SSO 2013

NATIONAL HARBOR, MD. – Small surgical margins can mean big trouble for patients with breast cancer, said investigators at the annual Society of Surgical Oncology Cancer Symposium.

A retrospective study of outcomes for 2,377 women who underwent either breast-conserving therapy or mastectomy revealed that margins less than 2 mm resulted in a substantial risk of residual disease for all patients, reported Dr. Erin Garvey, a general surgery resident at Mayo Clinic Arizona in Phoenix.

"A policy of re-excision for margins less than 2 mm, coupled with a standardized multidisciplinary approach to breast cancer surgery, results in excellent re-excision and 5-year local recurrence rates. The local recurrence rate is higher, however, for those patients who complete breast-conserving therapy, thus warranting appropriate patient counseling regarding re-excision options and long-term outcome expectations," she said.

In a separate study, investigators from the University of Texas M.D. Anderson Cancer Center, Houston, reported 10-year follow-up data for women who opted for mastectomy to treat ductal carcinoma in situ (DCIS). They found that the incidence of local-regional recurrence (LRR) increased as the surgical margins shrank, and that close margins were the only independent predictor of LRR, reported Dr. Elizabeth FitzSullivan, a surgery fellow at M.D. Anderson.

"However, the local-regional recurrence rate in these patients is so low that routine postmastectomy radiation therapy is not warranted," she said.

No accord on margins

Despite multiple studies and meta-analyses, there is no standard for acceptable margin width in breast cancer, and surveys of both surgeons and radiation oncologists have shown wide variations in preferred margin widths, Dr. Garvey said.

Her group hypothesized that patients with invasive ductal carcinoma without an extensive in situ component who had surgical margins of at least 1 mm would have no evidence of residual disease on re-excision.

To test the idea, they took a retrospective look at records from a prospective database on 2,377 patients who underwent a total of 2,520 procedures from January 2000 through May 2012.

Of this group, 1,498 (63%) underwent lumpectomy, and 180 (12%) required re-excision surgery: 10% who had breast-conserving surgery alone, and 2% whose surgeries were converted to mastectomies. Of the 158 patients who had completed breast-conserving therapy following re-excision, 50 (32%) had residual disease, as did 20 of the 27 patients whose procedures were converted to mastectomies.

Of the 37% (879) who had up-front mastectomies, 2% (19) had re-excision, and of this group, 5 patients had residual disease.

In all, 40% of patients with positive margins had residual disease, compared with 38% of those with margins from 0.1 to 0.9 mm, and 33% for those with margins from 1.0 to 1.9 mm.

In univariate analysis, the presence of residual disease on re-excision did not show any significant association with age, race, menopausal status, width of the closest final margin, hormone receptor status, tumor histology, triple-negative disease, or the presence of angiolymphatic invasion. There was a trend, albeit nonsignificant, toward an association between residual disease and more than one margin narrower than 2 mm, Dr. Garvey noted.

At a median follow-up of 43 months (range, 0-140 months), 5-year local recurrence rates were 1.9% for patients who had breast-conserving therapy, and 1.1% for those who had mastectomy.

Patients who underwent breast-conserving therapy without re-excision had a 5-year local recurrence rate of 1.8%, compared with 4.3% for those who required re-excision, and 0% for those whose procedures were converted to mastectomy.

There was a nonsignificant trend toward higher local recurrence rates for breast-conserving therapy in patients who had re-excisions, which became significant when those patients who had conversion to mastectomy were excluded, with a hazard ratio compared with no re-excision of 2.56 (P = .04).

Narrower margins, larger risk

Dr. FitzSullivan and her M.D. Anderson colleagues reviewed the records of 810 women treated with mastectomy for DCIS from 1996 to 2009. They looked at the final width of histologic margins, defining disease-free margins as those of 3 mm or greater.

In all, 4 patients had positive margins, 59 had margins of 1 mm or smaller, and 35 had margins from 1.1 to 2.9 mm.

In multivariate analysis, independent predictors of close or positive margins were pathologic tumor size of 1.5 cm or greater (odds ratio, 5.11; P = .001), multicentric disease (OR, 5.44; P = .026), and the presence of necrosis (OR, 2.5; P = .003). Neither age, postmenopausal status, skin-sparing mastectomy, nor immediate breast reconstruction were significantly associated with close or positive margins, however.

None of seven patients who underwent postmastectomy radiotherapy had local-regional recurrences. Of the 803 patients who did not receive postsurgery radiation, 10-year LRR rates were 1%, consisting of 7 cases of invasive disease and 1 of DICS. Five patients had surgical management, and the remaining 3 had no further treatment.

