on average, that patients would face higher costsharing if categories are consolidated under Part D, but that the Medicare program would save money, because Part D has less generous coverage rules. The study looked at oral anticancer and antiemetic drugs, insulin, vaccinations, inhalants, and rheumatoid arthritis medications.
In separate comments, the American Society of Clinical Oncology (ASCO) also targeted potential prescribing shifts in cancer therapies. “The CMS should address perverse incentives that will arise due to the proposed rule’s exclusion of Part D drugs from the costs by which ACO savings are measured,” ASCO warned. “Safeguards are needed to ensure that cancer patients receive the most appropriate therapy, regardless of whether that therapy is typically covered under Part B or under Part D.” (In addition to chemotherapy administered in a physician’s office, some oral anticancer drugs are also covered under Part B when they are direct substitutions for physician-administered treatments.)
Even the medical device and diagnostics industry registered concern with possible prescribing shifts to Part D drugs. The Advanced Medical Technology Association pointed out that providers might have an artificial incentive to substitute a pain medication, for example, when a surgical procedure covered under Part A or B would be considered standard of care.
Carving out the cost of new technologies
A number of stakeholders flagged the prospect that cost issues might also deter the adoption of cuttingedge drugs or other medical technologies by ACOs. The BIO and the Medical Device Manufacturers Association are among the groups urging CMS to consider ways to carve out new technologies from the benchmark for ACO costs and from annual expenditure performance reviews.
As proposed, the ACO program would be risk-based: an organization would receive a bonus payment if cost and quality benchmarks are met but would also face penalties if they are not. As a result, providers may be reluctant to adopt expensive new technologies, based on the comments.
To ensure beneficiary access to new therapies and to retain incentives for innovation, the stakeholders have suggested that the CMS adapt the Medicare payment mechanisms already used for new technologies in the hospital setting. Medicare’s hospital payment scheme includes mechanisms for add-on (inpatient) or pass-through (outpatient) payments. They provide payment for new and relatively expensive treatments in their first few years on the market in addition to the reimbursement rate set for their therapeutic category. The system is designed to account for the fact that the CMS does not yet know with certainty how much the new treatment will add to hospital charges and costs. And the extra reimbursement is intended to make it more likely that hospitals can afford to “try out” new technologies while the CMS collects the data on cost and use patterns that will ultimately allow it to develop a more permanent payment level.
The BIO has suggested that the carveout should apply regardless of the healthcare setting and should include drugs and biologics provided in the physician’s office. “Such congruity is necessary to ensure that the policy does not create an incentive to perform procedures in the hospital rather than the physician’s office,” according to the comments.
Quality metrics and warfarin
In comments on other aspects of the proposal, the BIO has recommended that the CMS should establish a process to rapidly update the quality measures used to evaluate ACO performance. The proposal outlines a number of quality measures, including whether patients at risk of serious disease or frail elderly patients are taking drugs for managing chronic disease. The agency also will track patient measurements, such as cholesterol levels and hemoglobin A1c values, to monitor the effectiveness of disease management. The quality measures are expected to improve adherence to drugs for chronic disease.
Nevertheless, the BIO takes issue with the quality metric assessing adherence to warfarin therapy in patients with heart failure and paroxysmal or chronic atrial fibrillation; the concern is that it could discourage the use of newer agents. “This [warfarin] measure already is outdated,” the BIO stated. “There are more advanced therapeutic options available for these patients today as well as additional therapies in development. Leaving the measure in place may force ACOs to use a therapy that is no longer the only clinically appropriate, or even recommended, choice.”
The latest newcomer to the category is dabigatran (Pradaxa). Two other oral anticoagulants are in late-stage development: rivaroxaban (Xarelto) and apixaban. The biotech group’s concerns reflect the tension in the anticoagulant market segment among developers of newer agents and payer concerns over whether the advantages offered by such drugs can justify their increased cost.
A more prominent role for specialists
The ACO proposal is designed to emphasize the role of primary care providers. However, the BIO has advised that the CMS also ensure that specialists play an important part in the operation of ACOs, such as during the development of clinical guidelines and processes. Under the proposed rule, ACOs would be required to “develop and implement evidence-based medical practice or clinical guidelines and processes for delivering care consistent with the goals of better care for individuals, better health for populations, and lower growth in expenditures.”
ASCO echoed the BIO recommendation. “Given the prevalence of cancer in the Medicare community … [the] CMS should require ACOs to secure substantial input from specialists practicing in the community.” ASCO also requested that the CMS ensure that oncology practices do not face undue financial or procedural challenges under the ACO program.
“Oncologists already face a significant form of financial risk arising from the need to purchase and administer chemotherapy drugs within an environment in which Medicare and other health insurers often provide ambiguous coverage policies, payment rates that are inadequate to cover acquisition costs and slow preauthorization decisions,” ASCO stated. Furthermore, ASCO noted that Medicare contractors frequently require “excessive documentation” on the effectiveness of off-label drug indications, even when the indication is supported in the peer-reviewed literature.
The Centers for Medicare & Medicaid Services (CMS) should look for ways to ensure that financial incentives for Medicare accountable care organizations (ACOs) do not promote inappropriate shifts in prescribing from the Part B benefit to Part D, urged clinician and biotechnology groups in comments filed with the agency on June 6, 2011.
The Part B benefit covers drugs administered in a physician’s office, such as infused products. It excludes most oral and other medications that patients obtain at retail or mail-order pharmacies for use at home; these medications instead are covered under Part D. So in theory, the ACO program will encompass Part B expenditures, but not Part D. ACOs could save on the cost of drugs by shifting some prescriptions to Part D.
The comments respond to the proposed rule to establish ACOs that the CMS published on March 31, 2011. ACOs are part of a “shared savings” initiative, mandated by the Affordable Care Act, aimed at driving providers through financial incentives to better manage and coordinate healthcare delivery to improve quality and reduce costs. ACOs are slated to begin operating January 1, 2012. Under the proposed regulation, ACOs would be made up of physician practices, hospitals, and other healthcare providers. Primary care physicians are expected to serve as coordinators of individual patient care. The proposal posal does not specifically discuss potential shifts in prescribing. However, the Biotechnology Industry Organization (BIO) and others have urged the CMS to develop a process to ensure that patient care is not secondary to the pursuit of savings.
The implications of prescribing shifts
The BIO points out that the Congressional Research Service (CRS) identified prescribing shifts as a possible issue in an April 25 report on ACOs and the Medicare shared savings program. Such shifts might occur because the savings calculations used to evaluate ACOs are based on Parts A and B expenditures only, suggests the CRS, an analytic unit employed by Congress. Thus, “there may be instances where there is the appearance of cost savings as a result of providers unduly relying on Part D prescription medicines over other forms of care,” the CRS cautioned. The BIO added, “this issue may present problems from both the perspective of quality patient care as well as for Medicare program expenditures.” A shift to Part D drugs could actually increase patient outof- pocket costs, “which in turn may impact prescription drug adherence and ultimately clinical outcomes,” the comments stated.
The prospect of increasing patient costs under Part D is supported by a CMS-commissioned report released in August 2010. The report looked at the potential impact of consolidating Medicare reimbursement for drug categories with overlapping Parts B and D coverage. It found, on average, that patients would face higher costsharing if categories are consolidated under Part D, but that the Medicare program would save money, because Part D has less generous coverage rules. The study looked at oral anticancer and antiemetic drugs, insulin, vaccinations, inhalants, and rheumatoid arthritis medications.
In separate comments, the American Society of Clinical Oncology (ASCO) also targeted potential prescribing shifts in cancer therapies. “The CMS should address perverse incentives that will arise due to the proposed rule’s exclusion of Part D drugs from the costs by which ACO savings are measured,” ASCO warned. “Safeguards are needed to ensure that cancer patients receive the most appropriate therapy, regardless of whether that therapy is typically covered under Part B or under Part D.” (In addition to chemotherapy administered in a physician’s office, some oral anticancer drugs are also covered under Part B when they are direct substitutions for physician-administered treatments.)
Even the medical device and diagnostics industry registered concern with possible prescribing shifts to Part D drugs. The Advanced Medical Technology Association pointed out that providers might have an artificial incentive to substitute a pain medication, for example, when a surgical procedure covered under Part A or B would be considered standard of care.
Carving out the cost of new technologies
A number of stakeholders flagged the prospect that cost issues might also deter the adoption of cuttingedge drugs or other medical technologies by ACOs. The BIO and the Medical Device Manufacturers Association are among the groups urging CMS to consider ways to carve out new technologies from the benchmark for ACO costs and from annual expenditure performance reviews.
As proposed, the ACO program would be risk-based: an organization would receive a bonus payment if cost and quality benchmarks are met but would also face penalties if they are not. As a result, providers may be reluctant to adopt expensive new technologies, based on the comments.
To ensure beneficiary access to new therapies and to retain incentives for innovation, the stakeholders have suggested that the CMS adapt the Medicare payment mechanisms already used for new technologies in the hospital setting. Medicare’s hospital payment scheme includes mechanisms for add-on (inpatient) or pass-through (outpatient) payments. They provide payment for new and relatively expensive treatments in their first few years on the market in addition to the reimbursement rate set for their therapeutic category. The system is designed to account for the fact that the CMS does not yet know with certainty how much the new treatment will add to hospital charges and costs. And the extra reimbursement is intended to make it more likely that hospitals can afford to “try out” new technologies while the CMS collects the data on cost and use patterns that will ultimately allow it to develop a more permanent payment level.
The BIO has suggested that the carveout should apply regardless of the healthcare setting and should include drugs and biologics provided in the physician’s office. “Such congruity is necessary to ensure that the policy does not create an incentive to perform procedures in the hospital rather than the physician’s office,” according to the comments.
Quality metrics and warfarin
In comments on other aspects of the proposal, the BIO has recommended that the CMS should establish a process to rapidly update the quality measures used to evaluate ACO performance. The proposal outlines a number of quality measures, including whether patients at risk of serious disease or frail elderly patients are taking drugs for managing chronic disease. The agency also will track patient measurements, such as cholesterol levels and hemoglobin A1c values, to monitor the effectiveness of disease management. The quality measures are expected to improve adherence to drugs for chronic disease.
Nevertheless, the BIO takes issue with the quality metric assessing adherence to warfarin therapy in patients with heart failure and paroxysmal or chronic atrial fibrillation; the concern is that it could discourage the use of newer agents. “This [warfarin] measure already is outdated,” the BIO stated. “There are more advanced therapeutic options available for these patients today as well as additional therapies in development. Leaving the measure in place may force ACOs to use a therapy that is no longer the only clinically appropriate, or even recommended, choice.”
The latest newcomer to the category is dabigatran (Pradaxa). Two other oral anticoagulants are in late-stage development: rivaroxaban (Xarelto) and apixaban. The biotech group’s concerns reflect the tension in the anticoagulant market segment among developers of newer agents and payer concerns over whether the advantages offered by such drugs can justify their increased cost.
A more prominent role for specialists
The ACO proposal is designed to emphasize the role of primary care providers. However, the BIO has advised that the CMS also ensure that specialists play an important part in the operation of ACOs, such as during the development of clinical guidelines and processes. Under the proposed rule, ACOs would be required to “develop and implement evidence-based medical practice or clinical guidelines and processes for delivering care consistent with the goals of better care for individuals, better health for populations, and lower growth in expenditures.”
ASCO echoed the BIO recommendation. “Given the prevalence of cancer in the Medicare community … [the] CMS should require ACOs to secure substantial input from specialists practicing in the community.” ASCO also requested that the CMS ensure that oncology practices do not face undue financial or procedural challenges under the ACO program.
“Oncologists already face a significant form of financial risk arising from the need to purchase and administer chemotherapy drugs within an environment in which Medicare and other health insurers often provide ambiguous coverage policies, payment rates that are inadequate to cover acquisition costs and slow preauthorization decisions,” ASCO stated. Furthermore, ASCO noted that Medicare contractors frequently require “excessive documentation” on the effectiveness of off-label drug indications, even when the indication is supported in the peer-reviewed literature.
A case of lung cancer and hypercoagulability, complicated by suspected heparin-induced thrombocytopenia
Heparin-induced thrombocytopenia (HIT) is a life-threatening disorder that follows exposure to unfractionated heparin or (less commonly) low-molecular-weight heparin (LMWH). Patients classically present with a low platelet count (< 150,000 cells/mm3) or a relative decrease of 50% or more from baseline, although the fall may be less (eg, 30%–40%) in some patients.Thrombotic complications develop in approximately 20%–50% of patients.
HIT is caused by antibodies against complexes of platelet factor 4 and heparin.These antibodies are present in nearly all patients who receive a clinical diagnosis of the disorder and are also known to cause disease in animals.However,they are also present in many patients who have been exposed to heparin in various clinical settings but who do not develop clinical manifestations. It is uncertain why complications occur in some patients but not in others.1 We present a 73-year-old man who developed thrombocytopenia after starting LMWH and who has newly diagnosed adenocarcinoma of the lungs with extensive arterial and venous thrombosis and a negative serology for HIT.
Case presentation
A 73-year-old man presented to the emergency department after waking up in the morning with right-sided vague weakness and an inability to get out of bed. He had a history of right parietal stroke 1 month before the current presentation, when he was diagnosed with an aortic arch atheroma and started on warfarin. (At that time, CT scan of the head showed a right posterior temporoparietal lobe infarct in the posterior right middle cerebral artery distribution, and MRI of the brain and magnetic resonance angiography showed acute or subacute infarction in the distribution of the posterior division of the right middle cerebral artery, likely embolic, and tiny acute infarctions in the left frontal lobe.) This patient had been admitted 5 days prior to the current presentation for right lower extremity deep vein thrombosis (DVT) and was discharged after being prescribed enoxaparin (60 mg subcutaneously every 12 hours) and warfarin as per international normalized ratio (INR) daily.
Also included in the medical history was supraventricular tachycardia status post ablation, non–ST elevation myocardial infarction (NSTEMI), hypertension, hyperlipidemia, and macular degeneration. He had no surgical history. The patient had a family history of coronary artery disease. He had an extensive smoking history up until the day of admission. His medications on admission included atorvastatin (Lipitor; 20 mg daily), warfarin daily as per INR, enoxaparin (60 mg subcutaneously every 12 hours), amlodipine (5 mg daily), and aspirin (81 mg daily).
Pertinent initial laboratory results on admission were as follows: hemoglobin, 12.9 g/dL; white blood cell count, 8.6 ×109/L; platelet count, 183,000 cells/ mm3; INR, 1.2; and initial troponin level, negative.His admission chest x-ray showed a 4.5 cm × 5.5 cm lobulated density in the right hilum, suspicious for a hilar or subcarinal mass. Initial peripheral blood smear showed an isolated platelet decrease with increased size and no schistocytes. Initial CT of the head on admission showed no evidence of acute transcortical infarction and no definite evidence of acute intracranial hemorrhage but did show interval evolution of the right middle cerebral artery and left watershed distribution infarctions, with a probable small region of laminar necrosis in the right parietal lobe.
Clinical Course
The patient was initially thought to have had a transient ischemic attack causing aphasia, confusion, and rightsided weakness. He was started on therapeutic anticoagulation with dalteparin (Fragmin; 12,000 U subcutaneously daily),and enoxaparin was discontinued. The following day,his platelet count was 86,000 cells/mm3,down from an admission platelet count of 183,000 cells/mm3. A subsequent MRI of the brain showed a new hemorrhagic area in the right parietal infarct. The decision was made to stop anticoagulation, even though he had an embolic source from his aortic arch atheroma and lower extremity DVT.
The patient then underwent inferior vena cava (IVC) filter placement to prevent pulmonary thromboembolism and was transferred to the medical service due to low platelet count and an episode of nine beats of ventricular tachycardia. Subsequently, his troponin level was found to be elevated > 12 ng/mL, without significant electrocardiographic changes. He was diagnosed as having NSTEMI. Given his conversion from an ischemic to hemorrhagic CNS infarct and decrease in platelet count after LMWH exposure, HIT became a concern, and both anticoagulation and antiplatelet agents were held. The patient’s platelet count continued to trend downward over the next 3 days to a low of 27,000 cells/mm3. An HIT panel was negative by both immunologic and functional assays.
A CT scan of the brain 3 days after admission to monitor the hemorrhagic infarct showed multiple evolving infarcts and a new left occipital hemorrhagic infarct. The following day, a repeat CT scan of the head showed mulatiple evolving infarcts of varying ages, some with hemorrhage, and a mild interval increase in the previously described left medial parietal and left occipital lobe infarcts.
With worsening of his hemorrhagic infarct, along with his low platelet count and negative HIT panel,the decision was made to transfuse 2 units of platelets. His platelet count increased to 64,000 cells/mm3 after transfusion, subsequently dropping to 44,000 cells/mm3. However, during this time, the patient began to have worsening right lower extremity pain and left upper quadrant abdominal pain.
A CT scan of the thorax showed multifocal right hilar adenopathy suspicious for malignancy, either metastatic or representing a central lung carcinoma. It also showed nonocclusive segmental and possibly subsegmental pulmonary emboli in the right lower and middle lobes, as well as hypodense areas in the spleen, suggestive of areas of splenic infarction. Echocardiography showed an ejection fraction of 60%–70%, diastolic dysfunction, mildly elevated pulmonary artery pressure, and no evidence of patent foramen ovale.A cardiac stress test showed no reversible defects and an ejection fraction of 63%.
Risk of further bleeding into the brain was thought to be too great to initiate anticoagulation despite the CT thorax findings. The neurologist recommended waiting 2 weeks post hemorrhagic infarction before beginning anticoagulation. Antiphospholipid antibody syndrome was ruled out, with a negative lupus anticoagulant and anticardiolipin antibody. Also, negative blood cultures, normal fibrinogen levels, and normal haptoglobin levels ruled out disseminated intravascular coagulation. D-dimer was elevated but nonspecific, secondary to malignancy and multiple infarcts. He was started on aspirin (81 mg daily) 9 days after admission.
The patient had a repeat CT scan of the thorax and CT scan of the abdomen and pelvis due to continued abdominal pain. The CT scans showed multiple subsegmental pulmonary emboli, greatest in the right lower lobe, some of which were new since the prior study; continued evidence of multifocal splenic infarction (Figure 3); and multiple right and left kidney infarcts.
The patient then underwent endobronchial ultrasound (EBUS)-guided biopsy of his right hilar adenopathy to confirm the diagnosis of suspected malignancy. After the procedure, he developed right upper quadrant pleuritic pain with a low-grade fever. A repeat CT scan of the thorax showed a marked increase in the extent of the right lower lobe pulmonary emboli, with a new small embolus noted in the anterior segment of the right upper lobe. There was a thrombus inferior to the IVC filter, with probable mild extension of a thrombus superior to the filter as well, and again multiple splenic and bilateral renal infarcts.
With progression of thrombosis and now post EBUS, anticoagulation was initiated with argatroban and warfarin. His D-dimer was followed daily and remained high, despite therapeutic anticoagulation with warfarin. Given the persistently elevated D-dimer, the hematologist recommended discontinuing warfarin and starting fondaparinux (Arixtra) subcutaneously. His platelet count improved to a range of 156,000 cells/mm3 to 181,000 cells/mm3, even before the initiation of chemotherapy.
Follow-up
HIT was suspected clinically by classic drop in platelet count but was negative on enzyme-linked immunosorbent assay (ELISA) and serotonin release assay (SRA). The patient has been maintained on fondaparinux for anticoagulation, avoiding heparin. Factor V Leiden and lupus anticoagulant were negative.
Fondaparinux was discontinued after 3 months, and the patient presented again with swelling of his right lower extremity. Ultrasonography of the right lower extremity redemonstrated an occlusive thrombus in the peripheral portion of the right femoaral vein and throughout the right peroneal vein. The patient was restarted on fondaparinux (7.5 mg subcutaneously daily). During this follow-up, his platelet count ranged from 134,000 cells/mm3 to 193,000 cells/mm3.
Regarding management of nonsmall cell lung carcinoma of the left upper lobe (stains positive for TTF- 1 [thyroid transcription factor-1], CK7, and CK20; weakly positive for CK5/6; and negative for P63) with metastasis to bone and adrenal glands, he received 4 cycles of paclitaxel/carboplatin, with improved disease. A repeat CT of the chest, abdomen, and pelvis after chemotherapy showed improvement in mediastinal and hilar lymphadenopathy, resolution of extensive right lower lobe pulmonary consolidation, resolution of right-sided effusion, and no evidence of metastatic malignancy in the abdomen or pelvis and no osseous metastasis.
He was started on maintenance therapy with pemetrexed (Alimta), which was continued for 4 months, until repeat CT revealed progressive disease. He then received 4 cycles of vinorelbine. He had progression-free survival of 7 months from first-line chemotherapy and stable disease for 7 months after 4 cycles of vinorelbine.
Discussion
In summary, we have a 73-year-old man admitted with a hemorrhagbic infarct, NSTEMI, and recently diagnosed right lower extremity DVT with a decreasing platelet count in the setting of LMWH. Throughout the hospital course, he had worsening hemorrhagic infarcts, preventing proper anticoagulation for his progressive thromboembolic events in the lungs, spleen, kidneys, and legs. Incidentally, he was also found to have a mass on a chest x-ray, later identified by biopsy as adenocarcinoma.
Given that the 4T scoring system for HIT showed a high probability with 8 points—identified by a platelet count fall > 50%, a platelet nadir > 20,000 cells/mm3, clear onset between days 5 and 14 with exposure to heparin/LMWH, new thrombosis, and no apparent cause of thrombocytopenia—suspicion for HIT remained high. Both functional and immunologic assays were negative for HIT, when repeated 2 weeks apart. The assays for laboratory diagnosis of HIT are immunologic, done by ELISA with a sensitivity of > 95% and a specificity of 50%–89%, and functional, done by SRA with a sensitivity > 90% and a specificity > 90%.2 As neither assay is 100% sensitive and specific, we still had a high clinical suspicion for HIT.
The HIT diagnostics in the presence of other comorbid states that may also induce thrombocytopenia represent a specific clinical problem. Despite increasing awareness of the clinical features of HIT, laboratory detection of the pathogenic HIT antibodies remains central to diagnosis. 4–6 This is because thrombocytopenia during heparin anticoagulation does not necessarily indicate HIT. Indeed, several other disorders complicated by thrombosis and thrombocytopenia during or shortly following heparin treatment strongly resemble HIT.These “pseudo-HIT”disorders7,8 (eg, cancer, sepsis, disseminated intravascular coagulation, pulmonary embolism, antiphospholipid syndrome) can reliably be distinguished from HIT by negative results using sensitive tests for HIT antibodies.
Thrombosis is strongly associated with HIT, with an incidence of 50%– 67%.9,10 The most common complication of HIT is venous thrombosis (DVT being the most frequent, followed by pulmonary embolism).9,11 Arterial thrombosis commonly presents as limb ischemia followed by cerebral vascular accident and myocardial infarction. Our patient had DVT followed by NSTEMI, cerebral vascular accident, and pulmonary embolism. He also had splenic and renal infarctions, which are rare in HIT. A literature review revealed, in abstract form, a retrospective study from a single institution showing a high incidence of thrombosis in a patient with a high 4T score and negative SRAs.12
The most common causes of thrombocytopenia in cancer are related to cancer treatment and bone marrow invasion by tumor cells. Chemotherapy and radiation therapy are damaging to the bone marrow and can cause severe myelosuppression, which results in lowering of platelet counts as well as white and red blood cell counts. It commonly occurs in patients with leukemia and lymphoma, but there are many other cancer types that can spread to bone marrow. Other causes of thrombocytopenia in cancer include the syndrome of disseminated intravascular coagulation and thrombotic microangiopathy.13
Nonbacterial thrombotic endocarditis (NBTE) is a disease characterized by the presence of vegetations on cardiac valves, consisting of fibrin and platelet aggregates devoid of inflammation or bacteria. NBTE has increasingly been recognized as a condition associated with numerous diseases and a potentially life-threatening source of thromboembolism. NBTE is not a common entity; however, it is frequently underestimated, probably due to underlying diseases (cancer, autoimmune disorders, HIV). NBTE is difficult to diagnose and relies on strong clinical suspicion. NBTE is also difficult to manage, and each case should be individually managed by identifying and treating the underlying pathology.14 Even though our patient had thromboembolism, there was no evidence of vegetations on cardiac valves by transthoracic or transesophageal echocardiography.
Trousseau’s syndrome is a paraneoplastic syndrome characterized by hypercoagulability related to malignancy. Coagulation abnormalities may include disseminated intravascular coagulation, pulmonary embolism, various types of gangrene, thrombotic endocarditis, arterial thrombosis, and embolic stoke.15 We considered this with our patient; however, a literature review showed no cases of Trousseau’s syndrome associated with thrombocytopenia, although concurrent Trousseau’s syndrome and HIT could not be excluded.
In summary, we need to consider all the above differential diagnoses in a patient presenting with thrombocytopenia and thrombosis. Treatment relies on clinical correlation of all the findings and supporting data.
1. Arepally GM, Ortel TL. Heparin-induced thrombocytopenia. N Engl J Med 2006;355:809–817.
2. Cuker A, Crowther MA. 2009 Clinical Practice Guideline on the Evaluation andManagement of Heparin-Induced Thrombocytopenia (HIT). American Society of Hematology Quick Reference. http://www.hematology. org/Practice/Guidelines/2934.aspx. Accessed May 6, 2011.
3. Antonijevic NM, Radovanovic N, Obradovic S, et al. Obstacles in the diagnostics and therapy of heparin-induced thrombocytopenia. Srp Arh Celok Lek 2010;138(suppl 1):69–73.
4. Warkentin TE, Chong BH, Greinacher A.Heparin induced thrombocytopenia: towards consensus. Thromb Haemost 1998;79:1–7.
5. Warkentin TE. Heparin-induced thrombocytopenia: a clinicopathologic syndrome.
Thromb Haemost 1999; 82:439–447. 6. Warkentin TE, Greinacher A. Laboratory testing for heparin-induced thrombocytopenia.J Thromb Thrombolysis 2000:10(suppl 1):35–45.
7. Warkentin TE. Pseudo-heparin-induced thrombocytopenia. In: Warkentin TE, Greinacher A, eds. Heparin-Induced Thrombocytopenia. New York: Marcel Dekker, Inc.; 2000:245–260.
8. Warkentin TE. Venous limb gangrene (VLG) complicating warfarin treatment of deep-vein thrombosis (DVT) in metastatic carcinoma (abstract). Blood 1999;94(suppl 1):114b.
9. Warkentin TE, Kelton JG. A 14-year study of heparin-induced thrombocytopenia. Am J Med 1996;101:502–507.
10. Greinacher A, Volpe H, Janssens U, et al. Recombinant hirudin (lepirudin) provides safe and effective anticoagulation in patients with heparin-induced thrombocytopenia: a prospective study. Circulation 1999;99:73–80.
11. Nand S, Wong W, Yuen B, et al. Heparin induced thrombocytopenia with thrombosis: incidence, analysis of risk factors, and clinical outcomes in 108 consecutive patients treated at a single institution. Am J Hematol 1997;56:12–16.
12. Hueser C, Patel AJ, Allan JN. Incidence of thrombosis in serotonin release assay negative patients and correlation with pretest heparin- induced thrombocytopenia scoring system. Blood 2008;112:1816.
13. Prandoni P, Falanga A, Piccioli A. Cancer, thrombosis and heparin-induced thrombocytopenia. Thromb Res 2007;120(suppl 2):S137–S140.
14. Asopa S, Patel A, Khan OA, Sharma R, Ohri SK. Non-bacterial thrombotic endocarditis. Eur J Cardiothoracic Surg 2007;32:696– 701.
15. Lim BR, Henry DH. Stroke syndrome secondary to hypercoagulability of lung cancer. Commun Oncol 2008;5:595–596.
Heparin-induced thrombocytopenia (HIT) is a life-threatening disorder that follows exposure to unfractionated heparin or (less commonly) low-molecular-weight heparin (LMWH). Patients classically present with a low platelet count (< 150,000 cells/mm3) or a relative decrease of 50% or more from baseline, although the fall may be less (eg, 30%–40%) in some patients.Thrombotic complications develop in approximately 20%–50% of patients.
HIT is caused by antibodies against complexes of platelet factor 4 and heparin.These antibodies are present in nearly all patients who receive a clinical diagnosis of the disorder and are also known to cause disease in animals.However,they are also present in many patients who have been exposed to heparin in various clinical settings but who do not develop clinical manifestations. It is uncertain why complications occur in some patients but not in others.1 We present a 73-year-old man who developed thrombocytopenia after starting LMWH and who has newly diagnosed adenocarcinoma of the lungs with extensive arterial and venous thrombosis and a negative serology for HIT.
Case presentation
A 73-year-old man presented to the emergency department after waking up in the morning with right-sided vague weakness and an inability to get out of bed. He had a history of right parietal stroke 1 month before the current presentation, when he was diagnosed with an aortic arch atheroma and started on warfarin. (At that time, CT scan of the head showed a right posterior temporoparietal lobe infarct in the posterior right middle cerebral artery distribution, and MRI of the brain and magnetic resonance angiography showed acute or subacute infarction in the distribution of the posterior division of the right middle cerebral artery, likely embolic, and tiny acute infarctions in the left frontal lobe.) This patient had been admitted 5 days prior to the current presentation for right lower extremity deep vein thrombosis (DVT) and was discharged after being prescribed enoxaparin (60 mg subcutaneously every 12 hours) and warfarin as per international normalized ratio (INR) daily.
Also included in the medical history was supraventricular tachycardia status post ablation, non–ST elevation myocardial infarction (NSTEMI), hypertension, hyperlipidemia, and macular degeneration. He had no surgical history. The patient had a family history of coronary artery disease. He had an extensive smoking history up until the day of admission. His medications on admission included atorvastatin (Lipitor; 20 mg daily), warfarin daily as per INR, enoxaparin (60 mg subcutaneously every 12 hours), amlodipine (5 mg daily), and aspirin (81 mg daily).
Pertinent initial laboratory results on admission were as follows: hemoglobin, 12.9 g/dL; white blood cell count, 8.6 ×109/L; platelet count, 183,000 cells/ mm3; INR, 1.2; and initial troponin level, negative.His admission chest x-ray showed a 4.5 cm × 5.5 cm lobulated density in the right hilum, suspicious for a hilar or subcarinal mass. Initial peripheral blood smear showed an isolated platelet decrease with increased size and no schistocytes. Initial CT of the head on admission showed no evidence of acute transcortical infarction and no definite evidence of acute intracranial hemorrhage but did show interval evolution of the right middle cerebral artery and left watershed distribution infarctions, with a probable small region of laminar necrosis in the right parietal lobe.
Clinical Course
The patient was initially thought to have had a transient ischemic attack causing aphasia, confusion, and rightsided weakness. He was started on therapeutic anticoagulation with dalteparin (Fragmin; 12,000 U subcutaneously daily),and enoxaparin was discontinued. The following day,his platelet count was 86,000 cells/mm3,down from an admission platelet count of 183,000 cells/mm3. A subsequent MRI of the brain showed a new hemorrhagic area in the right parietal infarct. The decision was made to stop anticoagulation, even though he had an embolic source from his aortic arch atheroma and lower extremity DVT.
The patient then underwent inferior vena cava (IVC) filter placement to prevent pulmonary thromboembolism and was transferred to the medical service due to low platelet count and an episode of nine beats of ventricular tachycardia. Subsequently, his troponin level was found to be elevated > 12 ng/mL, without significant electrocardiographic changes. He was diagnosed as having NSTEMI. Given his conversion from an ischemic to hemorrhagic CNS infarct and decrease in platelet count after LMWH exposure, HIT became a concern, and both anticoagulation and antiplatelet agents were held. The patient’s platelet count continued to trend downward over the next 3 days to a low of 27,000 cells/mm3. An HIT panel was negative by both immunologic and functional assays.
A CT scan of the brain 3 days after admission to monitor the hemorrhagic infarct showed multiple evolving infarcts and a new left occipital hemorrhagic infarct. The following day, a repeat CT scan of the head showed mulatiple evolving infarcts of varying ages, some with hemorrhage, and a mild interval increase in the previously described left medial parietal and left occipital lobe infarcts.
With worsening of his hemorrhagic infarct, along with his low platelet count and negative HIT panel,the decision was made to transfuse 2 units of platelets. His platelet count increased to 64,000 cells/mm3 after transfusion, subsequently dropping to 44,000 cells/mm3. However, during this time, the patient began to have worsening right lower extremity pain and left upper quadrant abdominal pain.
A CT scan of the thorax showed multifocal right hilar adenopathy suspicious for malignancy, either metastatic or representing a central lung carcinoma. It also showed nonocclusive segmental and possibly subsegmental pulmonary emboli in the right lower and middle lobes, as well as hypodense areas in the spleen, suggestive of areas of splenic infarction. Echocardiography showed an ejection fraction of 60%–70%, diastolic dysfunction, mildly elevated pulmonary artery pressure, and no evidence of patent foramen ovale.A cardiac stress test showed no reversible defects and an ejection fraction of 63%.
Risk of further bleeding into the brain was thought to be too great to initiate anticoagulation despite the CT thorax findings. The neurologist recommended waiting 2 weeks post hemorrhagic infarction before beginning anticoagulation. Antiphospholipid antibody syndrome was ruled out, with a negative lupus anticoagulant and anticardiolipin antibody. Also, negative blood cultures, normal fibrinogen levels, and normal haptoglobin levels ruled out disseminated intravascular coagulation. D-dimer was elevated but nonspecific, secondary to malignancy and multiple infarcts. He was started on aspirin (81 mg daily) 9 days after admission.
The patient had a repeat CT scan of the thorax and CT scan of the abdomen and pelvis due to continued abdominal pain. The CT scans showed multiple subsegmental pulmonary emboli, greatest in the right lower lobe, some of which were new since the prior study; continued evidence of multifocal splenic infarction (Figure 3); and multiple right and left kidney infarcts.
The patient then underwent endobronchial ultrasound (EBUS)-guided biopsy of his right hilar adenopathy to confirm the diagnosis of suspected malignancy. After the procedure, he developed right upper quadrant pleuritic pain with a low-grade fever. A repeat CT scan of the thorax showed a marked increase in the extent of the right lower lobe pulmonary emboli, with a new small embolus noted in the anterior segment of the right upper lobe. There was a thrombus inferior to the IVC filter, with probable mild extension of a thrombus superior to the filter as well, and again multiple splenic and bilateral renal infarcts.
With progression of thrombosis and now post EBUS, anticoagulation was initiated with argatroban and warfarin. His D-dimer was followed daily and remained high, despite therapeutic anticoagulation with warfarin. Given the persistently elevated D-dimer, the hematologist recommended discontinuing warfarin and starting fondaparinux (Arixtra) subcutaneously. His platelet count improved to a range of 156,000 cells/mm3 to 181,000 cells/mm3, even before the initiation of chemotherapy.
Follow-up
HIT was suspected clinically by classic drop in platelet count but was negative on enzyme-linked immunosorbent assay (ELISA) and serotonin release assay (SRA). The patient has been maintained on fondaparinux for anticoagulation, avoiding heparin. Factor V Leiden and lupus anticoagulant were negative.
Fondaparinux was discontinued after 3 months, and the patient presented again with swelling of his right lower extremity. Ultrasonography of the right lower extremity redemonstrated an occlusive thrombus in the peripheral portion of the right femoaral vein and throughout the right peroneal vein. The patient was restarted on fondaparinux (7.5 mg subcutaneously daily). During this follow-up, his platelet count ranged from 134,000 cells/mm3 to 193,000 cells/mm3.
Regarding management of nonsmall cell lung carcinoma of the left upper lobe (stains positive for TTF- 1 [thyroid transcription factor-1], CK7, and CK20; weakly positive for CK5/6; and negative for P63) with metastasis to bone and adrenal glands, he received 4 cycles of paclitaxel/carboplatin, with improved disease. A repeat CT of the chest, abdomen, and pelvis after chemotherapy showed improvement in mediastinal and hilar lymphadenopathy, resolution of extensive right lower lobe pulmonary consolidation, resolution of right-sided effusion, and no evidence of metastatic malignancy in the abdomen or pelvis and no osseous metastasis.
He was started on maintenance therapy with pemetrexed (Alimta), which was continued for 4 months, until repeat CT revealed progressive disease. He then received 4 cycles of vinorelbine. He had progression-free survival of 7 months from first-line chemotherapy and stable disease for 7 months after 4 cycles of vinorelbine.
Discussion
In summary, we have a 73-year-old man admitted with a hemorrhagbic infarct, NSTEMI, and recently diagnosed right lower extremity DVT with a decreasing platelet count in the setting of LMWH. Throughout the hospital course, he had worsening hemorrhagic infarcts, preventing proper anticoagulation for his progressive thromboembolic events in the lungs, spleen, kidneys, and legs. Incidentally, he was also found to have a mass on a chest x-ray, later identified by biopsy as adenocarcinoma.
Given that the 4T scoring system for HIT showed a high probability with 8 points—identified by a platelet count fall > 50%, a platelet nadir > 20,000 cells/mm3, clear onset between days 5 and 14 with exposure to heparin/LMWH, new thrombosis, and no apparent cause of thrombocytopenia—suspicion for HIT remained high. Both functional and immunologic assays were negative for HIT, when repeated 2 weeks apart. The assays for laboratory diagnosis of HIT are immunologic, done by ELISA with a sensitivity of > 95% and a specificity of 50%–89%, and functional, done by SRA with a sensitivity > 90% and a specificity > 90%.2 As neither assay is 100% sensitive and specific, we still had a high clinical suspicion for HIT.
The HIT diagnostics in the presence of other comorbid states that may also induce thrombocytopenia represent a specific clinical problem. Despite increasing awareness of the clinical features of HIT, laboratory detection of the pathogenic HIT antibodies remains central to diagnosis. 4–6 This is because thrombocytopenia during heparin anticoagulation does not necessarily indicate HIT. Indeed, several other disorders complicated by thrombosis and thrombocytopenia during or shortly following heparin treatment strongly resemble HIT.These “pseudo-HIT”disorders7,8 (eg, cancer, sepsis, disseminated intravascular coagulation, pulmonary embolism, antiphospholipid syndrome) can reliably be distinguished from HIT by negative results using sensitive tests for HIT antibodies.
Thrombosis is strongly associated with HIT, with an incidence of 50%– 67%.9,10 The most common complication of HIT is venous thrombosis (DVT being the most frequent, followed by pulmonary embolism).9,11 Arterial thrombosis commonly presents as limb ischemia followed by cerebral vascular accident and myocardial infarction. Our patient had DVT followed by NSTEMI, cerebral vascular accident, and pulmonary embolism. He also had splenic and renal infarctions, which are rare in HIT. A literature review revealed, in abstract form, a retrospective study from a single institution showing a high incidence of thrombosis in a patient with a high 4T score and negative SRAs.12
The most common causes of thrombocytopenia in cancer are related to cancer treatment and bone marrow invasion by tumor cells. Chemotherapy and radiation therapy are damaging to the bone marrow and can cause severe myelosuppression, which results in lowering of platelet counts as well as white and red blood cell counts. It commonly occurs in patients with leukemia and lymphoma, but there are many other cancer types that can spread to bone marrow. Other causes of thrombocytopenia in cancer include the syndrome of disseminated intravascular coagulation and thrombotic microangiopathy.13
Nonbacterial thrombotic endocarditis (NBTE) is a disease characterized by the presence of vegetations on cardiac valves, consisting of fibrin and platelet aggregates devoid of inflammation or bacteria. NBTE has increasingly been recognized as a condition associated with numerous diseases and a potentially life-threatening source of thromboembolism. NBTE is not a common entity; however, it is frequently underestimated, probably due to underlying diseases (cancer, autoimmune disorders, HIV). NBTE is difficult to diagnose and relies on strong clinical suspicion. NBTE is also difficult to manage, and each case should be individually managed by identifying and treating the underlying pathology.14 Even though our patient had thromboembolism, there was no evidence of vegetations on cardiac valves by transthoracic or transesophageal echocardiography.
Trousseau’s syndrome is a paraneoplastic syndrome characterized by hypercoagulability related to malignancy. Coagulation abnormalities may include disseminated intravascular coagulation, pulmonary embolism, various types of gangrene, thrombotic endocarditis, arterial thrombosis, and embolic stoke.15 We considered this with our patient; however, a literature review showed no cases of Trousseau’s syndrome associated with thrombocytopenia, although concurrent Trousseau’s syndrome and HIT could not be excluded.
In summary, we need to consider all the above differential diagnoses in a patient presenting with thrombocytopenia and thrombosis. Treatment relies on clinical correlation of all the findings and supporting data.
Heparin-induced thrombocytopenia (HIT) is a life-threatening disorder that follows exposure to unfractionated heparin or (less commonly) low-molecular-weight heparin (LMWH). Patients classically present with a low platelet count (< 150,000 cells/mm3) or a relative decrease of 50% or more from baseline, although the fall may be less (eg, 30%–40%) in some patients.Thrombotic complications develop in approximately 20%–50% of patients.
HIT is caused by antibodies against complexes of platelet factor 4 and heparin.These antibodies are present in nearly all patients who receive a clinical diagnosis of the disorder and are also known to cause disease in animals.However,they are also present in many patients who have been exposed to heparin in various clinical settings but who do not develop clinical manifestations. It is uncertain why complications occur in some patients but not in others.1 We present a 73-year-old man who developed thrombocytopenia after starting LMWH and who has newly diagnosed adenocarcinoma of the lungs with extensive arterial and venous thrombosis and a negative serology for HIT.
Case presentation
A 73-year-old man presented to the emergency department after waking up in the morning with right-sided vague weakness and an inability to get out of bed. He had a history of right parietal stroke 1 month before the current presentation, when he was diagnosed with an aortic arch atheroma and started on warfarin. (At that time, CT scan of the head showed a right posterior temporoparietal lobe infarct in the posterior right middle cerebral artery distribution, and MRI of the brain and magnetic resonance angiography showed acute or subacute infarction in the distribution of the posterior division of the right middle cerebral artery, likely embolic, and tiny acute infarctions in the left frontal lobe.) This patient had been admitted 5 days prior to the current presentation for right lower extremity deep vein thrombosis (DVT) and was discharged after being prescribed enoxaparin (60 mg subcutaneously every 12 hours) and warfarin as per international normalized ratio (INR) daily.
Also included in the medical history was supraventricular tachycardia status post ablation, non–ST elevation myocardial infarction (NSTEMI), hypertension, hyperlipidemia, and macular degeneration. He had no surgical history. The patient had a family history of coronary artery disease. He had an extensive smoking history up until the day of admission. His medications on admission included atorvastatin (Lipitor; 20 mg daily), warfarin daily as per INR, enoxaparin (60 mg subcutaneously every 12 hours), amlodipine (5 mg daily), and aspirin (81 mg daily).
Pertinent initial laboratory results on admission were as follows: hemoglobin, 12.9 g/dL; white blood cell count, 8.6 ×109/L; platelet count, 183,000 cells/ mm3; INR, 1.2; and initial troponin level, negative.His admission chest x-ray showed a 4.5 cm × 5.5 cm lobulated density in the right hilum, suspicious for a hilar or subcarinal mass. Initial peripheral blood smear showed an isolated platelet decrease with increased size and no schistocytes. Initial CT of the head on admission showed no evidence of acute transcortical infarction and no definite evidence of acute intracranial hemorrhage but did show interval evolution of the right middle cerebral artery and left watershed distribution infarctions, with a probable small region of laminar necrosis in the right parietal lobe.
Clinical Course
The patient was initially thought to have had a transient ischemic attack causing aphasia, confusion, and rightsided weakness. He was started on therapeutic anticoagulation with dalteparin (Fragmin; 12,000 U subcutaneously daily),and enoxaparin was discontinued. The following day,his platelet count was 86,000 cells/mm3,down from an admission platelet count of 183,000 cells/mm3. A subsequent MRI of the brain showed a new hemorrhagic area in the right parietal infarct. The decision was made to stop anticoagulation, even though he had an embolic source from his aortic arch atheroma and lower extremity DVT.
The patient then underwent inferior vena cava (IVC) filter placement to prevent pulmonary thromboembolism and was transferred to the medical service due to low platelet count and an episode of nine beats of ventricular tachycardia. Subsequently, his troponin level was found to be elevated > 12 ng/mL, without significant electrocardiographic changes. He was diagnosed as having NSTEMI. Given his conversion from an ischemic to hemorrhagic CNS infarct and decrease in platelet count after LMWH exposure, HIT became a concern, and both anticoagulation and antiplatelet agents were held. The patient’s platelet count continued to trend downward over the next 3 days to a low of 27,000 cells/mm3. An HIT panel was negative by both immunologic and functional assays.
A CT scan of the brain 3 days after admission to monitor the hemorrhagic infarct showed multiple evolving infarcts and a new left occipital hemorrhagic infarct. The following day, a repeat CT scan of the head showed mulatiple evolving infarcts of varying ages, some with hemorrhage, and a mild interval increase in the previously described left medial parietal and left occipital lobe infarcts.
With worsening of his hemorrhagic infarct, along with his low platelet count and negative HIT panel,the decision was made to transfuse 2 units of platelets. His platelet count increased to 64,000 cells/mm3 after transfusion, subsequently dropping to 44,000 cells/mm3. However, during this time, the patient began to have worsening right lower extremity pain and left upper quadrant abdominal pain.
A CT scan of the thorax showed multifocal right hilar adenopathy suspicious for malignancy, either metastatic or representing a central lung carcinoma. It also showed nonocclusive segmental and possibly subsegmental pulmonary emboli in the right lower and middle lobes, as well as hypodense areas in the spleen, suggestive of areas of splenic infarction. Echocardiography showed an ejection fraction of 60%–70%, diastolic dysfunction, mildly elevated pulmonary artery pressure, and no evidence of patent foramen ovale.A cardiac stress test showed no reversible defects and an ejection fraction of 63%.
Risk of further bleeding into the brain was thought to be too great to initiate anticoagulation despite the CT thorax findings. The neurologist recommended waiting 2 weeks post hemorrhagic infarction before beginning anticoagulation. Antiphospholipid antibody syndrome was ruled out, with a negative lupus anticoagulant and anticardiolipin antibody. Also, negative blood cultures, normal fibrinogen levels, and normal haptoglobin levels ruled out disseminated intravascular coagulation. D-dimer was elevated but nonspecific, secondary to malignancy and multiple infarcts. He was started on aspirin (81 mg daily) 9 days after admission.
The patient had a repeat CT scan of the thorax and CT scan of the abdomen and pelvis due to continued abdominal pain. The CT scans showed multiple subsegmental pulmonary emboli, greatest in the right lower lobe, some of which were new since the prior study; continued evidence of multifocal splenic infarction (Figure 3); and multiple right and left kidney infarcts.
The patient then underwent endobronchial ultrasound (EBUS)-guided biopsy of his right hilar adenopathy to confirm the diagnosis of suspected malignancy. After the procedure, he developed right upper quadrant pleuritic pain with a low-grade fever. A repeat CT scan of the thorax showed a marked increase in the extent of the right lower lobe pulmonary emboli, with a new small embolus noted in the anterior segment of the right upper lobe. There was a thrombus inferior to the IVC filter, with probable mild extension of a thrombus superior to the filter as well, and again multiple splenic and bilateral renal infarcts.