When the researchers stratified the local recurrence rates by margin status, they saw that 5% of patients with margins of 1 mm or smaller had LRRs within 10 years, as did 3.6% of those with margins from 1.1 to 2.9 mm, compared with just 0.07% of those with disease-free margins (P less than .001). There was no difference in LRR between the two narrow-margin groups.

Among 546 patients with an intact contralateral breast, the 10-year rate of contralateral breast disease was 6.4%.

On univariate analysis, significant predictors of LRR included margin status (P = .002), multicentric disease (P = .005), and necrosis (P = .005). On multivariate analysis, however, only margin status remained significant, with an HR of 8.0 (P = .006).

Dr. FitzSullivan said that the low rate of LRR of DCIS treated with mastectomy and close surgical margins, compared with the rate of contralateral breast cancer, suggests that routine postmastectomy radiation therapy is not warranted, and should be reserved only for those patients with close or positive surgical margins that cannot be surgically excised.

Each study was internally funded. Dr. Garvey and Dr. FitzSullivan reported having no financial disclosures.

[email protected]

NATIONAL HARBOR, MD. – Small surgical margins can mean big trouble for patients with breast cancer, said investigators at the annual Society of Surgical Oncology Cancer Symposium.

A retrospective study of outcomes for 2,377 women who underwent either breast-conserving therapy or mastectomy revealed that margins less than 2 mm resulted in a substantial risk of residual disease for all patients, reported Dr. Erin Garvey, a general surgery resident at Mayo Clinic Arizona in Phoenix.

"A policy of re-excision for margins less than 2 mm, coupled with a standardized multidisciplinary approach to breast cancer surgery, results in excellent re-excision and 5-year local recurrence rates. The local recurrence rate is higher, however, for those patients who complete breast-conserving therapy, thus warranting appropriate patient counseling regarding re-excision options and long-term outcome expectations," she said.

In a separate study, investigators from the University of Texas M.D. Anderson Cancer Center, Houston, reported 10-year follow-up data for women who opted for mastectomy to treat ductal carcinoma in situ (DCIS). They found that the incidence of local-regional recurrence (LRR) increased as the surgical margins shrank, and that close margins were the only independent predictor of LRR, reported Dr. Elizabeth FitzSullivan, a surgery fellow at M.D. Anderson.

"However, the local-regional recurrence rate in these patients is so low that routine postmastectomy radiation therapy is not warranted," she said.

No accord on margins

Despite multiple studies and meta-analyses, there is no standard for acceptable margin width in breast cancer, and surveys of both surgeons and radiation oncologists have shown wide variations in preferred margin widths, Dr. Garvey said.

Her group hypothesized that patients with invasive ductal carcinoma without an extensive in situ component who had surgical margins of at least 1 mm would have no evidence of residual disease on re-excision.

To test the idea, they took a retrospective look at records from a prospective database on 2,377 patients who underwent a total of 2,520 procedures from January 2000 through May 2012.

Of this group, 1,498 (63%) underwent lumpectomy, and 180 (12%) required re-excision surgery: 10% who had breast-conserving surgery alone, and 2% whose surgeries were converted to mastectomies. Of the 158 patients who had completed breast-conserving therapy following re-excision, 50 (32%) had residual disease, as did 20 of the 27 patients whose procedures were converted to mastectomies.

Of the 37% (879) who had up-front mastectomies, 2% (19) had re-excision, and of this group, 5 patients had residual disease.

In all, 40% of patients with positive margins had residual disease, compared with 38% of those with margins from 0.1 to 0.9 mm, and 33% for those with margins from 1.0 to 1.9 mm.

In univariate analysis, the presence of residual disease on re-excision did not show any significant association with age, race, menopausal status, width of the closest final margin, hormone receptor status, tumor histology, triple-negative disease, or the presence of angiolymphatic invasion. There was a trend, albeit nonsignificant, toward an association between residual disease and more than one margin narrower than 2 mm, Dr. Garvey noted.

At a median follow-up of 43 months (range, 0-140 months), 5-year local recurrence rates were 1.9% for patients who had breast-conserving therapy, and 1.1% for those who had mastectomy.

Patients who underwent breast-conserving therapy without re-excision had a 5-year local recurrence rate of 1.8%, compared with 4.3% for those who required re-excision, and 0% for those whose procedures were converted to mastectomy.

There was a nonsignificant trend toward higher local recurrence rates for breast-conserving therapy in patients who had re-excisions, which became significant when those patients who had conversion to mastectomy were excluded, with a hazard ratio compared with no re-excision of 2.56 (P = .04).