With progression of thrombosis and now post EBUS, anticoagulation was initiated with argatroban and warfarin. His D-dimer was followed daily and remained high, despite therapeutic anticoagulation with warfarin. Given the persistently elevated D-dimer, the hematologist recommended discontinuing warfarin and starting fondaparinux (Arixtra) subcutaneously. His platelet count improved to a range of 156,000 cells/mm3 to 181,000 cells/mm3, even before the initiation of chemotherapy.
Follow-up
HIT was suspected clinically by classic drop in platelet count but was negative on enzyme-linked immunosorbent assay (ELISA) and serotonin release assay (SRA). The patient has been maintained on fondaparinux for anticoagulation, avoiding heparin. Factor V Leiden and lupus anticoagulant were negative.
Fondaparinux was discontinued after 3 months, and the patient presented again with swelling of his right lower extremity. Ultrasonography of the right lower extremity redemonstrated an occlusive thrombus in the peripheral portion of the right femoaral vein and throughout the right peroneal vein. The patient was restarted on fondaparinux (7.5 mg subcutaneously daily). During this follow-up, his platelet count ranged from 134,000 cells/mm3 to 193,000 cells/mm3.
Regarding management of nonsmall cell lung carcinoma of the left upper lobe (stains positive for TTF- 1 [thyroid transcription factor-1], CK7, and CK20; weakly positive for CK5/6; and negative for P63) with metastasis to bone and adrenal glands, he received 4 cycles of paclitaxel/carboplatin, with improved disease. A repeat CT of the chest, abdomen, and pelvis after chemotherapy showed improvement in mediastinal and hilar lymphadenopathy, resolution of extensive right lower lobe pulmonary consolidation, resolution of right-sided effusion, and no evidence of metastatic malignancy in the abdomen or pelvis and no osseous metastasis.
He was started on maintenance therapy with pemetrexed (Alimta), which was continued for 4 months, until repeat CT revealed progressive disease. He then received 4 cycles of vinorelbine. He had progression-free survival of 7 months from first-line chemotherapy and stable disease for 7 months after 4 cycles of vinorelbine.
Discussion
In summary, we have a 73-year-old man admitted with a hemorrhagbic infarct, NSTEMI, and recently diagnosed right lower extremity DVT with a decreasing platelet count in the setting of LMWH. Throughout the hospital course, he had worsening hemorrhagic infarcts, preventing proper anticoagulation for his progressive thromboembolic events in the lungs, spleen, kidneys, and legs. Incidentally, he was also found to have a mass on a chest x-ray, later identified by biopsy as adenocarcinoma.
Given that the 4T scoring system for HIT showed a high probability with 8 points—identified by a platelet count fall > 50%, a platelet nadir > 20,000 cells/mm3, clear onset between days 5 and 14 with exposure to heparin/LMWH, new thrombosis, and no apparent cause of thrombocytopenia—suspicion for HIT remained high. Both functional and immunologic assays were negative for HIT, when repeated 2 weeks apart. The assays for laboratory diagnosis of HIT are immunologic, done by ELISA with a sensitivity of > 95% and a specificity of 50%–89%, and functional, done by SRA with a sensitivity > 90% and a specificity > 90%.2 As neither assay is 100% sensitive and specific, we still had a high clinical suspicion for HIT.
The HIT diagnostics in the presence of other comorbid states that may also induce thrombocytopenia represent a specific clinical problem. Despite increasing awareness of the clinical features of HIT, laboratory detection of the pathogenic HIT antibodies remains central to diagnosis. 4–6 This is because thrombocytopenia during heparin anticoagulation does not necessarily indicate HIT. Indeed, several other disorders complicated by thrombosis and thrombocytopenia during or shortly following heparin treatment strongly resemble HIT.These “pseudo-HIT”disorders7,8 (eg, cancer, sepsis, disseminated intravascular coagulation, pulmonary embolism, antiphospholipid syndrome) can reliably be distinguished from HIT by negative results using sensitive tests for HIT antibodies.
Thrombosis is strongly associated with HIT, with an incidence of 50%– 67%.9,10 The most common complication of HIT is venous thrombosis (DVT being the most frequent, followed by pulmonary embolism).9,11 Arterial thrombosis commonly presents as limb ischemia followed by cerebral vascular accident and myocardial infarction. Our patient had DVT followed by NSTEMI, cerebral vascular accident, and pulmonary embolism. He also had splenic and renal infarctions, which are rare in HIT. A literature review revealed, in abstract form, a retrospective study from a single institution showing a high incidence of thrombosis in a patient with a high 4T score and negative SRAs.12
The most common causes of thrombocytopenia in cancer are related to cancer treatment and bone marrow invasion by tumor cells. Chemotherapy and radiation therapy are damaging to the bone marrow and can cause severe myelosuppression, which results in lowering of platelet counts as well as white and red blood cell counts. It commonly occurs in patients with leukemia and lymphoma, but there are many other cancer types that can spread to bone marrow. Other causes of thrombocytopenia in cancer include the syndrome of disseminated intravascular coagulation and thrombotic microangiopathy.13
Nonbacterial thrombotic endocarditis (NBTE) is a disease characterized by the presence of vegetations on cardiac valves, consisting of fibrin and platelet aggregates devoid of inflammation or bacteria. NBTE has increasingly been recognized as a condition associated with numerous diseases and a potentially life-threatening source of thromboembolism. NBTE is not a common entity; however, it is frequently underestimated, probably due to underlying diseases (cancer, autoimmune disorders, HIV). NBTE is difficult to diagnose and relies on strong clinical suspicion. NBTE is also difficult to manage, and each case should be individually managed by identifying and treating the underlying pathology.14 Even though our patient had thromboembolism, there was no evidence of vegetations on cardiac valves by transthoracic or transesophageal echocardiography.
Trousseau’s syndrome is a paraneoplastic syndrome characterized by hypercoagulability related to malignancy. Coagulation abnormalities may include disseminated intravascular coagulation, pulmonary embolism, various types of gangrene, thrombotic endocarditis, arterial thrombosis, and embolic stoke.15 We considered this with our patient; however, a literature review showed no cases of Trousseau’s syndrome associated with thrombocytopenia, although concurrent Trousseau’s syndrome and HIT could not be excluded.
In summary, we need to consider all the above differential diagnoses in a patient presenting with thrombocytopenia and thrombosis. Treatment relies on clinical correlation of all the findings and supporting data.
1. Arepally GM, Ortel TL. Heparin-induced thrombocytopenia. N Engl J Med 2006;355:809–817.
2. Cuker A, Crowther MA. 2009 Clinical Practice Guideline on the Evaluation andManagement of Heparin-Induced Thrombocytopenia (HIT). American Society of Hematology Quick Reference. http://www.hematology. org/Practice/Guidelines/2934.aspx. Accessed May 6, 2011.
3. Antonijevic NM, Radovanovic N, Obradovic S, et al. Obstacles in the diagnostics and therapy of heparin-induced thrombocytopenia. Srp Arh Celok Lek 2010;138(suppl 1):69–73.
4. Warkentin TE, Chong BH, Greinacher A.Heparin induced thrombocytopenia: towards consensus. Thromb Haemost 1998;79:1–7.
5. Warkentin TE. Heparin-induced thrombocytopenia: a clinicopathologic syndrome.
Thromb Haemost 1999; 82:439–447. 6. Warkentin TE, Greinacher A. Laboratory testing for heparin-induced thrombocytopenia.J Thromb Thrombolysis 2000:10(suppl 1):35–45.
7. Warkentin TE. Pseudo-heparin-induced thrombocytopenia. In: Warkentin TE, Greinacher A, eds. Heparin-Induced Thrombocytopenia. New York: Marcel Dekker, Inc.; 2000:245–260.
8. Warkentin TE. Venous limb gangrene (VLG) complicating warfarin treatment of deep-vein thrombosis (DVT) in metastatic carcinoma (abstract). Blood 1999;94(suppl 1):114b.
9. Warkentin TE, Kelton JG. A 14-year study of heparin-induced thrombocytopenia. Am J Med 1996;101:502–507.
10. Greinacher A, Volpe H, Janssens U, et al. Recombinant hirudin (lepirudin) provides safe and effective anticoagulation in patients with heparin-induced thrombocytopenia: a prospective study. Circulation 1999;99:73–80.
11. Nand S, Wong W, Yuen B, et al. Heparin induced thrombocytopenia with thrombosis: incidence, analysis of risk factors, and clinical outcomes in 108 consecutive patients treated at a single institution. Am J Hematol 1997;56:12–16.
12. Hueser C, Patel AJ, Allan JN. Incidence of thrombosis in serotonin release assay negative patients and correlation with pretest heparin- induced thrombocytopenia scoring system. Blood 2008;112:1816.
13. Prandoni P, Falanga A, Piccioli A. Cancer, thrombosis and heparin-induced thrombocytopenia. Thromb Res 2007;120(suppl 2):S137–S140.
14. Asopa S, Patel A, Khan OA, Sharma R, Ohri SK. Non-bacterial thrombotic endocarditis. Eur J Cardiothoracic Surg 2007;32:696– 701.
15. Lim BR, Henry DH. Stroke syndrome secondary to hypercoagulability of lung cancer. Commun Oncol 2008;5:595–596.
1. Arepally GM, Ortel TL. Heparin-induced thrombocytopenia. N Engl J Med 2006;355:809–817.
2. Cuker A, Crowther MA. 2009 Clinical Practice Guideline on the Evaluation andManagement of Heparin-Induced Thrombocytopenia (HIT). American Society of Hematology Quick Reference. http://www.hematology. org/Practice/Guidelines/2934.aspx. Accessed May 6, 2011.
3. Antonijevic NM, Radovanovic N, Obradovic S, et al. Obstacles in the diagnostics and therapy of heparin-induced thrombocytopenia. Srp Arh Celok Lek 2010;138(suppl 1):69–73.
4. Warkentin TE, Chong BH, Greinacher A.Heparin induced thrombocytopenia: towards consensus. Thromb Haemost 1998;79:1–7.
5. Warkentin TE. Heparin-induced thrombocytopenia: a clinicopathologic syndrome.
Thromb Haemost 1999; 82:439–447. 6. Warkentin TE, Greinacher A. Laboratory testing for heparin-induced thrombocytopenia.J Thromb Thrombolysis 2000:10(suppl 1):35–45.
7. Warkentin TE. Pseudo-heparin-induced thrombocytopenia. In: Warkentin TE, Greinacher A, eds. Heparin-Induced Thrombocytopenia. New York: Marcel Dekker, Inc.; 2000:245–260.
8. Warkentin TE. Venous limb gangrene (VLG) complicating warfarin treatment of deep-vein thrombosis (DVT) in metastatic carcinoma (abstract). Blood 1999;94(suppl 1):114b.
9. Warkentin TE, Kelton JG. A 14-year study of heparin-induced thrombocytopenia. Am J Med 1996;101:502–507.
10. Greinacher A, Volpe H, Janssens U, et al. Recombinant hirudin (lepirudin) provides safe and effective anticoagulation in patients with heparin-induced thrombocytopenia: a prospective study. Circulation 1999;99:73–80.
11. Nand S, Wong W, Yuen B, et al. Heparin induced thrombocytopenia with thrombosis: incidence, analysis of risk factors, and clinical outcomes in 108 consecutive patients treated at a single institution. Am J Hematol 1997;56:12–16.
12. Hueser C, Patel AJ, Allan JN. Incidence of thrombosis in serotonin release assay negative patients and correlation with pretest heparin- induced thrombocytopenia scoring system. Blood 2008;112:1816.
13. Prandoni P, Falanga A, Piccioli A. Cancer, thrombosis and heparin-induced thrombocytopenia. Thromb Res 2007;120(suppl 2):S137–S140.
14. Asopa S, Patel A, Khan OA, Sharma R, Ohri SK. Non-bacterial thrombotic endocarditis. Eur J Cardiothoracic Surg 2007;32:696– 701.
15. Lim BR, Henry DH. Stroke syndrome secondary to hypercoagulability of lung cancer. Commun Oncol 2008;5:595–596.
Challenges of ensuring adherence to oral therapy in patients with solid malignancies
Arthur P. Staddon, MD
Joan Karnell Cancer Center, University of Pennsylvania School of Medicine, and Department of Hematology and Oncology, Pennsylvania Hospital, Philadelphia, PA
Manuscript received December 22, 2010; accepted June 10, 2011.
Correspondence to: Arthur P. Staddon, MD, Department of Hematology and Oncology, Pennsylvania Hospital of the University of Pennsylvania Health System, 230 West Washington Square, Philadelphia, PA 19107; e-mail: [email protected].
Oral anticancer drugs, particularly targeted therapies, are used increasingly to treat many solid malignancies. Adherence to the prescribed regimen is essential to ensuring that patients derive maximal clinical benefit from these
oral agents. However, multiple patient-related, treatment-related, and healthcare-associated factors may adversely impact adherence, thus compromising patient outcome. Reliable methods are not readily available in clinical practice to identify which patients are nonadherent. Therefore, clinicians need to take a proactive approach by assessing their patients’ needs, providing education about what can be expected during the course of oral therapy, monitoring adherence and reinforcing key points at all office visits, and using follow-up phone calls to identify issues that may still have an impact on adherence. By identifying and addressing barriers to adherence, oncologists can help their patients realize the full potential of oral therapy, including the promise of improved clinical outcome and quality of life.
Parenteral cytotoxic chemotherapy has traditionally been the major component of treatment for many solid malignancies—both in the adjuvant/neoadjuvant setting for high-risk localized disease and in the primary treatment for advanced and unresectable disease. Some oral anticancer agents, including tamoxifen, prednisone, and cyclophosphamide, have been used to manage certain malignancies for many years,1 and other oral cytotoxic drugs, such as capecitabine (Xeloda), have recently increased in usage.2,3. However, the advent of targeted therapy with agents that block specific cellular processes thought to be important in cell growth, survival, and metastasis has led to a large increase in the number of oral drugs for cancer.4 With the increased use of oral cancer drugs comes a shift in the treatment paradigm as it relates to patient management, particularly patient adherence to therapy. Examples of such oral targeted agents include the KIT/plateletderived growth factor receptor (PDGFR) tyrosine kinase inhibitors imatinib (Gleevec)5 and sunitinib (Sutent),6 the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors erlotinib (Tarceva) 7 and gefitinib (Iressa),8 the vascular endothelial growth factor receptor (VEGFR)/PDGFR kinase inhibitor pazopanib (Votrient),9 the multikinase inhibitor sorafenib (Nexavar),10 and the mamma lian target of rapamycin (mTOR) inhibitor everolimus (Afinitor).11
Notably, oral targeted therapy has emerged as a primary treatment of certain solid malignancies, including advanced gastrointestinal stromal tumors (GIST) and metastatic renal cell carcinoma.12 For example, imatinib is now recognized as first-line treatment of advanced GIST, and sunitinib is used in the second-line setting when patients relapse or cannot tolerate imatinib.12 Imatinib is also used in adjuvant therapy following complete surgical resection of localized GIST in patients with an intermediate or high risk of disease recurrence.12,13 Although no oral therapies are currently approved for treatment of bone or soft-tissue sarcomas, several are in clinical trials—the most advanced of which include two placebo-controlled, phase III trials evaluating the mTOR inhibitor ridaforolimus and the tyrosine kinase inhibitor pazopanib. Ridaforolimus is being evaluated as maintenance therapy for patients with favorable responses to chemotherapy, in the Sarcoma Multicenter Clinical Evalua tion of the Efficacy of Ridaforolimus (SUCCEED) trial.14 In the Pazopanib Explored in Soft-Tissue Sarcoma a Phase III (PALETTE) trial, pazopanib is being investigated in patients whose disease has progressed during or following prior therapy.15
Cancer patients generally prefer oral therapy over intravenous (IV) therapy, provided that efficacy is not compromised.1,16,17 Oral therapy offers greater convenience because it eliminates the time constraints imposed by receiving IV therapy at a physician’s office or treatment clinic. 18,19 This benefit may be particularly important for patients who live a significant distance from a treatment center.20 Oral therapy also reduces the discomfort and anxiety of having a venous catheter inserted for each treatment and it is easier in cases where venous access is difficult.20 Clinicians also are generally reported to prefer oral therapy, as it prevents complications such as infections and clotting associated with venous access and infusion pumps.2
The benefits of oral therapy must be balanced against several potential disadvantages, not the least of which is patient adherence to prescribed treatment. Patients receiving IV treatment may see their oncologist at intervals of 1 to 3 weeks, whereas those on oral agents may have less frequent visits. This arrangement limits educational and monitoring opportunities, as well as positive reinforcing interactions between patients and clinicians.22 The positive reinforcement and education provided by oncology nurses during IV treatment sessions are also lost.23 Perhaps most important, however, is that oral therapy requires patient self-medication and depends on the ability of patients to adhere to a prescribed regimen, which may involve daily treatment. More complex regimens may include multiple daily doses, drug holidays, or dosing based on the timing of meals.
Poor patient adherence is a well recognized issue in the treatment of many chronic diseases, including hypertension, diabetes, asthma, and mental illness.24 With oral therapies, long-term treatment is generally necessary to maintain disease control. However, adherence tends to decrease during long-term treatment, and therefore poor adherence may be more problematic during long-term maintenance with a targeted agent than with short-term use of an oral cytotoxic drug (Figure 1).25 The goal of maintaining high-level treatment adherence is to achieve optimal control of disease and to prevent, to the highest extent possible, disease progression.
Treatment adherence with oral anticancer agents
An overview
The terms “adherence” and “compliance” are used to describe the degree or extent to which a patient conforms to day-to-day treatment recommendations, including the timing, dosage, mode of administration, and frequency of the medication(s).26 Adherence is a synonym of compliance and is often perceived to be less judgmental of the patient’s behavior.27 Optimal adherence is achieved when a patient follows the treatment regimen exactly as prescribed (ie, without missing doses or taking extra doses). Therefore, adherence simply reflects the percentage of doses that are taken as prescribed.26
Monitoring treatment adherence in clinical practice is challenging, with all methods having significant limitations.1,27 Patient self-reporting may not accurately reflect treatment-taking behavior, because patients who want to please their clinicians will often overreport actual adherence. Patients completing medication diaries may erroneously fill in doses they failed to take, particularly at times close to scheduled visits. Additionally, these methods may provide unreliable information about the timing of doses, which is also important in ensuring appropriate systemic exposure to oral therapy. Other approaches have been used in clinical studies, including monitoring prescription refills and the microelectronic monitoring system (MEMS), which records each time the cap of the medication bottle is opened. However, opening the bottle does not ensure that the patient has taken the medication.
Therapeutic drug monitoring, measured through assessment of serum or urine drug levels, when available, may help to determine whether a patient has been adherent; however, it provides information for only the brief time leading up to testing and is limited by substantial interpatient variability in pharmacokinetics. Therapeutic drug monitoring also requires additional costs, further limiting its utility in clinical practice.
Adherence and the cancer patient
Cancer patients are generally thought to be highly motivated to follow their prescribed regimen.1,25 However, studies show that adherence with oral anticancer drugs is not optimal and that self-reporting and pill counts may not accurately reflect true patient behavior. Most information about medication adherence in patients with solid tumors comes from studies of tamoxifen in breast cancer.
In a study of a small cohort of breast cancer patients receiving tamoxifen for a mean of 3 months, the average adherence rates based on self-reports and pill counts were 98% (range, 91%–100%) and 92% (range, 74%–109%), respectively, with several patients taking more doses than prescribed. In comparison, MEMS showed an average rate of 69% (range, 33%–94%).28 Similar differences in adherence rates based on self-reports versus prescription refills have been seen with adjuvant tamoxifen and anastrozole. 29 Although self-reported adherence of 100% was claimed for both agents, only 80% and 69% of the women on tamoxifen and anastrozole, respectively, were still classified as adherent after controlling for prescription refills (P < 0.01 and P < 0.01 vs self-report).29 In general, electronic adherence measures, such as MEMS, show higher agreement with refilled pharmacy prescriptions than with patient self-reports.30
Adherence to oral cytotoxic drugs may also not be optimal in cancer patients. Adherence to oral capecitabine was assessed using MEMS in a cohort of 161 elderly women receiving adjuvant therapy for breast cancer. Twenty-four percent of the patients took fewer than 80% of the planned doses.31 Similarly, a nonadherence rate of 43% was reported for 51 breast cancer patients receiving oral cyclophosphamide, with higher nonadherence observed when women were treated in community practice settings than at academic centers.32 To date, assessments of treatment adherence to oral targeted therapies have shown a relatively high proportion of nonadherence in patients with hematologic malignancies (eg, chronic myelocytic leukemia).33 Adherence in patients with solid tumors has not been studied as extensively. Clinicians should suspect poor adherence when patients fail to achieve expected treatment responses within a certain time or when prescriptions are not filled as often as expected.
Impact of poor adherence
Patients with poor adherence may not receive the full benefit of treatment and may consequently experience poor clinical outcomes. Two sample populations illustrate this point—women with breast cancer and patients with advanced GIST. Adjuvant tamoxifen is well recognized to improve survival of women with hormone receptor-positive early- stage breast cancer.34 The impact of tamoxifen adherence on outcome was evaluated in a cohort of 1,633 women who received treatment for a median of 2.4 years.35 Median adherence was 93%; however, 315 women (19%) had adherence rates of less than 80%. In the multivariate analysis, a low adherence rate was independently associated with a higher risk of death. At the median study follow-up, the hazard ratio for mortality with adherence of less than 80% was 1.10 (95% confidence interval: 1.001–1.21; P = 0.046).35
Oral imatinib changed the treatment of advanced GIST, extending median survival to nearly 5 years.5 Poor adherence with daily imatinib leads to low serum drug levels, which in turn has been associated with poor clinical outcomes. Patients with serum imatinib levels less than 1,100 ng/mL—the lowest quartile in a recent analysis—had a median time to disease progression (TTP) of 11.3 months, which was significantly shorter than the TTP of greater than 30 months seen in the other three serum imatinib quartiles.36
Another randomized study compared planned interruption versus continuation of imatinib therapy in patients with advanced GIST.37 Most patients assigned to stop imatinib after achieving objective responses or stable disease on long-term therapy had rapid disease progression. Conversely, most patients who continued imatinib without interruption maintained the clinical benefit. Although treatment interruption was a planned event in this study, these results do suggest that poor adherence due to treatment interruptions may lead to disease progression. Patient adherence to imatinib in GIST has not been formally assessed in a clinical study. However, in patients with chronic myeloid leukemia, poor adherence to imatinib has been shown to have adverse consequences, resulting in suboptimal responses, disease relapse, and higher healthcare utilization and costs.33,38,39 Therefore, efforts to improve adherence may be expected to lead to better patient outcomes.
Barriers to adherence
Numerous factors have been identified as barriers to adherence in patients with chronic diseases.1,40 These factors can be grouped into three categories: patient-related factors, treatment-related factors, and healthcare system-related factors (Figure 2). When considered from a population perspective rather than from the perspective of an individual patient, poor adherence likely reflects a complex interplay among multiple factors.
Patient-related factors
These factors include sociodemographic, psychosocial, and employment status; comorbid conditions; polypharmacy; and social and family support characteristics; as well as the patient’s health beliefs, self-efficacy, and health literacy. Age may be a factor associated with poor adherence, which is particularly problematic among adolescents.41 Older individuals may also be prone to adherence issues for a variety of reasons, including the need to take multiple medications for comorbid conditions, visual and cognitive deficits, lack of social support if living alone, medication cost if living on a fixed budget, and higher risk of side effects due to drug-drug interactions and altered drug pharmacokinetics.42,43 Other demographic factors, such as race, educational level, and socioeconomic status, may indirectly affect adherence due to their impact on access to healthcare.1
Patient expectations are also key factors influencing adherence. On the one hand, patients who are not convinced about the importance of their therapy or who believe that their fate is governed largely by chance are more likely to exhibit poor adherence. 1,40 On the other hand, patients who believe that medication will be effective and that their own actions can influence the course of disease are more likely to adhere to treatment, even when faced with side effects. With these expectations in mind, clinicians can provide relevant information and encouragement and address their patients’ concerns.
Treatment-related factors
These factors include the complexity of the regimen; pill burden; duration of treatment; timing of drug administration; and type, frequency, and severity of side effects. Adherence is negatively affected by more complex regimens, such as those with multiple drugs and those that are inconvenient. 40 Medications that must be taken with respect to meals (eg, with meals or several hours before or after meals) may limit adherence in some patients.
Side effects, or the fear of side effects, have generally been shown to reduce adherence across multiple chronic diseases.40 However, clinical studies in cancer patients provide conflicting evidence about the impact of side effects on adherence.1,44 Side effects may occur early, before treatment benefits become evident, and obviously may be a significant barrier to adherence for some cancer patients. For instance, oral capecitabine is associated with a high incidence of grade 1/2 diarrhea,45 which may be expected to adversely impact adherence, although no data are available to support this theory. Lastly, as previously noted, adherence tends to decrease during long-term treatment.25
Healthcare system-related factors
The patient-provider relationship, along with patient access to and patient satisfaction with medical care, can affect adherence.27,40 Positive, constructive relationships between patients and clinicians are key to minimizing treatment-related anxiety and may improve adherence.46 The high cost of medication may be a major barrier for patients who do not have medical insurance and also for those whose insurance does not provide coverage or requires high copayments for oral anticancer medications. 3,47
Overcoming barriers to adherence in patient management
Set realistic patient expectations on treatment outcomes and side effects
Patient education and a collaborative patient-physician relationship are key components for overcoming barriers to adherence. In cases in which patients cannot afford oral treatment, patient-assistance programs may be helpful. Some patients, however, may refuse educational or assistance programs because they do not want to appear as being needy or as having failed.
Patient-centered education about what patients can expect from treatment— both in terms of benefits and side effects, as well as the steps to take to manage key side effects should they occur—has been shown to improve adherence.46 Realistic expectations about treatment benefits should be offered, as patients may wrongfully conclude that their treatment is not working if their too-high expectations are not met, leading some to stop treatment. Hence, patients on maintenance therapy with targeted drugs, such as erlotinib in non-small cell lung cancer or ridaforolimus in the ongoing SUCCEED trial in bone or soft-tissue sarcomas, should be informed that treatment is designed to keep their disease stable and not necessarily to further shrink the size of their tumors.48
Oral targeted agents may offer an improved tolerability profile compared with cytotoxic chemotherapy but nevertheless may cause specific, dose-limiting side effects related to target inhibition in normal cells (Table 1). Key side effects associated with each prescribed agent, as well as prophylactic steps that can be taken to prevent or minimize potential side effects and instructions to follow should side effects occur, should be discussed with patients before treatment is started.
For example, combined data from nine studies show that patients receiving the standard dose of sorafenib run a significant risk of hypertension.49 Physicians should closely monitor these patients during treatment and prescribe appropriate antihypertensive agents as needed.
An acneiform rash is common with erlotinib and other EGFR-targeting drugs; it is amenable to topical steroid therapy in mild cases and tetracyclines in moderate cases.50 ‘‘Aphthous-like’’ oral lesions are common with mTOR inhibitors, such as temsirolimus (Torisel), everolimus, and ridaforolimus, but they differ from the classic mucositis encountered with cytotoxic chemotherapy.51 Oral ulcerations are often amenable to prophylactic strategies: practicing good oral hygiene with brushing and flossing after each meal, avoiding spicy and acidic foods, drinking warm rather than hot beverages, and cleansing the mouth with baking soda rinses.52
Nausea and vomiting are well recognized side effects associated with cytotoxic chemotherapy and may be problematic with some oral anticancer drugs (Figure 3). Oral agents with a moderate emesis risk (30%–90%) include cyclophosphamide, etoposide, imatinib, and temozolomide (Temodar), and those with a low risk (10%– 30%) include capecitabine, oral fludarabine (Oforta), and pazopanib. 9,53,54 Most orally active targeted agents, except for imatinib, have a very low risk of emesis (ie, < 10%); they include erlotinib, gefitinib, sorafenib, and sunitinib. 54
Several classes of antiemetic agents are available for preventing nausea and vomiting, including serotonin 5-HT3 (5-hydroxytryptamine) agonists (eg, ondansetron, granisetron, dolasetron [Anzemet]), dexamethasone, and the neurokinin-1 (NK1) receptor antagonist aprepitant (Emend).54,55 Each drug is available in an oral formulation, which provides comparable prophylaxis as the corresponding IV formulation.56 Guidelines for prophylactic use of these agents are available for IV chemotherapy drugs and regimens based on their emetic risk.56,57 However, prospective data on the use of these agents during daily dosing with oral anticancer drugs are limited, and therefore treatment is largely empirical. 5
Dose reduction may be an option to manage side effects effectively while allowing patients to remain on treatment. In some cases, side effects may be an indicator of the efficacy of treatment—as in the case of rash with EGFR inhibitors—and the patient should be made aware of this possibility. 58,59
Provide effective patient education
Education can be offered in many formats but should be tailored to patient preferences, whenever possible, with sufficient time made available to assess patient needs and for patients to express concerns and ask questions. It is helpful to include a family member or caregiver who can reinforce educational information at home and encourage the patient to maintain adherence. Patient-oriented written materials and reliable online information sources may be offered to reinforce and supplement learning points made by the healthcare team.
In addition to providing information, clinicians can help to improve patient adherence by offering encouragement and empathy during each interaction with their patients.22,60 Regular follow-up to emphasize the need for adherence, answer additional questions, and obtain feedback about treatment is also an integral part of an effective strategy for promoting adherence. Patients may forget much of what their physicians tell them, particularly when the prognosis is poor.61 This scenario underscores the importance of frequent interactions and written materials for patient reference at home. Follow-up can be provided by periodic telephone calls from the oncology nursing staff or by group educational sessions—the latter should be for patients who have expressed an interest in attending such sessions.25,62
Perform routine monitoring and documentation of adherence
From the provider’s perspective, improvements can be achieved by routine monitoring of patient adherence, as well as assessment and reinforcement of patient understanding of the treatment goals and the need to adhere to treatment recommendations. This process also includes documentation of the patient’s treatment history, whether in written files or electronic medical records. IV chemotherapy drugs are documented in the chemotherapy flow sheet, which is a best clinical practice in oncology. In contrast, oral anticancer agents may be entered into patient records in the same manner as oral drugs used for other medical conditions (eg, antihypertensive drugs). Instead, it is preferable to enter information about oral anticancer agents directly onto the chemotherapy flow sheet as a care plan, including dosage, schedule, and all dosage adjustments. This approach makes the oral agent prominent in the cancer plan and may help to improve adherence through better documentation and patient follow-up.
Use experimental approaches for monitoring/improving adherence
Conventional methods for monitoring adherence, such as patient selfreports, medication diaries, and pill counts, are not always reliable in clinical practice, as already discussed. As a result, a number of other methods are being evaluated to monitor and improve patient adherence.
The feasibility of an automated voice-response system (AVRS) coupled with nursing intervention was recently evaluated in patients with solid tumors who were receiving oral chemotherapy agents.63 Patients received weekly calls from the AVRS and answered questions about adherence as well as the severity of 15 symptoms. Patients reporting adherence below 100% or symptom severity of 4 or higher, on a 0 to 10 scale, for 3 consecutive weeks were called by a nurse for assistance with treatment adherence and symptomatology. In the study cohort, nonadherence to oral chemotherapy was 23.3% and was related to both symptoms and missed or forgotten medication. Notably, better symptom management— and not symptom severity per se—was associated with higher rates of adherence.
Other technologies that may enhance patient adherence to oral drug therapy include daily cell phone alarms, text messages, or smartphone reminder applications. Although the use of text-message reminders requiring a patient response has been shown to increase adherence in patients with human immunodeficiency virus receiving antiretroviral therapy,64 the use of interactive mobile web or smartphone applications to enhance patient adherence to antihypertensive medications is still being investigated. 65
Building on the MEMS strategy, GlowCaps (Vitality, Inc., Cambridge, MA) is a pill bottle cap designed to replace the conventional cap provided by retail pharmacies. The electronic cap flashes and plays a ringtone when it is time for the next dose, and its wireless transmitter sends a signal to a reminder light plug that also flashes. If the cap is not opened, the transmitter dials the patient’s telephone with an additional reminder. The cap also creates weekly adherence reports that can be e‑mailed to a friend or family member, as well as monthly adherence reports that can be mailed to the healthcare provider. The GlowCap should improve adherence for patients who forget to take their medication, but like MEMS, it only measures whether the bottle was opened and not whether the medication was actually taken. Conventional bottle caps are supplied at no additional charge; however, this device carries a retail price of $99, and therefore cost can be an issue for patients on a fixed budget and for those requiring multiple medications.
A miniscule edible “chip” (Proteus Biomedical Inc., Redwood City, CA) may go one step farther by confirming that medication has actually been ingested. The chip, which is a digestible sensor made from food ingredients, is activated by the low pH in the stomach to send a signal to a microelectronic receiver located in a bandagestyle skin patch. The receiver records medication-related information, including the date, time, and dose taken. A pilot study showed that this technology substantially improved adherence for an antihypertensive agent from 30% to 80% over a 6-month period. 66 The cost of the edible chip is only a few cents each when made in large quantities, suggesting that it may be economically viable. The involvement of pharmacists in monitoring prescription refills and in providing patient reminders is also being evaluated for effectiveness (Medco Health Solutions, Inc., Fairfield, OH).
Consider economic issues
Oncologists in the United States are reimbursed for administering IV chemotherapy agents. This reimbursement includes the cost of the medication and an additional small percentage above the acquisition cost (eg, 6% from Medicare). Profit from IV chemotherapy (ie, reimbursement exceeding acquisition costs) may contribute to the financial viability of many oncology practices.67 However, a similar financial incentive for use of oral anticancer agents is not provided by payors, and therefore the current system favors use of IV therapy in cancer management. To rectify this situation, an alternate method will need to be developed to provide comparable incentives for oral anticancer drugs, particularly for those agents that improve patient outcome or reduce healthcare costs.
The added cost of patient education and adherence monitoring— both in terms of oncology staff and time—is another financial issue to be addressed. These costs cannot be borne by oncology practices but will need to be covered by payors or patients. Episode-based or monthly management fees provided by payors, which encompass the coordination of oncology care, may represent viable options for covering patient education and adherence monitoring.67
Conclusion
Oral agents are increasingly used to treat many solid malignancies— both as primary treatment in advanced cancers and as maintenance therapy in patients after response to first-line chemotherapy. In these settings, long-term use of oral agents offers the promise of transforming cancer into a chronic disease. With these changes in treatment regimen and disease state, patient adherence becomes increasingly important. Poor adherence with tamoxifen has already been associated with poor outcomes in women with breast cancer.35 As other oral anticancer agents are used for longer periods, it is likely that additional associations between poor adherence and adverse outcomes will be shown.
Clinicians cannot simply depend on self-reports or pill counts to identify nonadherent patients. Instead, they need to adopt a proactive role, which includes assessing patient needs and understanding, educating patients before initiating treatment, key points at all subsequent visits, and using follow-up phone calls to identify issues that impact adherence. If barriers to adherence are identified, whether attributable to patient, treatment, or healthcare system-related factors, oncologists and their staffs have an opportunity to play an essential role in addressing and reducing, if not eliminating, those barriers. Overcoming adherence barriers should result in better clinical outcomes and improved quality of life for patients with sarcomas and other solid tumors.
Acknowledgments: The author would like to thank Brigitte Teissedre, PhD, and Joseph J. Abrajano, PhD, of Medicus International New York, for editorial assistance in the preparation of this manuscript. Editorial support was funded by Merck & Co., Inc. The author was fully responsible for all content and editorial decisions and received no financial support or other compensation related to the development of this article.
References1. Partridge AH, Avorn J, Wang PS, Winer EP. Adherence to therapy with oral antineoplastic agents. J Natl Cancer Inst 2002;94:652– 661.
2. Gelmon K, Chan A, Harbeck N. The role of capecitabine in first-line treatment for patients with metastatic breast cancer. Oncologist 2006;11(suppl 1):42–51.
3. Aisner J. Overview of the changing paradigm in cancer treatment: oral chemotherapy. Am J Health Syst Pharm 2007;64(9 suppl 5):S4–S7.
4. Ma WW, Adjei AA. Novel agents on the horizon for cancer therapy. CA Cancer J Clin 2009;59:111–137.
5. Blanke CD, Demetri GD, von Mehren M, et al. Long-term results from a randomized phase II trial of standard- versus higher- dose imatinib mesylate for patients with unresectable or metastatic gastrointestinal stromal tumors expressing KIT. J Clin Oncol 2008;26:620–625.
6. Demetri GD, van Oosterom AT, Garrett CR, et al. Efficacy and safety of sunitinib in patients with advanced gastrointestinal stromal tumour after failure of imatinib: a randomised controlled trial. Lancet 2006;368:1329–1338.
7. Shepherd FA, Rodrigues Pereira J, Ciuleanu T, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005;353:123–132.
8. Sanford M, Scott LJ. Gefitinib: a review of its use in the treatment of locally advanced/ metastatic non-small cell lung cancer. Drugs 2009;69:2303–2328.
9. Sternberg CN, Davis ID, Mardiak J, et al. Pazopanib in locally advanced or metastatic renal cell carcinoma: results of a randomized phase III trial. J Clin Oncol 2010;28:1061– 1068.
10. Escudier B, Eisen T, Stadler WM, et al. Sorafenib for treatment of renal cell carcinoma: final efficacy and safety results of the phase III treatment approaches in renal cancer global evaluation trial. J Clin Oncol 2009;27:3312– 3318.
11. Motzer RJ, Escudier B, Oudard S, et al. Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial. Lancet 2008;372:449–456.
12. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology™. Soft tissue sarcoma. V1.2010. Available at: www.nccn.org. Accessed April 2, 2010.
13. Dematteo RP, Ballman KV, Antonescu CR, et al. Adjuvant imatinib mesylate after resection of localised, primary gastrointestinal stromal tumour: a randomised, double-blind, placebo-controlled trial. Lancet 2009;373:1097–1104.
14. Ariad Pharmaceuticals. European Organization for Research and Treatment of Cancer. Ridaforolimus in treatment of sarcoma- SUCCEED (Sarcoma Multi-Center Clinical Eval. of the Efficacy of Ridaforolimus). In: ClinicalTrials.gov. Bethesda (MD): National Library of Medicine (US). http://clinicaltrials. gov/ct2/show/NCT00538239?term=0053823 9&rank=1. NLM Identifier: NCT00538239. Accessed February 5, 2010.
15. GlaxoSmithKline. A randomized double blind phase III trial of pazopanib versus placebo in patients with soft tissue sarcoma whose disease has progressed during or following prior therapy (PALETTE). In: Clinical- Trials.gov. Bethesda (MD): National Library of Medicine (US). http://clinicaltrials.gov/ct2/ show/NCT00753688?term=palette&rank=1. NLM Identifier: NCT00753688. Accessed February 11, 2010.
16. Liu G, Franssen E, Fitch MI, Warner E. Patient preferences for oral versus intravenous palliative chemotherapy. J Clin Oncol 1997;15:110–115.
17. Fallowfield L, Atkins L, Catt S, et al. Patients’ preference for administration of endocrine treatments by injection or tablets: results from a study of women with breast cancer. Ann Oncol 2006;17:205–210.
18. Twelves C, Gollins S, Grieve R, Samuel L. A randomised cross-over trial comparing patient preference for oral capecitabine and 5-fluorouracil/leucovorin regimens in patients with advanced colorectal cancer. Ann Oncol 2006;17:239–245.
19. Jensen LH, Osterlind K, Rytter C. Randomized cross-over study of patient preference for oral or intravenous vinorelbine in combination with carboplatin in the treatment of advanced NSCLC. Lung Cancer 2008;62:85–91.
20. Findlay M, von Minckwitz G, Wardley A. Effective oral chemotherapy for breast cancer: pillars of strength. Ann Oncol 2008;19:212–222.
21. Pelusi J. Capecitabine versus 5-FU in metastatic colorectal cancer: considerations for treatment decision-making. Commun Oncol 2006;3:19–27.
22. Pollak KI, Arnold RM, Jeffreys AS, et al. Oncologist communication about emotion during visits with patients with advanced cancer. J Clin Oncol 2007;25:5748–5752.
23. Palmieri FM, Barton DL. Challenges of oral medications in patients with advanced breast cancer. Semin Oncol Nurs 2007;23(4 suppl 2):S17–S22.
24. Ingersoll KS, Cohen J. The impact of medication regimen factors on adherence to chronic treatment: a review of literature. J Behav Med 2008;31:213–224.
25. O’Neill VJ, Twelves CJ. Oral cancer treatment: developments in chemotherapy and beyond. Br J Cancer 2002;87:933–937.
26. Cramer JA, Roy A, Burrell A, et al. Medication compliance and persistence: terminology and definitions. Value Health 2008;11:44–47.
27. Ruddy K, Mayer E, Partridge A. Patient adherence and persistence with oral anticancer treatment. CA Cancer J Clin 2009;59:56–66.
28. Waterhouse DM, Calzone KA, Mele C, Brenner DE. Adherence to oral tamoxifen: a comparison of patient self-report, pill counts, and microelectronic monitoring. J Clin Oncol 1993;11:1189–1197.
29. Ziller V, Kalder M, Albert US, et al. Adherence to adjuvant endocrine therapy in postmenopausal women with breast cancer. Ann Oncol 2009;20:431–436.
30. Hansen RA, Kim MM, Song L, Tu W, Wu J, Murray MD. Comparison of methods to assess medication adherence and classify nonadherence. Ann Pharmacother 2009;43:413–422.
31. Muss HB, Berry DA, Cirrincione CT, et al. Adjuvant chemotherapy in older women with early-stage breast cancer. N Engl J Med 2009;360:2055–2065.
32. Lebovits AH, Strain JJ, Schleifer SJ, Tanaka JS, Bhardwaj S, Messe MR. Patient noncompliance with self-administered chemotherapy. Cancer 1990;65:17–22.
33. Noens L, van Lierde MA, De Bock R, et al. Prevalence, determinants, and outcomes of nonadherence to imatinib therapy in patients with chronic myeloid leukemia: the ADAGIO study. Blood 2009;113:5401–5411.
34. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005;365:1687–1717.
35. McCowan C, Shearer J, Donnan PT, et al. Cohort study examining tamoxifen adherence and its relationship to mortality in women with breast cancer. Br J Cancer 2008;99:1763–1768.
36. Demetri GD, Wang Y, Wehrle E, et al. Imatinib plasma levels are correlated with clinical benefit in patients with unresectable/metastatic gastrointestinal stromal tumors. J Clin Oncol 2009;27:3141–3147.
37. Blay JY, Le Cesne A, Ray-Coquard I, et al. Prospective multicentric randomized phase III study of imatinib in patients with advanced gastrointestinal stromal tumors comparing interruption versus continuation of treatment beyond 1 year: the French Sarcoma Group. J Clin Oncol 2007;25:1107–1113.
38. O’Dwyer M, Atallah E. Practical considerations for the management of patients in the tyrosine kinase inhibitor era. Semin Hematol 2009;46(2 suppl 3):S16–S21.
39. Wu EQ, Johnson S, Beaulieu N, et al. Healthcare resource utilization and costs associated with non-adherence to imatinib treatment in chronic myeloid leukemia patients. Curr Med Res Opin 2010;26:61–69.
40. Krueger KP, Berger BA, Felkey B. Medication adherence and persistence: a comprehensive review. Adv Ther 2005;22:313–356.
41. Tebbi CK. Treatment compliance in childhood and adolescence. Cancer 1993;71(10 suppl):3441–3449.
42. MacLaughlin EJ, Raehl CL, Treadway AK, Sterling TL, Zoller DP, Bond CA. Assessing medication adherence in the elderly: which tools to use in clinical practice? Drugs Aging 2005;22:231–255.
43. Tam-McDevitt J. Polypharmacy, aging, and cancer. Oncology 2008;22:1052–1055; discussion 1055, 1058, 1060.
44. Richardson JL, Marks G, Levine A. The influence of symptoms of disease and side effects of treatment on compliance with cancer therapy. J Clin Oncol 1988;6:1746–1752.
45. Walko CM, Lindley C. Capecitabine: a review. Clin Ther 2005;27:23–44.
46. Fallowfield LJ. Treatment decision- making in breast cancer: the patient-doctor relationship. Breast Cancer Res Treat 2008;112(suppl 1):5–13.
47. Elliott R. Non-adherence to medicines: not solved but solvable. J Health Serv Res Policy 2009;14:58–61.
48. Mok TS, Ramalingam SS. Maintenance therapy in nonsmall-cell lung cancer: a new treatment paradigm. Cancer 2009;115:5143– 5154.
49. Wu S, Chen JJ, Kudelka A, Lu J, Zhu X. Incidence and risk of hypertension with sorafenib in patients with cancer: a systematic review and meta-analysis. Lancet Oncol 2008;9:117–123.
50. Segaert S, Van Cutsem E. Clinical signs, pathophysiology and management of skin toxicity during therapy with epidermal growth factor receptor inhibitors. Ann Oncol 2005;16:1425–1433.
51. Sonis S, Treister N, Chawla S, Demetri G, Haluska F. Preliminary characterization of oral lesions associated with inhibitors of mammalian target of rapamycin in cancer patients. Cancer 2010;116:210–215.
52. Scully C. Clinical practice: aphthous ulceration. N Engl J Med. 2006;355:165–172.
53. Hurwitz HI, Dowlati A, Saini S, et al. Phase I trial of pazopanib in patients with advanced cancer. Clin Cancer Res 2009;15:4220– 4227.
54. Jordan K, Sippel C, Schmoll HJ. Guidelines for antiemetic treatment of chemotherapy- induced nausea and vomiting: past, present, and future recommendations. Oncologist 2007;12:1143–1150.
55. Hesketh PJ. Chemotherapy-induced nausea and vomiting. N Engl J Med 2008;358: 2482–2494.
56. Kris MG, Hesketh PJ, Somerfield MR, et al. American Society of Clinical Oncology guideline for antiemetics in oncology: update 2006. J Clin Oncol 2006;24:2932–2947.
57. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology™. Antiemesis. V1. 2010. Available at: http://www.nccn.org. Accessed May 5, 2010.
58. Perez-Soler R. Rash as a surrogate marker for efficacy of epidermal growth factor receptor inhibitors in lung cancer. Clin Lung Cancer 2006;8(suppl 1):S7–S14.
59. Wacker B, Nagrani T, Weinberg J, Witt K, Clark G, Cagnoni PJ. Correlation between development of rash and efficacy in patients treated with the epidermal growth factor receptor tyrosine kinase inhibitor erlotinib in two large phase III studies. Clin Cancer Res 2007;13:3913–3921.
60. Morse DS, Edwardsen EA, Gordon HS. Missed opportunities for interval empathy in lung cancer communication. Arch Intern Med 2008;168:1853–1858.
61. Jansen J, van Weert J, van der Meulen N, van Dulmen S, Heeren T, Bensing J. Recall in older cancer patients: measuring memory for medical information. Gerontologist 2008;48:149–157.
62. Moore S. Facilitating oral chemotherapy treatment and compliance through patient/ family-focused education. Cancer Nurs 2007;30:112–122; quiz 123–124.
63. Decker V, Spoelstra S, Miezo E, et al. A pilot study of an automated voice response system and nursing intervention to monitor adherence to oral chemotherapy agents. Cancer Nurs 2009;32:E20–E29.
64. Lester RT, Ritvo P, Mills EJ, et al. Effects of a mobile phone short message service on antiretroviral treatment adherence in Kenya (WelTel Kenya1): a randomised trial. Lancet 2010;376:1838–1845.
65. Medication Adherence and mHealth: The George Washington University and Wireless Reach Pill Phone Study. Available at: http://multivu.prnewswire.com/mnr/qualcomm/ 48588/docs/48588-PillPhone_CS_FINAL_ r2.pdf. Accessed May 3, 2011.
66. Saenz A. Ingestible chip that reminds you to take medicine begins trial. Singularity Hub. Available at: http://singularityhub. com/2009/09/24/ingestible-chip-that-remindsyou- to-take-medicine-begins-trials-withnovartis/# more-7574. Accessed April 6, 2010.
67. Smith TJ, Hillner BE. Concrete options and ideas for increasing value in oncology care: the view from one trench. Oncologist 2010;15(suppl 1):65–72.
Arthur P. Staddon, MD
Joan Karnell Cancer Center, University of Pennsylvania School of Medicine, and Department of Hematology and Oncology, Pennsylvania Hospital, Philadelphia, PA
Manuscript received December 22, 2010; accepted June 10, 2011.
Correspondence to: Arthur P. Staddon, MD, Department of Hematology and Oncology, Pennsylvania Hospital of the University of Pennsylvania Health System, 230 West Washington Square, Philadelphia, PA 19107; e-mail: [email protected].
Oral anticancer drugs, particularly targeted therapies, are used increasingly to treat many solid malignancies. Adherence to the prescribed regimen is essential to ensuring that patients derive maximal clinical benefit from these
oral agents. However, multiple patient-related, treatment-related, and healthcare-associated factors may adversely impact adherence, thus compromising patient outcome. Reliable methods are not readily available in clinical practice to identify which patients are nonadherent. Therefore, clinicians need to take a proactive approach by assessing their patients’ needs, providing education about what can be expected during the course of oral therapy, monitoring adherence and reinforcing key points at all office visits, and using follow-up phone calls to identify issues that may still have an impact on adherence. By identifying and addressing barriers to adherence, oncologists can help their patients realize the full potential of oral therapy, including the promise of improved clinical outcome and quality of life.
Parenteral cytotoxic chemotherapy has traditionally been the major component of treatment for many solid malignancies—both in the adjuvant/neoadjuvant setting for high-risk localized disease and in the primary treatment for advanced and unresectable disease. Some oral anticancer agents, including tamoxifen, prednisone, and cyclophosphamide, have been used to manage certain malignancies for many years,1 and other oral cytotoxic drugs, such as capecitabine (Xeloda), have recently increased in usage.2,3. However, the advent of targeted therapy with agents that block specific cellular processes thought to be important in cell growth, survival, and metastasis has led to a large increase in the number of oral drugs for cancer.4 With the increased use of oral cancer drugs comes a shift in the treatment paradigm as it relates to patient management, particularly patient adherence to therapy. Examples of such oral targeted agents include the KIT/plateletderived growth factor receptor (PDGFR) tyrosine kinase inhibitors imatinib (Gleevec)5 and sunitinib (Sutent),6 the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors erlotinib (Tarceva) 7 and gefitinib (Iressa),8 the vascular endothelial growth factor receptor (VEGFR)/PDGFR kinase inhibitor pazopanib (Votrient),9 the multikinase inhibitor sorafenib (Nexavar),10 and the mamma lian target of rapamycin (mTOR) inhibitor everolimus (Afinitor).11
Notably, oral targeted therapy has emerged as a primary treatment of certain solid malignancies, including advanced gastrointestinal stromal tumors (GIST) and metastatic renal cell carcinoma.12 For example, imatinib is now recognized as first-line treatment of advanced GIST, and sunitinib is used in the second-line setting when patients relapse or cannot tolerate imatinib.12 Imatinib is also used in adjuvant therapy following complete surgical resection of localized GIST in patients with an intermediate or high risk of disease recurrence.12,13 Although no oral therapies are currently approved for treatment of bone or soft-tissue sarcomas, several are in clinical trials—the most advanced of which include two placebo-controlled, phase III trials evaluating the mTOR inhibitor ridaforolimus and the tyrosine kinase inhibitor pazopanib. Ridaforolimus is being evaluated as maintenance therapy for patients with favorable responses to chemotherapy, in the Sarcoma Multicenter Clinical Evalua tion of the Efficacy of Ridaforolimus (SUCCEED) trial.14 In the Pazopanib Explored in Soft-Tissue Sarcoma a Phase III (PALETTE) trial, pazopanib is being investigated in patients whose disease has progressed during or following prior therapy.15
Cancer patients generally prefer oral therapy over intravenous (IV) therapy, provided that efficacy is not compromised.1,16,17 Oral therapy offers greater convenience because it eliminates the time constraints imposed by receiving IV therapy at a physician’s office or treatment clinic. 18,19 This benefit may be particularly important for patients who live a significant distance from a treatment center.20 Oral therapy also reduces the discomfort and anxiety of having a venous catheter inserted for each treatment and it is easier in cases where venous access is difficult.20 Clinicians also are generally reported to prefer oral therapy, as it prevents complications such as infections and clotting associated with venous access and infusion pumps.2
The benefits of oral therapy must be balanced against several potential disadvantages, not the least of which is patient adherence to prescribed treatment. Patients receiving IV treatment may see their oncologist at intervals of 1 to 3 weeks, whereas those on oral agents may have less frequent visits. This arrangement limits educational and monitoring opportunities, as well as positive reinforcing interactions between patients and clinicians.22 The positive reinforcement and education provided by oncology nurses during IV treatment sessions are also lost.23 Perhaps most important, however, is that oral therapy requires patient self-medication and depends on the ability of patients to adhere to a prescribed regimen, which may involve daily treatment. More complex regimens may include multiple daily doses, drug holidays, or dosing based on the timing of meals.
Poor patient adherence is a well recognized issue in the treatment of many chronic diseases, including hypertension, diabetes, asthma, and mental illness.24 With oral therapies, long-term treatment is generally necessary to maintain disease control. However, adherence tends to decrease during long-term treatment, and therefore poor adherence may be more problematic during long-term maintenance with a targeted agent than with short-term use of an oral cytotoxic drug (Figure 1).25 The goal of maintaining high-level treatment adherence is to achieve optimal control of disease and to prevent, to the highest extent possible, disease progression.
Treatment adherence with oral anticancer agents
An overview
The terms “adherence” and “compliance” are used to describe the degree or extent to which a patient conforms to day-to-day treatment recommendations, including the timing, dosage, mode of administration, and frequency of the medication(s).26 Adherence is a synonym of compliance and is often perceived to be less judgmental of the patient’s behavior.27 Optimal adherence is achieved when a patient follows the treatment regimen exactly as prescribed (ie, without missing doses or taking extra doses). Therefore, adherence simply reflects the percentage of doses that are taken as prescribed.26
Monitoring treatment adherence in clinical practice is challenging, with all methods having significant limitations.1,27 Patient self-reporting may not accurately reflect treatment-taking behavior, because patients who want to please their clinicians will often overreport actual adherence. Patients completing medication diaries may erroneously fill in doses they failed to take, particularly at times close to scheduled visits. Additionally, these methods may provide unreliable information about the timing of doses, which is also important in ensuring appropriate systemic exposure to oral therapy. Other approaches have been used in clinical studies, including monitoring prescription refills and the microelectronic monitoring system (MEMS), which records each time the cap of the medication bottle is opened. However, opening the bottle does not ensure that the patient has taken the medication.
Therapeutic drug monitoring, measured through assessment of serum or urine drug levels, when available, may help to determine whether a patient has been adherent; however, it provides information for only the brief time leading up to testing and is limited by substantial interpatient variability in pharmacokinetics. Therapeutic drug monitoring also requires additional costs, further limiting its utility in clinical practice.
Adherence and the cancer patient
Cancer patients are generally thought to be highly motivated to follow their prescribed regimen.1,25 However, studies show that adherence with oral anticancer drugs is not optimal and that self-reporting and pill counts may not accurately reflect true patient behavior. Most information about medication adherence in patients with solid tumors comes from studies of tamoxifen in breast cancer.
In a study of a small cohort of breast cancer patients receiving tamoxifen for a mean of 3 months, the average adherence rates based on self-reports and pill counts were 98% (range, 91%–100%) and 92% (range, 74%–109%), respectively, with several patients taking more doses than prescribed. In comparison, MEMS showed an average rate of 69% (range, 33%–94%).28 Similar differences in adherence rates based on self-reports versus prescription refills have been seen with adjuvant tamoxifen and anastrozole. 29 Although self-reported adherence of 100% was claimed for both agents, only 80% and 69% of the women on tamoxifen and anastrozole, respectively, were still classified as adherent after controlling for prescription refills (P < 0.01 and P < 0.01 vs self-report).29 In general, electronic adherence measures, such as MEMS, show higher agreement with refilled pharmacy prescriptions than with patient self-reports.30
Adherence to oral cytotoxic drugs may also not be optimal in cancer patients. Adherence to oral capecitabine was assessed using MEMS in a cohort of 161 elderly women receiving adjuvant therapy for breast cancer. Twenty-four percent of the patients took fewer than 80% of the planned doses.31 Similarly, a nonadherence rate of 43% was reported for 51 breast cancer patients receiving oral cyclophosphamide, with higher nonadherence observed when women were treated in community practice settings than at academic centers.32 To date, assessments of treatment adherence to oral targeted therapies have shown a relatively high proportion of nonadherence in patients with hematologic malignancies (eg, chronic myelocytic leukemia).33 Adherence in patients with solid tumors has not been studied as extensively. Clinicians should suspect poor adherence when patients fail to achieve expected treatment responses within a certain time or when prescriptions are not filled as often as expected.
Impact of poor adherence
Patients with poor adherence may not receive the full benefit of treatment and may consequently experience poor clinical outcomes. Two sample populations illustrate this point—women with breast cancer and patients with advanced GIST. Adjuvant tamoxifen is well recognized to improve survival of women with hormone receptor-positive early- stage breast cancer.34 The impact of tamoxifen adherence on outcome was evaluated in a cohort of 1,633 women who received treatment for a median of 2.4 years.35 Median adherence was 93%; however, 315 women (19%) had adherence rates of less than 80%. In the multivariate analysis, a low adherence rate was independently associated with a higher risk of death. At the median study follow-up, the hazard ratio for mortality with adherence of less than 80% was 1.10 (95% confidence interval: 1.001–1.21; P = 0.046).35
Oral imatinib changed the treatment of advanced GIST, extending median survival to nearly 5 years.5 Poor adherence with daily imatinib leads to low serum drug levels, which in turn has been associated with poor clinical outcomes. Patients with serum imatinib levels less than 1,100 ng/mL—the lowest quartile in a recent analysis—had a median time to disease progression (TTP) of 11.3 months, which was significantly shorter than the TTP of greater than 30 months seen in the other three serum imatinib quartiles.36
Another randomized study compared planned interruption versus continuation of imatinib therapy in patients with advanced GIST.37 Most patients assigned to stop imatinib after achieving objective responses or stable disease on long-term therapy had rapid disease progression. Conversely, most patients who continued imatinib without interruption maintained the clinical benefit. Although treatment interruption was a planned event in this study, these results do suggest that poor adherence due to treatment interruptions may lead to disease progression. Patient adherence to imatinib in GIST has not been formally assessed in a clinical study. However, in patients with chronic myeloid leukemia, poor adherence to imatinib has been shown to have adverse consequences, resulting in suboptimal responses, disease relapse, and higher healthcare utilization and costs.33,38,39 Therefore, efforts to improve adherence may be expected to lead to better patient outcomes.
Barriers to adherence
Numerous factors have been identified as barriers to adherence in patients with chronic diseases.1,40 These factors can be grouped into three categories: patient-related factors, treatment-related factors, and healthcare system-related factors (Figure 2). When considered from a population perspective rather than from the perspective of an individual patient, poor adherence likely reflects a complex interplay among multiple factors.
Patient-related factors
These factors include sociodemographic, psychosocial, and employment status; comorbid conditions; polypharmacy; and social and family support characteristics; as well as the patient’s health beliefs, self-efficacy, and health literacy. Age may be a factor associated with poor adherence, which is particularly problematic among adolescents.41 Older individuals may also be prone to adherence issues for a variety of reasons, including the need to take multiple medications for comorbid conditions, visual and cognitive deficits, lack of social support if living alone, medication cost if living on a fixed budget, and higher risk of side effects due to drug-drug interactions and altered drug pharmacokinetics.42,43 Other demographic factors, such as race, educational level, and socioeconomic status, may indirectly affect adherence due to their impact on access to healthcare.1
Patient expectations are also key factors influencing adherence. On the one hand, patients who are not convinced about the importance of their therapy or who believe that their fate is governed largely by chance are more likely to exhibit poor adherence. 1,40 On the other hand, patients who believe that medication will be effective and that their own actions can influence the course of disease are more likely to adhere to treatment, even when faced with side effects. With these expectations in mind, clinicians can provide relevant information and encouragement and address their patients’ concerns.
Treatment-related factors
These factors include the complexity of the regimen; pill burden; duration of treatment; timing of drug administration; and type, frequency, and severity of side effects. Adherence is negatively affected by more complex regimens, such as those with multiple drugs and those that are inconvenient. 40 Medications that must be taken with respect to meals (eg, with meals or several hours before or after meals) may limit adherence in some patients.
Side effects, or the fear of side effects, have generally been shown to reduce adherence across multiple chronic diseases.40 However, clinical studies in cancer patients provide conflicting evidence about the impact of side effects on adherence.1,44 Side effects may occur early, before treatment benefits become evident, and obviously may be a significant barrier to adherence for some cancer patients. For instance, oral capecitabine is associated with a high incidence of grade 1/2 diarrhea,45 which may be expected to adversely impact adherence, although no data are available to support this theory. Lastly, as previously noted, adherence tends to decrease during long-term treatment.25
Healthcare system-related factors
The patient-provider relationship, along with patient access to and patient satisfaction with medical care, can affect adherence.27,40 Positive, constructive relationships between patients and clinicians are key to minimizing treatment-related anxiety and may improve adherence.46 The high cost of medication may be a major barrier for patients who do not have medical insurance and also for those whose insurance does not provide coverage or requires high copayments for oral anticancer medications. 3,47
Overcoming barriers to adherence in patient management
Set realistic patient expectations on treatment outcomes and side effects
Patient education and a collaborative patient-physician relationship are key components for overcoming barriers to adherence. In cases in which patients cannot afford oral treatment, patient-assistance programs may be helpful. Some patients, however, may refuse educational or assistance programs because they do not want to appear as being needy or as having failed.
Patient-centered education about what patients can expect from treatment— both in terms of benefits and side effects, as well as the steps to take to manage key side effects should they occur—has been shown to improve adherence.46 Realistic expectations about treatment benefits should be offered, as patients may wrongfully conclude that their treatment is not working if their too-high expectations are not met, leading some to stop treatment. Hence, patients on maintenance therapy with targeted drugs, such as erlotinib in non-small cell lung cancer or ridaforolimus in the ongoing SUCCEED trial in bone or soft-tissue sarcomas, should be informed that treatment is designed to keep their disease stable and not necessarily to further shrink the size of their tumors.48
Oral targeted agents may offer an improved tolerability profile compared with cytotoxic chemotherapy but nevertheless may cause specific, dose-limiting side effects related to target inhibition in normal cells (Table 1). Key side effects associated with each prescribed agent, as well as prophylactic steps that can be taken to prevent or minimize potential side effects and instructions to follow should side effects occur, should be discussed with patients before treatment is started.
For example, combined data from nine studies show that patients receiving the standard dose of sorafenib run a significant risk of hypertension.49 Physicians should closely monitor these patients during treatment and prescribe appropriate antihypertensive agents as needed.
An acneiform rash is common with erlotinib and other EGFR-targeting drugs; it is amenable to topical steroid therapy in mild cases and tetracyclines in moderate cases.50 ‘‘Aphthous-like’’ oral lesions are common with mTOR inhibitors, such as temsirolimus (Torisel), everolimus, and ridaforolimus, but they differ from the classic mucositis encountered with cytotoxic chemotherapy.51 Oral ulcerations are often amenable to prophylactic strategies: practicing good oral hygiene with brushing and flossing after each meal, avoiding spicy and acidic foods, drinking warm rather than hot beverages, and cleansing the mouth with baking soda rinses.52
Nausea and vomiting are well recognized side effects associated with cytotoxic chemotherapy and may be problematic with some oral anticancer drugs (Figure 3). Oral agents with a moderate emesis risk (30%–90%) include cyclophosphamide, etoposide, imatinib, and temozolomide (Temodar), and those with a low risk (10%– 30%) include capecitabine, oral fludarabine (Oforta), and pazopanib. 9,53,54 Most orally active targeted agents, except for imatinib, have a very low risk of emesis (ie, < 10%); they include erlotinib, gefitinib, sorafenib, and sunitinib. 54
Several classes of antiemetic agents are available for preventing nausea and vomiting, including serotonin 5-HT3 (5-hydroxytryptamine) agonists (eg, ondansetron, granisetron, dolasetron [Anzemet]), dexamethasone, and the neurokinin-1 (NK1) receptor antagonist aprepitant (Emend).54,55 Each drug is available in an oral formulation, which provides comparable prophylaxis as the corresponding IV formulation.56 Guidelines for prophylactic use of these agents are available for IV chemotherapy drugs and regimens based on their emetic risk.56,57 However, prospective data on the use of these agents during daily dosing with oral anticancer drugs are limited, and therefore treatment is largely empirical. 5
Dose reduction may be an option to manage side effects effectively while allowing patients to remain on treatment. In some cases, side effects may be an indicator of the efficacy of treatment—as in the case of rash with EGFR inhibitors—and the patient should be made aware of this possibility. 58,59
Provide effective patient education
Education can be offered in many formats but should be tailored to patient preferences, whenever possible, with sufficient time made available to assess patient needs and for patients to express concerns and ask questions. It is helpful to include a family member or caregiver who can reinforce educational information at home and encourage the patient to maintain adherence. Patient-oriented written materials and reliable online information sources may be offered to reinforce and supplement learning points made by the healthcare team.
In addition to providing information, clinicians can help to improve patient adherence by offering encouragement and empathy during each interaction with their patients.22,60 Regular follow-up to emphasize the need for adherence, answer additional questions, and obtain feedback about treatment is also an integral part of an effective strategy for promoting adherence. Patients may forget much of what their physicians tell them, particularly when the prognosis is poor.61 This scenario underscores the importance of frequent interactions and written materials for patient reference at home. Follow-up can be provided by periodic telephone calls from the oncology nursing staff or by group educational sessions—the latter should be for patients who have expressed an interest in attending such sessions.25,62
Perform routine monitoring and documentation of adherence
From the provider’s perspective, improvements can be achieved by routine monitoring of patient adherence, as well as assessment and reinforcement of patient understanding of the treatment goals and the need to adhere to treatment recommendations. This process also includes documentation of the patient’s treatment history, whether in written files or electronic medical records. IV chemotherapy drugs are documented in the chemotherapy flow sheet, which is a best clinical practice in oncology. In contrast, oral anticancer agents may be entered into patient records in the same manner as oral drugs used for other medical conditions (eg, antihypertensive drugs). Instead, it is preferable to enter information about oral anticancer agents directly onto the chemotherapy flow sheet as a care plan, including dosage, schedule, and all dosage adjustments. This approach makes the oral agent prominent in the cancer plan and may help to improve adherence through better documentation and patient follow-up.
Use experimental approaches for monitoring/improving adherence
Conventional methods for monitoring adherence, such as patient selfreports, medication diaries, and pill counts, are not always reliable in clinical practice, as already discussed. As a result, a number of other methods are being evaluated to monitor and improve patient adherence.
The feasibility of an automated voice-response system (AVRS) coupled with nursing intervention was recently evaluated in patients with solid tumors who were receiving oral chemotherapy agents.63 Patients received weekly calls from the AVRS and answered questions about adherence as well as the severity of 15 symptoms. Patients reporting adherence below 100% or symptom severity of 4 or higher, on a 0 to 10 scale, for 3 consecutive weeks were called by a nurse for assistance with treatment adherence and symptomatology. In the study cohort, nonadherence to oral chemotherapy was 23.3% and was related to both symptoms and missed or forgotten medication. Notably, better symptom management— and not symptom severity per se—was associated with higher rates of adherence.
Other technologies that may enhance patient adherence to oral drug therapy include daily cell phone alarms, text messages, or smartphone reminder applications. Although the use of text-message reminders requiring a patient response has been shown to increase adherence in patients with human immunodeficiency virus receiving antiretroviral therapy,64 the use of interactive mobile web or smartphone applications to enhance patient adherence to antihypertensive medications is still being investigated. 65
Building on the MEMS strategy, GlowCaps (Vitality, Inc., Cambridge, MA) is a pill bottle cap designed to replace the conventional cap provided by retail pharmacies. The electronic cap flashes and plays a ringtone when it is time for the next dose, and its wireless transmitter sends a signal to a reminder light plug that also flashes. If the cap is not opened, the transmitter dials the patient’s telephone with an additional reminder. The cap also creates weekly adherence reports that can be e‑mailed to a friend or family member, as well as monthly adherence reports that can be mailed to the healthcare provider. The GlowCap should improve adherence for patients who forget to take their medication, but like MEMS, it only measures whether the bottle was opened and not whether the medication was actually taken. Conventional bottle caps are supplied at no additional charge; however, this device carries a retail price of $99, and therefore cost can be an issue for patients on a fixed budget and for those requiring multiple medications.
A miniscule edible “chip” (Proteus Biomedical Inc., Redwood City, CA) may go one step farther by confirming that medication has actually been ingested. The chip, which is a digestible sensor made from food ingredients, is activated by the low pH in the stomach to send a signal to a microelectronic receiver located in a bandagestyle skin patch. The receiver records medication-related information, including the date, time, and dose taken. A pilot study showed that this technology substantially improved adherence for an antihypertensive agent from 30% to 80% over a 6-month period. 66 The cost of the edible chip is only a few cents each when made in large quantities, suggesting that it may be economically viable. The involvement of pharmacists in monitoring prescription refills and in providing patient reminders is also being evaluated for effectiveness (Medco Health Solutions, Inc., Fairfield, OH).
Consider economic issues
Oncologists in the United States are reimbursed for administering IV chemotherapy agents. This reimbursement includes the cost of the medication and an additional small percentage above the acquisition cost (eg, 6% from Medicare). Profit from IV chemotherapy (ie, reimbursement exceeding acquisition costs) may contribute to the financial viability of many oncology practices.67 However, a similar financial incentive for use of oral anticancer agents is not provided by payors, and therefore the current system favors use of IV therapy in cancer management. To rectify this situation, an alternate method will need to be developed to provide comparable incentives for oral anticancer drugs, particularly for those agents that improve patient outcome or reduce healthcare costs.
The added cost of patient education and adherence monitoring— both in terms of oncology staff and time—is another financial issue to be addressed. These costs cannot be borne by oncology practices but will need to be covered by payors or patients. Episode-based or monthly management fees provided by payors, which encompass the coordination of oncology care, may represent viable options for covering patient education and adherence monitoring.67
Conclusion
Oral agents are increasingly used to treat many solid malignancies— both as primary treatment in advanced cancers and as maintenance therapy in patients after response to first-line chemotherapy. In these settings, long-term use of oral agents offers the promise of transforming cancer into a chronic disease. With these changes in treatment regimen and disease state, patient adherence becomes increasingly important. Poor adherence with tamoxifen has already been associated with poor outcomes in women with breast cancer.35 As other oral anticancer agents are used for longer periods, it is likely that additional associations between poor adherence and adverse outcomes will be shown.
Clinicians cannot simply depend on self-reports or pill counts to identify nonadherent patients. Instead, they need to adopt a proactive role, which includes assessing patient needs and understanding, educating patients before initiating treatment, key points at all subsequent visits, and using follow-up phone calls to identify issues that impact adherence. If barriers to adherence are identified, whether attributable to patient, treatment, or healthcare system-related factors, oncologists and their staffs have an opportunity to play an essential role in addressing and reducing, if not eliminating, those barriers. Overcoming adherence barriers should result in better clinical outcomes and improved quality of life for patients with sarcomas and other solid tumors.
Acknowledgments: The author would like to thank Brigitte Teissedre, PhD, and Joseph J. Abrajano, PhD, of Medicus International New York, for editorial assistance in the preparation of this manuscript. Editorial support was funded by Merck & Co., Inc. The author was fully responsible for all content and editorial decisions and received no financial support or other compensation related to the development of this article.
References1. Partridge AH, Avorn J, Wang PS, Winer EP. Adherence to therapy with oral antineoplastic agents. J Natl Cancer Inst 2002;94:652– 661.
2. Gelmon K, Chan A, Harbeck N. The role of capecitabine in first-line treatment for patients with metastatic breast cancer. Oncologist 2006;11(suppl 1):42–51.
3. Aisner J. Overview of the changing paradigm in cancer treatment: oral chemotherapy. Am J Health Syst Pharm 2007;64(9 suppl 5):S4–S7.
4. Ma WW, Adjei AA. Novel agents on the horizon for cancer therapy. CA Cancer J Clin 2009;59:111–137.
5. Blanke CD, Demetri GD, von Mehren M, et al. Long-term results from a randomized phase II trial of standard- versus higher- dose imatinib mesylate for patients with unresectable or metastatic gastrointestinal stromal tumors expressing KIT. J Clin Oncol 2008;26:620–625.
6. Demetri GD, van Oosterom AT, Garrett CR, et al. Efficacy and safety of sunitinib in patients with advanced gastrointestinal stromal tumour after failure of imatinib: a randomised controlled trial. Lancet 2006;368:1329–1338.
7. Shepherd FA, Rodrigues Pereira J, Ciuleanu T, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005;353:123–132.
8. Sanford M, Scott LJ. Gefitinib: a review of its use in the treatment of locally advanced/ metastatic non-small cell lung cancer. Drugs 2009;69:2303–2328.
9. Sternberg CN, Davis ID, Mardiak J, et al. Pazopanib in locally advanced or metastatic renal cell carcinoma: results of a randomized phase III trial. J Clin Oncol 2010;28:1061– 1068.
10. Escudier B, Eisen T, Stadler WM, et al. Sorafenib for treatment of renal cell carcinoma: final efficacy and safety results of the phase III treatment approaches in renal cancer global evaluation trial. J Clin Oncol 2009;27:3312– 3318.
11. Motzer RJ, Escudier B, Oudard S, et al. Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial. Lancet 2008;372:449–456.
12. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology™. Soft tissue sarcoma. V1.2010. Available at: www.nccn.org. Accessed April 2, 2010.
13. Dematteo RP, Ballman KV, Antonescu CR, et al. Adjuvant imatinib mesylate after resection of localised, primary gastrointestinal stromal tumour: a randomised, double-blind, placebo-controlled trial. Lancet 2009;373:1097–1104.
14. Ariad Pharmaceuticals. European Organization for Research and Treatment of Cancer. Ridaforolimus in treatment of sarcoma- SUCCEED (Sarcoma Multi-Center Clinical Eval. of the Efficacy of Ridaforolimus). In: ClinicalTrials.gov. Bethesda (MD): National Library of Medicine (US). http://clinicaltrials. gov/ct2/show/NCT00538239?term=0053823 9&rank=1. NLM Identifier: NCT00538239. Accessed February 5, 2010.
15. GlaxoSmithKline. A randomized double blind phase III trial of pazopanib versus placebo in patients with soft tissue sarcoma whose disease has progressed during or following prior therapy (PALETTE). In: Clinical- Trials.gov. Bethesda (MD): National Library of Medicine (US). http://clinicaltrials.gov/ct2/ show/NCT00753688?term=palette&rank=1. NLM Identifier: NCT00753688. Accessed February 11, 2010.
16. Liu G, Franssen E, Fitch MI, Warner E. Patient preferences for oral versus intravenous palliative chemotherapy. J Clin Oncol 1997;15:110–115.
17. Fallowfield L, Atkins L, Catt S, et al. Patients’ preference for administration of endocrine treatments by injection or tablets: results from a study of women with breast cancer. Ann Oncol 2006;17:205–210.
18. Twelves C, Gollins S, Grieve R, Samuel L. A randomised cross-over trial comparing patient preference for oral capecitabine and 5-fluorouracil/leucovorin regimens in patients with advanced colorectal cancer. Ann Oncol 2006;17:239–245.
19. Jensen LH, Osterlind K, Rytter C. Randomized cross-over study of patient preference for oral or intravenous vinorelbine in combination with carboplatin in the treatment of advanced NSCLC. Lung Cancer 2008;62:85–91.
20. Findlay M, von Minckwitz G, Wardley A. Effective oral chemotherapy for breast cancer: pillars of strength. Ann Oncol 2008;19:212–222.
21. Pelusi J. Capecitabine versus 5-FU in metastatic colorectal cancer: considerations for treatment decision-making. Commun Oncol 2006;3:19–27.
22. Pollak KI, Arnold RM, Jeffreys AS, et al. Oncologist communication about emotion during visits with patients with advanced cancer. J Clin Oncol 2007;25:5748–5752.
23. Palmieri FM, Barton DL. Challenges of oral medications in patients with advanced breast cancer. Semin Oncol Nurs 2007;23(4 suppl 2):S17–S22.
24. Ingersoll KS, Cohen J. The impact of medication regimen factors on adherence to chronic treatment: a review of literature. J Behav Med 2008;31:213–224.
25. O’Neill VJ, Twelves CJ. Oral cancer treatment: developments in chemotherapy and beyond. Br J Cancer 2002;87:933–937.
26. Cramer JA, Roy A, Burrell A, et al. Medication compliance and persistence: terminology and definitions. Value Health 2008;11:44–47.
27. Ruddy K, Mayer E, Partridge A. Patient adherence and persistence with oral anticancer treatment. CA Cancer J Clin 2009;59:56–66.
28. Waterhouse DM, Calzone KA, Mele C, Brenner DE. Adherence to oral tamoxifen: a comparison of patient self-report, pill counts, and microelectronic monitoring. J Clin Oncol 1993;11:1189–1197.
29. Ziller V, Kalder M, Albert US, et al. Adherence to adjuvant endocrine therapy in postmenopausal women with breast cancer. Ann Oncol 2009;20:431–436.
30. Hansen RA, Kim MM, Song L, Tu W, Wu J, Murray MD. Comparison of methods to assess medication adherence and classify nonadherence. Ann Pharmacother 2009;43:413–422.
31. Muss HB, Berry DA, Cirrincione CT, et al. Adjuvant chemotherapy in older women with early-stage breast cancer. N Engl J Med 2009;360:2055–2065.
32. Lebovits AH, Strain JJ, Schleifer SJ, Tanaka JS, Bhardwaj S, Messe MR. Patient noncompliance with self-administered chemotherapy. Cancer 1990;65:17–22.
33. Noens L, van Lierde MA, De Bock R, et al. Prevalence, determinants, and outcomes of nonadherence to imatinib therapy in patients with chronic myeloid leukemia: the ADAGIO study. Blood 2009;113:5401–5411.
34. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005;365:1687–1717.
35. McCowan C, Shearer J, Donnan PT, et al. Cohort study examining tamoxifen adherence and its relationship to mortality in women with breast cancer. Br J Cancer 2008;99:1763–1768.
36. Demetri GD, Wang Y, Wehrle E, et al. Imatinib plasma levels are correlated with clinical benefit in patients with unresectable/metastatic gastrointestinal stromal tumors. J Clin Oncol 2009;27:3141–3147.
37. Blay JY, Le Cesne A, Ray-Coquard I, et al. Prospective multicentric randomized phase III study of imatinib in patients with advanced gastrointestinal stromal tumors comparing interruption versus continuation of treatment beyond 1 year: the French Sarcoma Group. J Clin Oncol 2007;25:1107–1113.
38. O’Dwyer M, Atallah E. Practical considerations for the management of patients in the tyrosine kinase inhibitor era. Semin Hematol 2009;46(2 suppl 3):S16–S21.
39. Wu EQ, Johnson S, Beaulieu N, et al. Healthcare resource utilization and costs associated with non-adherence to imatinib treatment in chronic myeloid leukemia patients. Curr Med Res Opin 2010;26:61–69.
40. Krueger KP, Berger BA, Felkey B. Medication adherence and persistence: a comprehensive review. Adv Ther 2005;22:313–356.
41. Tebbi CK. Treatment compliance in childhood and adolescence. Cancer 1993;71(10 suppl):3441–3449.
42. MacLaughlin EJ, Raehl CL, Treadway AK, Sterling TL, Zoller DP, Bond CA. Assessing medication adherence in the elderly: which tools to use in clinical practice? Drugs Aging 2005;22:231–255.
43. Tam-McDevitt J. Polypharmacy, aging, and cancer. Oncology 2008;22:1052–1055; discussion 1055, 1058, 1060.
44. Richardson JL, Marks G, Levine A. The influence of symptoms of disease and side effects of treatment on compliance with cancer therapy. J Clin Oncol 1988;6:1746–1752.
45. Walko CM, Lindley C. Capecitabine: a review. Clin Ther 2005;27:23–44.
46. Fallowfield LJ. Treatment decision- making in breast cancer: the patient-doctor relationship. Breast Cancer Res Treat 2008;112(suppl 1):5–13.
47. Elliott R. Non-adherence to medicines: not solved but solvable. J Health Serv Res Policy 2009;14:58–61.
48. Mok TS, Ramalingam SS. Maintenance therapy in nonsmall-cell lung cancer: a new treatment paradigm. Cancer 2009;115:5143– 5154.
49. Wu S, Chen JJ, Kudelka A, Lu J, Zhu X. Incidence and risk of hypertension with sorafenib in patients with cancer: a systematic review and meta-analysis. Lancet Oncol 2008;9:117–123.
50. Segaert S, Van Cutsem E. Clinical signs, pathophysiology and management of skin toxicity during therapy with epidermal growth factor receptor inhibitors. Ann Oncol 2005;16:1425–1433.
51. Sonis S, Treister N, Chawla S, Demetri G, Haluska F. Preliminary characterization of oral lesions associated with inhibitors of mammalian target of rapamycin in cancer patients. Cancer 2010;116:210–215.
52. Scully C. Clinical practice: aphthous ulceration. N Engl J Med. 2006;355:165–172.
53. Hurwitz HI, Dowlati A, Saini S, et al. Phase I trial of pazopanib in patients with advanced cancer. Clin Cancer Res 2009;15:4220– 4227.
54. Jordan K, Sippel C, Schmoll HJ. Guidelines for antiemetic treatment of chemotherapy- induced nausea and vomiting: past, present, and future recommendations. Oncologist 2007;12:1143–1150.
55. Hesketh PJ. Chemotherapy-induced nausea and vomiting. N Engl J Med 2008;358: 2482–2494.
56. Kris MG, Hesketh PJ, Somerfield MR, et al. American Society of Clinical Oncology guideline for antiemetics in oncology: update 2006. J Clin Oncol 2006;24:2932–2947.
57. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology™. Antiemesis. V1. 2010. Available at: http://www.nccn.org. Accessed May 5, 2010.
58. Perez-Soler R. Rash as a surrogate marker for efficacy of epidermal growth factor receptor inhibitors in lung cancer. Clin Lung Cancer 2006;8(suppl 1):S7–S14.
59. Wacker B, Nagrani T, Weinberg J, Witt K, Clark G, Cagnoni PJ. Correlation between development of rash and efficacy in patients treated with the epidermal growth factor receptor tyrosine kinase inhibitor erlotinib in two large phase III studies. Clin Cancer Res 2007;13:3913–3921.
60. Morse DS, Edwardsen EA, Gordon HS. Missed opportunities for interval empathy in lung cancer communication. Arch Intern Med 2008;168:1853–1858.
61. Jansen J, van Weert J, van der Meulen N, van Dulmen S, Heeren T, Bensing J. Recall in older cancer patients: measuring memory for medical information. Gerontologist 2008;48:149–157.
62. Moore S. Facilitating oral chemotherapy treatment and compliance through patient/ family-focused education. Cancer Nurs 2007;30:112–122; quiz 123–124.
63. Decker V, Spoelstra S, Miezo E, et al. A pilot study of an automated voice response system and nursing intervention to monitor adherence to oral chemotherapy agents. Cancer Nurs 2009;32:E20–E29.
64. Lester RT, Ritvo P, Mills EJ, et al. Effects of a mobile phone short message service on antiretroviral treatment adherence in Kenya (WelTel Kenya1): a randomised trial. Lancet 2010;376:1838–1845.
65. Medication Adherence and mHealth: The George Washington University and Wireless Reach Pill Phone Study. Available at: http://multivu.prnewswire.com/mnr/qualcomm/ 48588/docs/48588-PillPhone_CS_FINAL_ r2.pdf. Accessed May 3, 2011.
66. Saenz A. Ingestible chip that reminds you to take medicine begins trial. Singularity Hub. Available at: http://singularityhub. com/2009/09/24/ingestible-chip-that-remindsyou- to-take-medicine-begins-trials-withnovartis/# more-7574. Accessed April 6, 2010.
67. Smith TJ, Hillner BE. Concrete options and ideas for increasing value in oncology care: the view from one trench. Oncologist 2010;15(suppl 1):65–72.
Arthur P. Staddon, MD
Joan Karnell Cancer Center, University of Pennsylvania School of Medicine, and Department of Hematology and Oncology, Pennsylvania Hospital, Philadelphia, PA
Manuscript received December 22, 2010; accepted June 10, 2011.
Correspondence to: Arthur P. Staddon, MD, Department of Hematology and Oncology, Pennsylvania Hospital of the University of Pennsylvania Health System, 230 West Washington Square, Philadelphia, PA 19107; e-mail: [email protected].
Oral anticancer drugs, particularly targeted therapies, are used increasingly to treat many solid malignancies. Adherence to the prescribed regimen is essential to ensuring that patients derive maximal clinical benefit from these
oral agents. However, multiple patient-related, treatment-related, and healthcare-associated factors may adversely impact adherence, thus compromising patient outcome. Reliable methods are not readily available in clinical practice to identify which patients are nonadherent. Therefore, clinicians need to take a proactive approach by assessing their patients’ needs, providing education about what can be expected during the course of oral therapy, monitoring adherence and reinforcing key points at all office visits, and using follow-up phone calls to identify issues that may still have an impact on adherence. By identifying and addressing barriers to adherence, oncologists can help their patients realize the full potential of oral therapy, including the promise of improved clinical outcome and quality of life.
Parenteral cytotoxic chemotherapy has traditionally been the major component of treatment for many solid malignancies—both in the adjuvant/neoadjuvant setting for high-risk localized disease and in the primary treatment for advanced and unresectable disease. Some oral anticancer agents, including tamoxifen, prednisone, and cyclophosphamide, have been used to manage certain malignancies for many years,1 and other oral cytotoxic drugs, such as capecitabine (Xeloda), have recently increased in usage.2,3. However, the advent of targeted therapy with agents that block specific cellular processes thought to be important in cell growth, survival, and metastasis has led to a large increase in the number of oral drugs for cancer.4 With the increased use of oral cancer drugs comes a shift in the treatment paradigm as it relates to patient management, particularly patient adherence to therapy. Examples of such oral targeted agents include the KIT/plateletderived growth factor receptor (PDGFR) tyrosine kinase inhibitors imatinib (Gleevec)5 and sunitinib (Sutent),6 the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors erlotinib (Tarceva) 7 and gefitinib (Iressa),8 the vascular endothelial growth factor receptor (VEGFR)/PDGFR kinase inhibitor pazopanib (Votrient),9 the multikinase inhibitor sorafenib (Nexavar),10 and the mamma lian target of rapamycin (mTOR) inhibitor everolimus (Afinitor).11
Notably, oral targeted therapy has emerged as a primary treatment of certain solid malignancies, including advanced gastrointestinal stromal tumors (GIST) and metastatic renal cell carcinoma.12 For example, imatinib is now recognized as first-line treatment of advanced GIST, and sunitinib is used in the second-line setting when patients relapse or cannot tolerate imatinib.12 Imatinib is also used in adjuvant therapy following complete surgical resection of localized GIST in patients with an intermediate or high risk of disease recurrence.12,13 Although no oral therapies are currently approved for treatment of bone or soft-tissue sarcomas, several are in clinical trials—the most advanced of which include two placebo-controlled, phase III trials evaluating the mTOR inhibitor ridaforolimus and the tyrosine kinase inhibitor pazopanib. Ridaforolimus is being evaluated as maintenance therapy for patients with favorable responses to chemotherapy, in the Sarcoma Multicenter Clinical Evalua tion of the Efficacy of Ridaforolimus (SUCCEED) trial.14 In the Pazopanib Explored in Soft-Tissue Sarcoma a Phase III (PALETTE) trial, pazopanib is being investigated in patients whose disease has progressed during or following prior therapy.15
Cancer patients generally prefer oral therapy over intravenous (IV) therapy, provided that efficacy is not compromised.1,16,17 Oral therapy offers greater convenience because it eliminates the time constraints imposed by receiving IV therapy at a physician’s office or treatment clinic. 18,19 This benefit may be particularly important for patients who live a significant distance from a treatment center.20 Oral therapy also reduces the discomfort and anxiety of having a venous catheter inserted for each treatment and it is easier in cases where venous access is difficult.20 Clinicians also are generally reported to prefer oral therapy, as it prevents complications such as infections and clotting associated with venous access and infusion pumps.2
The benefits of oral therapy must be balanced against several potential disadvantages, not the least of which is patient adherence to prescribed treatment. Patients receiving IV treatment may see their oncologist at intervals of 1 to 3 weeks, whereas those on oral agents may have less frequent visits. This arrangement limits educational and monitoring opportunities, as well as positive reinforcing interactions between patients and clinicians.22 The positive reinforcement and education provided by oncology nurses during IV treatment sessions are also lost.23 Perhaps most important, however, is that oral therapy requires patient self-medication and depends on the ability of patients to adhere to a prescribed regimen, which may involve daily treatment. More complex regimens may include multiple daily doses, drug holidays, or dosing based on the timing of meals.
Poor patient adherence is a well recognized issue in the treatment of many chronic diseases, including hypertension, diabetes, asthma, and mental illness.24 With oral therapies, long-term treatment is generally necessary to maintain disease control. However, adherence tends to decrease during long-term treatment, and therefore poor adherence may be more problematic during long-term maintenance with a targeted agent than with short-term use of an oral cytotoxic drug (Figure 1).25 The goal of maintaining high-level treatment adherence is to achieve optimal control of disease and to prevent, to the highest extent possible, disease progression.
Treatment adherence with oral anticancer agents
An overview
The terms “adherence” and “compliance” are used to describe the degree or extent to which a patient conforms to day-to-day treatment recommendations, including the timing, dosage, mode of administration, and frequency of the medication(s).26 Adherence is a synonym of compliance and is often perceived to be less judgmental of the patient’s behavior.27 Optimal adherence is achieved when a patient follows the treatment regimen exactly as prescribed (ie, without missing doses or taking extra doses). Therefore, adherence simply reflects the percentage of doses that are taken as prescribed.26
Monitoring treatment adherence in clinical practice is challenging, with all methods having significant limitations.1,27 Patient self-reporting may not accurately reflect treatment-taking behavior, because patients who want to please their clinicians will often overreport actual adherence. Patients completing medication diaries may erroneously fill in doses they failed to take, particularly at times close to scheduled visits. Additionally, these methods may provide unreliable information about the timing of doses, which is also important in ensuring appropriate systemic exposure to oral therapy. Other approaches have been used in clinical studies, including monitoring prescription refills and the microelectronic monitoring system (MEMS), which records each time the cap of the medication bottle is opened. However, opening the bottle does not ensure that the patient has taken the medication.
Therapeutic drug monitoring, measured through assessment of serum or urine drug levels, when available, may help to determine whether a patient has been adherent; however, it provides information for only the brief time leading up to testing and is limited by substantial interpatient variability in pharmacokinetics. Therapeutic drug monitoring also requires additional costs, further limiting its utility in clinical practice.
Adherence and the cancer patient
Cancer patients are generally thought to be highly motivated to follow their prescribed regimen.1,25 However, studies show that adherence with oral anticancer drugs is not optimal and that self-reporting and pill counts may not accurately reflect true patient behavior. Most information about medication adherence in patients with solid tumors comes from studies of tamoxifen in breast cancer.
In a study of a small cohort of breast cancer patients receiving tamoxifen for a mean of 3 months, the average adherence rates based on self-reports and pill counts were 98% (range, 91%–100%) and 92% (range, 74%–109%), respectively, with several patients taking more doses than prescribed. In comparison, MEMS showed an average rate of 69% (range, 33%–94%).28 Similar differences in adherence rates based on self-reports versus prescription refills have been seen with adjuvant tamoxifen and anastrozole. 29 Although self-reported adherence of 100% was claimed for both agents, only 80% and 69% of the women on tamoxifen and anastrozole, respectively, were still classified as adherent after controlling for prescription refills (P < 0.01 and P < 0.01 vs self-report).29 In general, electronic adherence measures, such as MEMS, show higher agreement with refilled pharmacy prescriptions than with patient self-reports.30
Adherence to oral cytotoxic drugs may also not be optimal in cancer patients. Adherence to oral capecitabine was assessed using MEMS in a cohort of 161 elderly women receiving adjuvant therapy for breast cancer. Twenty-four percent of the patients took fewer than 80% of the planned doses.31 Similarly, a nonadherence rate of 43% was reported for 51 breast cancer patients receiving oral cyclophosphamide, with higher nonadherence observed when women were treated in community practice settings than at academic centers.32 To date, assessments of treatment adherence to oral targeted therapies have shown a relatively high proportion of nonadherence in patients with hematologic malignancies (eg, chronic myelocytic leukemia).33 Adherence in patients with solid tumors has not been studied as extensively. Clinicians should suspect poor adherence when patients fail to achieve expected treatment responses within a certain time or when prescriptions are not filled as often as expected.
Impact of poor adherence
Patients with poor adherence may not receive the full benefit of treatment and may consequently experience poor clinical outcomes. Two sample populations illustrate this point—women with breast cancer and patients with advanced GIST. Adjuvant tamoxifen is well recognized to improve survival of women with hormone receptor-positive early- stage breast cancer.34 The impact of tamoxifen adherence on outcome was evaluated in a cohort of 1,633 women who received treatment for a median of 2.4 years.35 Median adherence was 93%; however, 315 women (19%) had adherence rates of less than 80%. In the multivariate analysis, a low adherence rate was independently associated with a higher risk of death. At the median study follow-up, the hazard ratio for mortality with adherence of less than 80% was 1.10 (95% confidence interval: 1.001–1.21; P = 0.046).35
Oral imatinib changed the treatment of advanced GIST, extending median survival to nearly 5 years.5 Poor adherence with daily imatinib leads to low serum drug levels, which in turn has been associated with poor clinical outcomes. Patients with serum imatinib levels less than 1,100 ng/mL—the lowest quartile in a recent analysis—had a median time to disease progression (TTP) of 11.3 months, which was significantly shorter than the TTP of greater than 30 months seen in the other three serum imatinib quartiles.36
Another randomized study compared planned interruption versus continuation of imatinib therapy in patients with advanced GIST.37 Most patients assigned to stop imatinib after achieving objective responses or stable disease on long-term therapy had rapid disease progression. Conversely, most patients who continued imatinib without interruption maintained the clinical benefit. Although treatment interruption was a planned event in this study, these results do suggest that poor adherence due to treatment interruptions may lead to disease progression. Patient adherence to imatinib in GIST has not been formally assessed in a clinical study. However, in patients with chronic myeloid leukemia, poor adherence to imatinib has been shown to have adverse consequences, resulting in suboptimal responses, disease relapse, and higher healthcare utilization and costs.33,38,39 Therefore, efforts to improve adherence may be expected to lead to better patient outcomes.
Barriers to adherence
Numerous factors have been identified as barriers to adherence in patients with chronic diseases.1,40 These factors can be grouped into three categories: patient-related factors, treatment-related factors, and healthcare system-related factors (Figure 2). When considered from a population perspective rather than from the perspective of an individual patient, poor adherence likely reflects a complex interplay among multiple factors.
Patient-related factors
These factors include sociodemographic, psychosocial, and employment status; comorbid conditions; polypharmacy; and social and family support characteristics; as well as the patient’s health beliefs, self-efficacy, and health literacy. Age may be a factor associated with poor adherence, which is particularly problematic among adolescents.41 Older individuals may also be prone to adherence issues for a variety of reasons, including the need to take multiple medications for comorbid conditions, visual and cognitive deficits, lack of social support if living alone, medication cost if living on a fixed budget, and higher risk of side effects due to drug-drug interactions and altered drug pharmacokinetics.42,43 Other demographic factors, such as race, educational level, and socioeconomic status, may indirectly affect adherence due to their impact on access to healthcare.1
Patient expectations are also key factors influencing adherence. On the one hand, patients who are not convinced about the importance of their therapy or who believe that their fate is governed largely by chance are more likely to exhibit poor adherence. 1,40 On the other hand, patients who believe that medication will be effective and that their own actions can influence the course of disease are more likely to adhere to treatment, even when faced with side effects. With these expectations in mind, clinicians can provide relevant information and encouragement and address their patients’ concerns.
Treatment-related factors
These factors include the complexity of the regimen; pill burden; duration of treatment; timing of drug administration; and type, frequency, and severity of side effects. Adherence is negatively affected by more complex regimens, such as those with multiple drugs and those that are inconvenient. 40 Medications that must be taken with respect to meals (eg, with meals or several hours before or after meals) may limit adherence in some patients.
Side effects, or the fear of side effects, have generally been shown to reduce adherence across multiple chronic diseases.40 However, clinical studies in cancer patients provide conflicting evidence about the impact of side effects on adherence.1,44 Side effects may occur early, before treatment benefits become evident, and obviously may be a significant barrier to adherence for some cancer patients. For instance, oral capecitabine is associated with a high incidence of grade 1/2 diarrhea,45 which may be expected to adversely impact adherence, although no data are available to support this theory. Lastly, as previously noted, adherence tends to decrease during long-term treatment.25
Healthcare system-related factors
The patient-provider relationship, along with patient access to and patient satisfaction with medical care, can affect adherence.27,40 Positive, constructive relationships between patients and clinicians are key to minimizing treatment-related anxiety and may improve adherence.46 The high cost of medication may be a major barrier for patients who do not have medical insurance and also for those whose insurance does not provide coverage or requires high copayments for oral anticancer medications. 3,47
Overcoming barriers to adherence in patient management
Set realistic patient expectations on treatment outcomes and side effects
Patient education and a collaborative patient-physician relationship are key components for overcoming barriers to adherence. In cases in which patients cannot afford oral treatment, patient-assistance programs may be helpful. Some patients, however, may refuse educational or assistance programs because they do not want to appear as being needy or as having failed.
Patient-centered education about what patients can expect from treatment— both in terms of benefits and side effects, as well as the steps to take to manage key side effects should they occur—has been shown to improve adherence.46 Realistic expectations about treatment benefits should be offered, as patients may wrongfully conclude that their treatment is not working if their too-high expectations are not met, leading some to stop treatment. Hence, patients on maintenance therapy with targeted drugs, such as erlotinib in non-small cell lung cancer or ridaforolimus in the ongoing SUCCEED trial in bone or soft-tissue sarcomas, should be informed that treatment is designed to keep their disease stable and not necessarily to further shrink the size of their tumors.48
Oral targeted agents may offer an improved tolerability profile compared with cytotoxic chemotherapy but nevertheless may cause specific, dose-limiting side effects related to target inhibition in normal cells (Table 1). Key side effects associated with each prescribed agent, as well as prophylactic steps that can be taken to prevent or minimize potential side effects and instructions to follow should side effects occur, should be discussed with patients before treatment is started.
For example, combined data from nine studies show that patients receiving the standard dose of sorafenib run a significant risk of hypertension.49 Physicians should closely monitor these patients during treatment and prescribe appropriate antihypertensive agents as needed.
An acneiform rash is common with erlotinib and other EGFR-targeting drugs; it is amenable to topical steroid therapy in mild cases and tetracyclines in moderate cases.50 ‘‘Aphthous-like’’ oral lesions are common with mTOR inhibitors, such as temsirolimus (Torisel), everolimus, and ridaforolimus, but they differ from the classic mucositis encountered with cytotoxic chemotherapy.51 Oral ulcerations are often amenable to prophylactic strategies: practicing good oral hygiene with brushing and flossing after each meal, avoiding spicy and acidic foods, drinking warm rather than hot beverages, and cleansing the mouth with baking soda rinses.52
Nausea and vomiting are well recognized side effects associated with cytotoxic chemotherapy and may be problematic with some oral anticancer drugs (Figure 3). Oral agents with a moderate emesis risk (30%–90%) include cyclophosphamide, etoposide, imatinib, and temozolomide (Temodar), and those with a low risk (10%– 30%) include capecitabine, oral fludarabine (Oforta), and pazopanib. 9,53,54 Most orally active targeted agents, except for imatinib, have a very low risk of emesis (ie, < 10%); they include erlotinib, gefitinib, sorafenib, and sunitinib. 54
Several classes of antiemetic agents are available for preventing nausea and vomiting, including serotonin 5-HT3 (5-hydroxytryptamine) agonists (eg, ondansetron, granisetron, dolasetron [Anzemet]), dexamethasone, and the neurokinin-1 (NK1) receptor antagonist aprepitant (Emend).54,55 Each drug is available in an oral formulation, which provides comparable prophylaxis as the corresponding IV formulation.56 Guidelines for prophylactic use of these agents are available for IV chemotherapy drugs and regimens based on their emetic risk.56,57 However, prospective data on the use of these agents during daily dosing with oral anticancer drugs are limited, and therefore treatment is largely empirical. 5
Dose reduction may be an option to manage side effects effectively while allowing patients to remain on treatment. In some cases, side effects may be an indicator of the efficacy of treatment—as in the case of rash with EGFR inhibitors—and the patient should be made aware of this possibility. 58,59
Provide effective patient education
Education can be offered in many formats but should be tailored to patient preferences, whenever possible, with sufficient time made available to assess patient needs and for patients to express concerns and ask questions. It is helpful to include a family member or caregiver who can reinforce educational information at home and encourage the patient to maintain adherence. Patient-oriented written materials and reliable online information sources may be offered to reinforce and supplement learning points made by the healthcare team.
In addition to providing information, clinicians can help to improve patient adherence by offering encouragement and empathy during each interaction with their patients.22,60 Regular follow-up to emphasize the need for adherence, answer additional questions, and obtain feedback about treatment is also an integral part of an effective strategy for promoting adherence. Patients may forget much of what their physicians tell them, particularly when the prognosis is poor.61 This scenario underscores the importance of frequent interactions and written materials for patient reference at home. Follow-up can be provided by periodic telephone calls from the oncology nursing staff or by group educational sessions—the latter should be for patients who have expressed an interest in attending such sessions.25,62
Perform routine monitoring and documentation of adherence
From the provider’s perspective, improvements can be achieved by routine monitoring of patient adherence, as well as assessment and reinforcement of patient understanding of the treatment goals and the need to adhere to treatment recommendations. This process also includes documentation of the patient’s treatment history, whether in written files or electronic medical records. IV chemotherapy drugs are documented in the chemotherapy flow sheet, which is a best clinical practice in oncology. In contrast, oral anticancer agents may be entered into patient records in the same manner as oral drugs used for other medical conditions (eg, antihypertensive drugs). Instead, it is preferable to enter information about oral anticancer agents directly onto the chemotherapy flow sheet as a care plan, including dosage, schedule, and all dosage adjustments. This approach makes the oral agent prominent in the cancer plan and may help to improve adherence through better documentation and patient follow-up.
Use experimental approaches for monitoring/improving adherence
Conventional methods for monitoring adherence, such as patient selfreports, medication diaries, and pill counts, are not always reliable in clinical practice, as already discussed. As a result, a number of other methods are being evaluated to monitor and improve patient adherence.
The feasibility of an automated voice-response system (AVRS) coupled with nursing intervention was recently evaluated in patients with solid tumors who were receiving oral chemotherapy agents.63 Patients received weekly calls from the AVRS and answered questions about adherence as well as the severity of 15 symptoms. Patients reporting adherence below 100% or symptom severity of 4 or higher, on a 0 to 10 scale, for 3 consecutive weeks were called by a nurse for assistance with treatment adherence and symptomatology. In the study cohort, nonadherence to oral chemotherapy was 23.3% and was related to both symptoms and missed or forgotten medication. Notably, better symptom management— and not symptom severity per se—was associated with higher rates of adherence.
Other technologies that may enhance patient adherence to oral drug therapy include daily cell phone alarms, text messages, or smartphone reminder applications. Although the use of text-message reminders requiring a patient response has been shown to increase adherence in patients with human immunodeficiency virus receiving antiretroviral therapy,64 the use of interactive mobile web or smartphone applications to enhance patient adherence to antihypertensive medications is still being investigated. 65
Building on the MEMS strategy, GlowCaps (Vitality, Inc., Cambridge, MA) is a pill bottle cap designed to replace the conventional cap provided by retail pharmacies. The electronic cap flashes and plays a ringtone when it is time for the next dose, and its wireless transmitter sends a signal to a reminder light plug that also flashes. If the cap is not opened, the transmitter dials the patient’s telephone with an additional reminder. The cap also creates weekly adherence reports that can be e‑mailed to a friend or family member, as well as monthly adherence reports that can be mailed to the healthcare provider. The GlowCap should improve adherence for patients who forget to take their medication, but like MEMS, it only measures whether the bottle was opened and not whether the medication was actually taken. Conventional bottle caps are supplied at no additional charge; however, this device carries a retail price of $99, and therefore cost can be an issue for patients on a fixed budget and for those requiring multiple medications.
A miniscule edible “chip” (Proteus Biomedical Inc., Redwood City, CA) may go one step farther by confirming that medication has actually been ingested. The chip, which is a digestible sensor made from food ingredients, is activated by the low pH in the stomach to send a signal to a microelectronic receiver located in a bandagestyle skin patch. The receiver records medication-related information, including the date, time, and dose taken. A pilot study showed that this technology substantially improved adherence for an antihypertensive agent from 30% to 80% over a 6-month period. 66 The cost of the edible chip is only a few cents each when made in large quantities, suggesting that it may be economically viable. The involvement of pharmacists in monitoring prescription refills and in providing patient reminders is also being evaluated for effectiveness (Medco Health Solutions, Inc., Fairfield, OH).
Consider economic issues
Oncologists in the United States are reimbursed for administering IV chemotherapy agents. This reimbursement includes the cost of the medication and an additional small percentage above the acquisition cost (eg, 6% from Medicare). Profit from IV chemotherapy (ie, reimbursement exceeding acquisition costs) may contribute to the financial viability of many oncology practices.67 However, a similar financial incentive for use of oral anticancer agents is not provided by payors, and therefore the current system favors use of IV therapy in cancer management. To rectify this situation, an alternate method will need to be developed to provide comparable incentives for oral anticancer drugs, particularly for those agents that improve patient outcome or reduce healthcare costs.
The added cost of patient education and adherence monitoring— both in terms of oncology staff and time—is another financial issue to be addressed. These costs cannot be borne by oncology practices but will need to be covered by payors or patients. Episode-based or monthly management fees provided by payors, which encompass the coordination of oncology care, may represent viable options for covering patient education and adherence monitoring.67
Conclusion
Oral agents are increasingly used to treat many solid malignancies— both as primary treatment in advanced cancers and as maintenance therapy in patients after response to first-line chemotherapy. In these settings, long-term use of oral agents offers the promise of transforming cancer into a chronic disease. With these changes in treatment regimen and disease state, patient adherence becomes increasingly important. Poor adherence with tamoxifen has already been associated with poor outcomes in women with breast cancer.35 As other oral anticancer agents are used for longer periods, it is likely that additional associations between poor adherence and adverse outcomes will be shown.
Clinicians cannot simply depend on self-reports or pill counts to identify nonadherent patients. Instead, they need to adopt a proactive role, which includes assessing patient needs and understanding, educating patients before initiating treatment, key points at all subsequent visits, and using follow-up phone calls to identify issues that impact adherence. If barriers to adherence are identified, whether attributable to patient, treatment, or healthcare system-related factors, oncologists and their staffs have an opportunity to play an essential role in addressing and reducing, if not eliminating, those barriers. Overcoming adherence barriers should result in better clinical outcomes and improved quality of life for patients with sarcomas and other solid tumors.
Acknowledgments: The author would like to thank Brigitte Teissedre, PhD, and Joseph J. Abrajano, PhD, of Medicus International New York, for editorial assistance in the preparation of this manuscript. Editorial support was funded by Merck & Co., Inc. The author was fully responsible for all content and editorial decisions and received no financial support or other compensation related to the development of this article.
References1. Partridge AH, Avorn J, Wang PS, Winer EP. Adherence to therapy with oral antineoplastic agents. J Natl Cancer Inst 2002;94:652– 661.
2. Gelmon K, Chan A, Harbeck N. The role of capecitabine in first-line treatment for patients with metastatic breast cancer. Oncologist 2006;11(suppl 1):42–51.
3. Aisner J. Overview of the changing paradigm in cancer treatment: oral chemotherapy. Am J Health Syst Pharm 2007;64(9 suppl 5):S4–S7.
4. Ma WW, Adjei AA. Novel agents on the horizon for cancer therapy. CA Cancer J Clin 2009;59:111–137.
5. Blanke CD, Demetri GD, von Mehren M, et al. Long-term results from a randomized phase II trial of standard- versus higher- dose imatinib mesylate for patients with unresectable or metastatic gastrointestinal stromal tumors expressing KIT. J Clin Oncol 2008;26:620–625.
6. Demetri GD, van Oosterom AT, Garrett CR, et al. Efficacy and safety of sunitinib in patients with advanced gastrointestinal stromal tumour after failure of imatinib: a randomised controlled trial. Lancet 2006;368:1329–1338.
7. Shepherd FA, Rodrigues Pereira J, Ciuleanu T, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005;353:123–132.
8. Sanford M, Scott LJ. Gefitinib: a review of its use in the treatment of locally advanced/ metastatic non-small cell lung cancer. Drugs 2009;69:2303–2328.
9. Sternberg CN, Davis ID, Mardiak J, et al. Pazopanib in locally advanced or metastatic renal cell carcinoma: results of a randomized phase III trial. J Clin Oncol 2010;28:1061– 1068.
10. Escudier B, Eisen T, Stadler WM, et al. Sorafenib for treatment of renal cell carcinoma: final efficacy and safety results of the phase III treatment approaches in renal cancer global evaluation trial. J Clin Oncol 2009;27:3312– 3318.
11. Motzer RJ, Escudier B, Oudard S, et al. Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial. Lancet 2008;372:449–456.
12. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology™. Soft tissue sarcoma. V1.2010. Available at: www.nccn.org. Accessed April 2, 2010.
13. Dematteo RP, Ballman KV, Antonescu CR, et al. Adjuvant imatinib mesylate after resection of localised, primary gastrointestinal stromal tumour: a randomised, double-blind, placebo-controlled trial. Lancet 2009;373:1097–1104.
14. Ariad Pharmaceuticals. European Organization for Research and Treatment of Cancer. Ridaforolimus in treatment of sarcoma- SUCCEED (Sarcoma Multi-Center Clinical Eval. of the Efficacy of Ridaforolimus). In: ClinicalTrials.gov. Bethesda (MD): National Library of Medicine (US). http://clinicaltrials. gov/ct2/show/NCT00538239?term=0053823 9&rank=1. NLM Identifier: NCT00538239. Accessed February 5, 2010.
15. GlaxoSmithKline. A randomized double blind phase III trial of pazopanib versus placebo in patients with soft tissue sarcoma whose disease has progressed during or following prior therapy (PALETTE). In: Clinical- Trials.gov. Bethesda (MD): National Library of Medicine (US). http://clinicaltrials.gov/ct2/ show/NCT00753688?term=palette&rank=1. NLM Identifier: NCT00753688. Accessed February 11, 2010.
16. Liu G, Franssen E, Fitch MI, Warner E. Patient preferences for oral versus intravenous palliative chemotherapy. J Clin Oncol 1997;15:110–115.
17. Fallowfield L, Atkins L, Catt S, et al. Patients’ preference for administration of endocrine treatments by injection or tablets: results from a study of women with breast cancer. Ann Oncol 2006;17:205–210.
18. Twelves C, Gollins S, Grieve R, Samuel L. A randomised cross-over trial comparing patient preference for oral capecitabine and 5-fluorouracil/leucovorin regimens in patients with advanced colorectal cancer. Ann Oncol 2006;17:239–245.
19. Jensen LH, Osterlind K, Rytter C. Randomized cross-over study of patient preference for oral or intravenous vinorelbine in combination with carboplatin in the treatment of advanced NSCLC. Lung Cancer 2008;62:85–91.
20. Findlay M, von Minckwitz G, Wardley A. Effective oral chemotherapy for breast cancer: pillars of strength. Ann Oncol 2008;19:212–222.
21. Pelusi J. Capecitabine versus 5-FU in metastatic colorectal cancer: considerations for treatment decision-making. Commun Oncol 2006;3:19–27.
22. Pollak KI, Arnold RM, Jeffreys AS, et al. Oncologist communication about emotion during visits with patients with advanced cancer. J Clin Oncol 2007;25:5748–5752.
23. Palmieri FM, Barton DL. Challenges of oral medications in patients with advanced breast cancer. Semin Oncol Nurs 2007;23(4 suppl 2):S17–S22.
24. Ingersoll KS, Cohen J. The impact of medication regimen factors on adherence to chronic treatment: a review of literature. J Behav Med 2008;31:213–224.
25. O’Neill VJ, Twelves CJ. Oral cancer treatment: developments in chemotherapy and beyond. Br J Cancer 2002;87:933–937.
26. Cramer JA, Roy A, Burrell A, et al. Medication compliance and persistence: terminology and definitions. Value Health 2008;11:44–47.
27. Ruddy K, Mayer E, Partridge A. Patient adherence and persistence with oral anticancer treatment. CA Cancer J Clin 2009;59:56–66.
28. Waterhouse DM, Calzone KA, Mele C, Brenner DE. Adherence to oral tamoxifen: a comparison of patient self-report, pill counts, and microelectronic monitoring. J Clin Oncol 1993;11:1189–1197.
29. Ziller V, Kalder M, Albert US, et al. Adherence to adjuvant endocrine therapy in postmenopausal women with breast cancer. Ann Oncol 2009;20:431–436.
30. Hansen RA, Kim MM, Song L, Tu W, Wu J, Murray MD. Comparison of methods to assess medication adherence and classify nonadherence. Ann Pharmacother 2009;43:413–422.
31. Muss HB, Berry DA, Cirrincione CT, et al. Adjuvant chemotherapy in older women with early-stage breast cancer. N Engl J Med 2009;360:2055–2065.
32. Lebovits AH, Strain JJ, Schleifer SJ, Tanaka JS, Bhardwaj S, Messe MR. Patient noncompliance with self-administered chemotherapy. Cancer 1990;65:17–22.
33. Noens L, van Lierde MA, De Bock R, et al. Prevalence, determinants, and outcomes of nonadherence to imatinib therapy in patients with chronic myeloid leukemia: the ADAGIO study. Blood 2009;113:5401–5411.
34. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005;365:1687–1717.
35. McCowan C, Shearer J, Donnan PT, et al. Cohort study examining tamoxifen adherence and its relationship to mortality in women with breast cancer. Br J Cancer 2008;99:1763–1768.
36. Demetri GD, Wang Y, Wehrle E, et al. Imatinib plasma levels are correlated with clinical benefit in patients with unresectable/metastatic gastrointestinal stromal tumors. J Clin Oncol 2009;27:3141–3147.
37. Blay JY, Le Cesne A, Ray-Coquard I, et al. Prospective multicentric randomized phase III study of imatinib in patients with advanced gastrointestinal stromal tumors comparing interruption versus continuation of treatment beyond 1 year: the French Sarcoma Group. J Clin Oncol 2007;25:1107–1113.
38. O’Dwyer M, Atallah E. Practical considerations for the management of patients in the tyrosine kinase inhibitor era. Semin Hematol 2009;46(2 suppl 3):S16–S21.
39. Wu EQ, Johnson S, Beaulieu N, et al. Healthcare resource utilization and costs associated with non-adherence to imatinib treatment in chronic myeloid leukemia patients. Curr Med Res Opin 2010;26:61–69.
40. Krueger KP, Berger BA, Felkey B. Medication adherence and persistence: a comprehensive review. Adv Ther 2005;22:313–356.
41. Tebbi CK. Treatment compliance in childhood and adolescence. Cancer 1993;71(10 suppl):3441–3449.
42. MacLaughlin EJ, Raehl CL, Treadway AK, Sterling TL, Zoller DP, Bond CA. Assessing medication adherence in the elderly: which tools to use in clinical practice? Drugs Aging 2005;22:231–255.
43. Tam-McDevitt J. Polypharmacy, aging, and cancer. Oncology 2008;22:1052–1055; discussion 1055, 1058, 1060.
44. Richardson JL, Marks G, Levine A. The influence of symptoms of disease and side effects of treatment on compliance with cancer therapy. J Clin Oncol 1988;6:1746–1752.
45. Walko CM, Lindley C. Capecitabine: a review. Clin Ther 2005;27:23–44.
46. Fallowfield LJ. Treatment decision- making in breast cancer: the patient-doctor relationship. Breast Cancer Res Treat 2008;112(suppl 1):5–13.
47. Elliott R. Non-adherence to medicines: not solved but solvable. J Health Serv Res Policy 2009;14:58–61.
48. Mok TS, Ramalingam SS. Maintenance therapy in nonsmall-cell lung cancer: a new treatment paradigm. Cancer 2009;115:5143– 5154.
49. Wu S, Chen JJ, Kudelka A, Lu J, Zhu X. Incidence and risk of hypertension with sorafenib in patients with cancer: a systematic review and meta-analysis. Lancet Oncol 2008;9:117–123.
50. Segaert S, Van Cutsem E. Clinical signs, pathophysiology and management of skin toxicity during therapy with epidermal growth factor receptor inhibitors. Ann Oncol 2005;16:1425–1433.
51. Sonis S, Treister N, Chawla S, Demetri G, Haluska F. Preliminary characterization of oral lesions associated with inhibitors of mammalian target of rapamycin in cancer patients. Cancer 2010;116:210–215.
52. Scully C. Clinical practice: aphthous ulceration. N Engl J Med. 2006;355:165–172.
53. Hurwitz HI, Dowlati A, Saini S, et al. Phase I trial of pazopanib in patients with advanced cancer. Clin Cancer Res 2009;15:4220– 4227.
54. Jordan K, Sippel C, Schmoll HJ. Guidelines for antiemetic treatment of chemotherapy- induced nausea and vomiting: past, present, and future recommendations. Oncologist 2007;12:1143–1150.
55. Hesketh PJ. Chemotherapy-induced nausea and vomiting. N Engl J Med 2008;358: 2482–2494.
56. Kris MG, Hesketh PJ, Somerfield MR, et al. American Society of Clinical Oncology guideline for antiemetics in oncology: update 2006. J Clin Oncol 2006;24:2932–2947.
57. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology™. Antiemesis. V1. 2010. Available at: http://www.nccn.org. Accessed May 5, 2010.
58. Perez-Soler R. Rash as a surrogate marker for efficacy of epidermal growth factor receptor inhibitors in lung cancer. Clin Lung Cancer 2006;8(suppl 1):S7–S14.
59. Wacker B, Nagrani T, Weinberg J, Witt K, Clark G, Cagnoni PJ. Correlation between development of rash and efficacy in patients treated with the epidermal growth factor receptor tyrosine kinase inhibitor erlotinib in two large phase III studies. Clin Cancer Res 2007;13:3913–3921.
60. Morse DS, Edwardsen EA, Gordon HS. Missed opportunities for interval empathy in lung cancer communication. Arch Intern Med 2008;168:1853–1858.
61. Jansen J, van Weert J, van der Meulen N, van Dulmen S, Heeren T, Bensing J. Recall in older cancer patients: measuring memory for medical information. Gerontologist 2008;48:149–157.
62. Moore S. Facilitating oral chemotherapy treatment and compliance through patient/ family-focused education. Cancer Nurs 2007;30:112–122; quiz 123–124.
63. Decker V, Spoelstra S, Miezo E, et al. A pilot study of an automated voice response system and nursing intervention to monitor adherence to oral chemotherapy agents. Cancer Nurs 2009;32:E20–E29.
64. Lester RT, Ritvo P, Mills EJ, et al. Effects of a mobile phone short message service on antiretroviral treatment adherence in Kenya (WelTel Kenya1): a randomised trial. Lancet 2010;376:1838–1845.
65. Medication Adherence and mHealth: The George Washington University and Wireless Reach Pill Phone Study. Available at: http://multivu.prnewswire.com/mnr/qualcomm/ 48588/docs/48588-PillPhone_CS_FINAL_ r2.pdf. Accessed May 3, 2011.
66. Saenz A. Ingestible chip that reminds you to take medicine begins trial. Singularity Hub. Available at: http://singularityhub. com/2009/09/24/ingestible-chip-that-remindsyou- to-take-medicine-begins-trials-withnovartis/# more-7574. Accessed April 6, 2010.
67. Smith TJ, Hillner BE. Concrete options and ideas for increasing value in oncology care: the view from one trench. Oncologist 2010;15(suppl 1):65–72.
Video Report: What's the Future of Prostate Cancer Care?
Dr. Walter Stadler of the University of Chicago speaks about the changing landscape of prostate cancer care for physicians and patients. Newer modalities mean growing complexities in treating prostate disease.
Dr. Walter Stadler of the University of Chicago speaks about the changing landscape of prostate cancer care for physicians and patients. Newer modalities mean growing complexities in treating prostate disease.
Dr. Walter Stadler of the University of Chicago speaks about the changing landscape of prostate cancer care for physicians and patients. Newer modalities mean growing complexities in treating prostate disease.
New Payment Systems in Oncology: Aligning Incentives for Value and Accountability
Community Oncology Editor Dr. Linda Bosserman explains how new payment systems are affecting oncology practices.
Linda D. Bosserman, M.D., is with the Wilshire Oncology Medical Group in La Verne, CA.
Download the New Payment Systems Powerpoint Slideshow 
Community Oncology Editor Dr. Linda Bosserman explains how new payment systems are affecting oncology practices.
Linda D. Bosserman, M.D., is with the Wilshire Oncology Medical Group in La Verne, CA.
Download the New Payment Systems Powerpoint Slideshow
Community Oncology Editor Dr. Linda Bosserman explains how new payment systems are affecting oncology practices.
Linda D. Bosserman, M.D., is with the Wilshire Oncology Medical Group in La Verne, CA.
Download the New Payment Systems Powerpoint Slideshow
The House hears SGR alternatives, vows action
A plan to finally replace Medicare’s much-maligned sustainable growth rate (SGR) payment formula could be unveiled by this summer, federal lawmakers said at a committee hearing. “Here’s the bottom line: If we get to December and we’re doing an extension, that’s a failure on our part,” Rep. Michael Burgess (R– Tex) said at the hearing. “We need a permanent solution that’s predictable, updatable, and reasonable for this year—and nothing else will do.”
“Whatever virtues the SGR had when it was created 14 years ago…, it’s clear that they have vanished,” added Rep. Henry A. Waxman (D– Calif). He noted that in the past 2 years, Congress has had to pass legislation six times, blocking fee cuts of up to 21% or more.
About 30 medical associations, including the American Society of Clinical Oncology (ASCO), responded to the House subcommittee’s request for suggestions and proposals in developing a new system. On May 5, 2011, House subcommittee members met with a five-person panel of experts from medical associations and health policy organizations to consider alternatives to the current SGR formula, which some participants labeled as anything but sustainable.
One size won’t fit all
Although the details of ASCO’s plan and others vary, they also show a consensus on several fronts: repealing the SGR, moving away from the traditional fee-for-services payment model, and providing a 4- to 5-year transition period during which providers can experiment with a variety of payment systems.
In a letter accompanying the ASCO recommendations, the president, Dr. George Sledge, and CEO, Dr. Allen Lichter, stressed that SGR reforms in general should be linked to existing “robust” systems that promote evidence-based medicine. For oncology in particular, that effort should leverage the Quality Oncology Practice Initiative (QOPI), a comprehensive, field-tested program that more than one-quarter of outpatient oncology practices in the United States already participate in. More than 80% of oncology care is provided in that setting.
“The current SGR system has created an uncertain and unstable environment— a situation that threatens the viability of practices and access to care for thousands of cancer patients,” they concluded.
In its recommendations, ASCO asserted that evidence-based medicine is “both warranted and necessary” because:
- Medicare beneficiaries account for more than half of all new cancer diagnoses in the United States, and treatment and prevention of the disease comprise almost 10% of costs under fee-for-service Medicare;
- The care is complex, treatment can span many specialties, and treatment strategies change rapidly to keep pace with scientific advances; and
- These complexities would not be adequately addressed if a multispecialty system (such as the Physician Quality Reporting System) were to be applied in the oncology setting.
The recommendations also detailed why the QOPI should be incorporated as the primary quality measurement program: 25%−30% of a range of practices—urban, rural, community, and academic—participate in it; it is free; some private insurers have adopted incentives for participation in the program; the performance measures are field-tested and up-to-date; and participation promotes high-quality, high-value care and can help identify and address discrepancies in oncology care.
Moreover, QOPI “protects the best interests of patients, reduces exposure to unnecessary treatments and tests, minimizes the use of suboptimal treatment options, promotes the coordination of care, and protects the Medicare program from costs associated with poor-quality care,” ASCO asserted in the recommendation.
Members of the expert panel also stressed the importance of avoiding a “one-size-fits-all” solution. “We should [be mindful] that what will work in one part of the country will not work in another part of the country, and that’s why we have continued to talk about a variety of options,” said Dr. Cecil Wilson, president of the American Medical Association (AMA). “There is a temptation to feel that we ought to figure out one rule …that solves it all.”
Dr. Wilson pointed to the provisions in the Affordable Care Act that allow for a variety of models of accountable care organizations, embodying the concept of options in the medical system. In that spirit, he said that the AMA has formed a physician leadership group to evaluate the effectiveness of alternative payment methods.
Dr. Roland A. Goertz, president of the American Academy of Family Physicians (AAFP), noted in written testimony to the committee that “the evidence shows that to achieve the savings Congress is looking for, and to improve the quality of health care delivered to millions of patients in the country, reform must include investment in primary care.”
To strengthen primary care’s role in Medicare, the AAFP backs payment reforms that would boost primary care reimbursement and support the concept of the patient-centered medical home (PCMH). The AAFP’s proposal would create a blended reimbursement system for primary care delivered within a PCMH: fee-forservice payments and pay for performance, plus care management fees for PCMH-related activities that do not involve direct patient care.
To prepare for that new payment system, the AAFP has proposed a 5-year transition period with mandated pay increases for primary care physicians, an increase in the Primary Care Incentive Care payment from 10% to 20%, and a rule that Medicaid payments to primary care physicians will always be at least equal to Medicare payments.
Dr. David Hoyt, executive director of the American College of Surgeons, said the College is analyzing the use of bundled payments for surgery. Dr. M. Todd Williamson, of the Coalition of State Medical and National Specialty Societies, introduced the option of private contracting, in which patients would be free to apply their benefits to a doctor of their choice, who would be free to opt out on a per-patient basis.
“Private contracting is a key principle of American freedom and liberty,” Dr. Williamson said. “[It] will help the federal government achieve fiscal stability while fulfilling its promise to Medicare beneficiaries.”
Harold Miller, executive director of the Center for Healthcare Quality and Payment Reform, suggested an episode-of-care payment plan through which hospitals and physicians jointly charge one price for all services included in a hospitalization. The model would also include a warranty stating that any infections or complications would be treated at no additional cost. Also, a physician practice would receive one payment for all patient needs associated with chronic diseases or other conditions.
Rep. Burgess, who is also a physician, said organizations should focus on ways to address patients with chronic conditions, adding that 80% of Medicare funding is spent by 20% of beneficiaries with chronic illnesses.
Is the IPAB the new SGR?
Rep. Fred Upton (R–Mich) raised concerns about the Independent Payment Advisory Board (IPAB), crecreated by the Affordable Care Act. The Board sets expenditure targets on which it bases spending cuts. In 2018, targets will be based on the gross domestic product. “Sounds a lot like the SGR, which we’re trying to get rid of,” Mr. Upton said. “Since hospitals are exempt from IPAB cuts through the rest of the decade, it seems that the IPAB has the potential to undermine any serious efforts at physician payment reform.”
Some panelists agreed. “It’s not impossible that [the IPAB] could serve a function,” Dr. Wilson said, “but as presently constituted, we see it [as] basically another target for physicians to meet, potential double jeopardy, with an SGR as well as the pronouncements from this body.”
The panelists also asserted their belief that whatever plan chosen should be physician led, with financial support of the government. “It would be very helpful if physicians could get better financial support in their own payment system to enable them to lead all of those efforts,” said Dr. Mark B. McClellan, director of the Engelberg Center for Health Care Reform and former administrator of the Centers for Medicare and Medicaid Services. “Right now, with fee-for-service staying the way it is, they’re staying behind.” Dr. McClellan added that physicians can be the best sources for innovative and costsaving mechanisms.
A plan to finally replace Medicare’s much-maligned sustainable growth rate (SGR) payment formula could be unveiled by this summer, federal lawmakers said at a committee hearing. “Here’s the bottom line: If we get to December and we’re doing an extension, that’s a failure on our part,” Rep. Michael Burgess (R– Tex) said at the hearing. “We need a permanent solution that’s predictable, updatable, and reasonable for this year—and nothing else will do.”
“Whatever virtues the SGR had when it was created 14 years ago…, it’s clear that they have vanished,” added Rep. Henry A. Waxman (D– Calif). He noted that in the past 2 years, Congress has had to pass legislation six times, blocking fee cuts of up to 21% or more.
About 30 medical associations, including the American Society of Clinical Oncology (ASCO), responded to the House subcommittee’s request for suggestions and proposals in developing a new system. On May 5, 2011, House subcommittee members met with a five-person panel of experts from medical associations and health policy organizations to consider alternatives to the current SGR formula, which some participants labeled as anything but sustainable.
One size won’t fit all
Although the details of ASCO’s plan and others vary, they also show a consensus on several fronts: repealing the SGR, moving away from the traditional fee-for-services payment model, and providing a 4- to 5-year transition period during which providers can experiment with a variety of payment systems.
In a letter accompanying the ASCO recommendations, the president, Dr. George Sledge, and CEO, Dr. Allen Lichter, stressed that SGR reforms in general should be linked to existing “robust” systems that promote evidence-based medicine. For oncology in particular, that effort should leverage the Quality Oncology Practice Initiative (QOPI), a comprehensive, field-tested program that more than one-quarter of outpatient oncology practices in the United States already participate in. More than 80% of oncology care is provided in that setting.
“The current SGR system has created an uncertain and unstable environment— a situation that threatens the viability of practices and access to care for thousands of cancer patients,” they concluded.
In its recommendations, ASCO asserted that evidence-based medicine is “both warranted and necessary” because:
- Medicare beneficiaries account for more than half of all new cancer diagnoses in the United States, and treatment and prevention of the disease comprise almost 10% of costs under fee-for-service Medicare;
- The care is complex, treatment can span many specialties, and treatment strategies change rapidly to keep pace with scientific advances; and
- These complexities would not be adequately addressed if a multispecialty system (such as the Physician Quality Reporting System) were to be applied in the oncology setting.
The recommendations also detailed why the QOPI should be incorporated as the primary quality measurement program: 25%−30% of a range of practices—urban, rural, community, and academic—participate in it; it is free; some private insurers have adopted incentives for participation in the program; the performance measures are field-tested and up-to-date; and participation promotes high-quality, high-value care and can help identify and address discrepancies in oncology care.
Moreover, QOPI “protects the best interests of patients, reduces exposure to unnecessary treatments and tests, minimizes the use of suboptimal treatment options, promotes the coordination of care, and protects the Medicare program from costs associated with poor-quality care,” ASCO asserted in the recommendation.
Members of the expert panel also stressed the importance of avoiding a “one-size-fits-all” solution. “We should [be mindful] that what will work in one part of the country will not work in another part of the country, and that’s why we have continued to talk about a variety of options,” said Dr. Cecil Wilson, president of the American Medical Association (AMA). “There is a temptation to feel that we ought to figure out one rule …that solves it all.”
Dr. Wilson pointed to the provisions in the Affordable Care Act that allow for a variety of models of accountable care organizations, embodying the concept of options in the medical system. In that spirit, he said that the AMA has formed a physician leadership group to evaluate the effectiveness of alternative payment methods.
Dr. Roland A. Goertz, president of the American Academy of Family Physicians (AAFP), noted in written testimony to the committee that “the evidence shows that to achieve the savings Congress is looking for, and to improve the quality of health care delivered to millions of patients in the country, reform must include investment in primary care.”
To strengthen primary care’s role in Medicare, the AAFP backs payment reforms that would boost primary care reimbursement and support the concept of the patient-centered medical home (PCMH). The AAFP’s proposal would create a blended reimbursement system for primary care delivered within a PCMH: fee-forservice payments and pay for performance, plus care management fees for PCMH-related activities that do not involve direct patient care.
To prepare for that new payment system, the AAFP has proposed a 5-year transition period with mandated pay increases for primary care physicians, an increase in the Primary Care Incentive Care payment from 10% to 20%, and a rule that Medicaid payments to primary care physicians will always be at least equal to Medicare payments.
Dr. David Hoyt, executive director of the American College of Surgeons, said the College is analyzing the use of bundled payments for surgery. Dr. M. Todd Williamson, of the Coalition of State Medical and National Specialty Societies, introduced the option of private contracting, in which patients would be free to apply their benefits to a doctor of their choice, who would be free to opt out on a per-patient basis.
“Private contracting is a key principle of American freedom and liberty,” Dr. Williamson said. “[It] will help the federal government achieve fiscal stability while fulfilling its promise to Medicare beneficiaries.”
Harold Miller, executive director of the Center for Healthcare Quality and Payment Reform, suggested an episode-of-care payment plan through which hospitals and physicians jointly charge one price for all services included in a hospitalization. The model would also include a warranty stating that any infections or complications would be treated at no additional cost. Also, a physician practice would receive one payment for all patient needs associated with chronic diseases or other conditions.
Rep. Burgess, who is also a physician, said organizations should focus on ways to address patients with chronic conditions, adding that 80% of Medicare funding is spent by 20% of beneficiaries with chronic illnesses.
Is the IPAB the new SGR?
Rep. Fred Upton (R–Mich) raised concerns about the Independent Payment Advisory Board (IPAB), crecreated by the Affordable Care Act. The Board sets expenditure targets on which it bases spending cuts. In 2018, targets will be based on the gross domestic product. “Sounds a lot like the SGR, which we’re trying to get rid of,” Mr. Upton said. “Since hospitals are exempt from IPAB cuts through the rest of the decade, it seems that the IPAB has the potential to undermine any serious efforts at physician payment reform.”
Some panelists agreed. “It’s not impossible that [the IPAB] could serve a function,” Dr. Wilson said, “but as presently constituted, we see it [as] basically another target for physicians to meet, potential double jeopardy, with an SGR as well as the pronouncements from this body.”
The panelists also asserted their belief that whatever plan chosen should be physician led, with financial support of the government. “It would be very helpful if physicians could get better financial support in their own payment system to enable them to lead all of those efforts,” said Dr. Mark B. McClellan, director of the Engelberg Center for Health Care Reform and former administrator of the Centers for Medicare and Medicaid Services. “Right now, with fee-for-service staying the way it is, they’re staying behind.” Dr. McClellan added that physicians can be the best sources for innovative and costsaving mechanisms.
A plan to finally replace Medicare’s much-maligned sustainable growth rate (SGR) payment formula could be unveiled by this summer, federal lawmakers said at a committee hearing. “Here’s the bottom line: If we get to December and we’re doing an extension, that’s a failure on our part,” Rep. Michael Burgess (R– Tex) said at the hearing. “We need a permanent solution that’s predictable, updatable, and reasonable for this year—and nothing else will do.”
“Whatever virtues the SGR had when it was created 14 years ago…, it’s clear that they have vanished,” added Rep. Henry A. Waxman (D– Calif). He noted that in the past 2 years, Congress has had to pass legislation six times, blocking fee cuts of up to 21% or more.
About 30 medical associations, including the American Society of Clinical Oncology (ASCO), responded to the House subcommittee’s request for suggestions and proposals in developing a new system. On May 5, 2011, House subcommittee members met with a five-person panel of experts from medical associations and health policy organizations to consider alternatives to the current SGR formula, which some participants labeled as anything but sustainable.
One size won’t fit all
Although the details of ASCO’s plan and others vary, they also show a consensus on several fronts: repealing the SGR, moving away from the traditional fee-for-services payment model, and providing a 4- to 5-year transition period during which providers can experiment with a variety of payment systems.
In a letter accompanying the ASCO recommendations, the president, Dr. George Sledge, and CEO, Dr. Allen Lichter, stressed that SGR reforms in general should be linked to existing “robust” systems that promote evidence-based medicine. For oncology in particular, that effort should leverage the Quality Oncology Practice Initiative (QOPI), a comprehensive, field-tested program that more than one-quarter of outpatient oncology practices in the United States already participate in. More than 80% of oncology care is provided in that setting.
“The current SGR system has created an uncertain and unstable environment— a situation that threatens the viability of practices and access to care for thousands of cancer patients,” they concluded.
In its recommendations, ASCO asserted that evidence-based medicine is “both warranted and necessary” because:
- Medicare beneficiaries account for more than half of all new cancer diagnoses in the United States, and treatment and prevention of the disease comprise almost 10% of costs under fee-for-service Medicare;
- The care is complex, treatment can span many specialties, and treatment strategies change rapidly to keep pace with scientific advances; and
- These complexities would not be adequately addressed if a multispecialty system (such as the Physician Quality Reporting System) were to be applied in the oncology setting.
The recommendations also detailed why the QOPI should be incorporated as the primary quality measurement program: 25%−30% of a range of practices—urban, rural, community, and academic—participate in it; it is free; some private insurers have adopted incentives for participation in the program; the performance measures are field-tested and up-to-date; and participation promotes high-quality, high-value care and can help identify and address discrepancies in oncology care.
Moreover, QOPI “protects the best interests of patients, reduces exposure to unnecessary treatments and tests, minimizes the use of suboptimal treatment options, promotes the coordination of care, and protects the Medicare program from costs associated with poor-quality care,” ASCO asserted in the recommendation.
Members of the expert panel also stressed the importance of avoiding a “one-size-fits-all” solution. “We should [be mindful] that what will work in one part of the country will not work in another part of the country, and that’s why we have continued to talk about a variety of options,” said Dr. Cecil Wilson, president of the American Medical Association (AMA). “There is a temptation to feel that we ought to figure out one rule …that solves it all.”
Dr. Wilson pointed to the provisions in the Affordable Care Act that allow for a variety of models of accountable care organizations, embodying the concept of options in the medical system. In that spirit, he said that the AMA has formed a physician leadership group to evaluate the effectiveness of alternative payment methods.
Dr. Roland A. Goertz, president of the American Academy of Family Physicians (AAFP), noted in written testimony to the committee that “the evidence shows that to achieve the savings Congress is looking for, and to improve the quality of health care delivered to millions of patients in the country, reform must include investment in primary care.”
To strengthen primary care’s role in Medicare, the AAFP backs payment reforms that would boost primary care reimbursement and support the concept of the patient-centered medical home (PCMH). The AAFP’s proposal would create a blended reimbursement system for primary care delivered within a PCMH: fee-forservice payments and pay for performance, plus care management fees for PCMH-related activities that do not involve direct patient care.
To prepare for that new payment system, the AAFP has proposed a 5-year transition period with mandated pay increases for primary care physicians, an increase in the Primary Care Incentive Care payment from 10% to 20%, and a rule that Medicaid payments to primary care physicians will always be at least equal to Medicare payments.
Dr. David Hoyt, executive director of the American College of Surgeons, said the College is analyzing the use of bundled payments for surgery. Dr. M. Todd Williamson, of the Coalition of State Medical and National Specialty Societies, introduced the option of private contracting, in which patients would be free to apply their benefits to a doctor of their choice, who would be free to opt out on a per-patient basis.
“Private contracting is a key principle of American freedom and liberty,” Dr. Williamson said. “[It] will help the federal government achieve fiscal stability while fulfilling its promise to Medicare beneficiaries.”
Harold Miller, executive director of the Center for Healthcare Quality and Payment Reform, suggested an episode-of-care payment plan through which hospitals and physicians jointly charge one price for all services included in a hospitalization. The model would also include a warranty stating that any infections or complications would be treated at no additional cost. Also, a physician practice would receive one payment for all patient needs associated with chronic diseases or other conditions.
Rep. Burgess, who is also a physician, said organizations should focus on ways to address patients with chronic conditions, adding that 80% of Medicare funding is spent by 20% of beneficiaries with chronic illnesses.
Is the IPAB the new SGR?
Rep. Fred Upton (R–Mich) raised concerns about the Independent Payment Advisory Board (IPAB), crecreated by the Affordable Care Act. The Board sets expenditure targets on which it bases spending cuts. In 2018, targets will be based on the gross domestic product. “Sounds a lot like the SGR, which we’re trying to get rid of,” Mr. Upton said. “Since hospitals are exempt from IPAB cuts through the rest of the decade, it seems that the IPAB has the potential to undermine any serious efforts at physician payment reform.”
Some panelists agreed. “It’s not impossible that [the IPAB] could serve a function,” Dr. Wilson said, “but as presently constituted, we see it [as] basically another target for physicians to meet, potential double jeopardy, with an SGR as well as the pronouncements from this body.”
The panelists also asserted their belief that whatever plan chosen should be physician led, with financial support of the government. “It would be very helpful if physicians could get better financial support in their own payment system to enable them to lead all of those efforts,” said Dr. Mark B. McClellan, director of the Engelberg Center for Health Care Reform and former administrator of the Centers for Medicare and Medicaid Services. “Right now, with fee-for-service staying the way it is, they’re staying behind.” Dr. McClellan added that physicians can be the best sources for innovative and costsaving mechanisms.
FROM THE FEDERAL REGISTER