Narrower margins, larger risk

Dr. FitzSullivan and her M.D. Anderson colleagues reviewed the records of 810 women treated with mastectomy for DCIS from 1996 to 2009. They looked at the final width of histologic margins, defining disease-free margins as those of 3 mm or greater.

In all, 4 patients had positive margins, 59 had margins of 1 mm or smaller, and 35 had margins from 1.1 to 2.9 mm.

In multivariate analysis, independent predictors of close or positive margins were pathologic tumor size of 1.5 cm or greater (odds ratio, 5.11; P = .001), multicentric disease (OR, 5.44; P = .026), and the presence of necrosis (OR, 2.5; P = .003). Neither age, postmenopausal status, skin-sparing mastectomy, nor immediate breast reconstruction were significantly associated with close or positive margins, however.

None of seven patients who underwent postmastectomy radiotherapy had local-regional recurrences. Of the 803 patients who did not receive postsurgery radiation, 10-year LRR rates were 1%, consisting of 7 cases of invasive disease and 1 of DICS. Five patients had surgical management, and the remaining 3 had no further treatment.

When the researchers stratified the local recurrence rates by margin status, they saw that 5% of patients with margins of 1 mm or smaller had LRRs within 10 years, as did 3.6% of those with margins from 1.1 to 2.9 mm, compared with just 0.07% of those with disease-free margins (P less than .001). There was no difference in LRR between the two narrow-margin groups.

Among 546 patients with an intact contralateral breast, the 10-year rate of contralateral breast disease was 6.4%.

On univariate analysis, significant predictors of LRR included margin status (P = .002), multicentric disease (P = .005), and necrosis (P = .005). On multivariate analysis, however, only margin status remained significant, with an HR of 8.0 (P = .006).

Dr. FitzSullivan said that the low rate of LRR of DCIS treated with mastectomy and close surgical margins, compared with the rate of contralateral breast cancer, suggests that routine postmastectomy radiation therapy is not warranted, and should be reserved only for those patients with close or positive surgical margins that cannot be surgically excised.

Each study was internally funded. Dr. Garvey and Dr. FitzSullivan reported having no financial disclosures.

[email protected]

NATIONAL HARBOR, MD. – Small surgical margins can mean big trouble for patients with breast cancer, said investigators at the annual Society of Surgical Oncology Cancer Symposium.

A retrospective study of outcomes for 2,377 women who underwent either breast-conserving therapy or mastectomy revealed that margins less than 2 mm resulted in a substantial risk of residual disease for all patients, reported Dr. Erin Garvey, a general surgery resident at Mayo Clinic Arizona in Phoenix.

"A policy of re-excision for margins less than 2 mm, coupled with a standardized multidisciplinary approach to breast cancer surgery, results in excellent re-excision and 5-year local recurrence rates. The local recurrence rate is higher, however, for those patients who complete breast-conserving therapy, thus warranting appropriate patient counseling regarding re-excision options and long-term outcome expectations," she said.

In a separate study, investigators from the University of Texas M.D. Anderson Cancer Center, Houston, reported 10-year follow-up data for women who opted for mastectomy to treat ductal carcinoma in situ (DCIS). They found that the incidence of local-regional recurrence (LRR) increased as the surgical margins shrank, and that close margins were the only independent predictor of LRR, reported Dr. Elizabeth FitzSullivan, a surgery fellow at M.D. Anderson.

"However, the local-regional recurrence rate in these patients is so low that routine postmastectomy radiation therapy is not warranted," she said.

No accord on margins

Despite multiple studies and meta-analyses, there is no standard for acceptable margin width in breast cancer, and surveys of both surgeons and radiation oncologists have shown wide variations in preferred margin widths, Dr. Garvey said.

Her group hypothesized that patients with invasive ductal carcinoma without an extensive in situ component who had surgical margins of at least 1 mm would have no evidence of residual disease on re-excision.

To test the idea, they took a retrospective look at records from a prospective database on 2,377 patients who underwent a total of 2,520 procedures from January 2000 through May 2012.

Of this group, 1,498 (63%) underwent lumpectomy, and 180 (12%) required re-excision surgery: 10% who had breast-conserving surgery alone, and 2% whose surgeries were converted to mastectomies. Of the 158 patients who had completed breast-conserving therapy following re-excision, 50 (32%) had residual disease, as did 20 of the 27 patients whose procedures were converted to mastectomies.

Of the 37% (879) who had up-front mastectomies, 2% (19) had re-excision, and of this group, 5 patients had residual disease.

In all, 40% of patients with positive margins had residual disease, compared with 38% of those with margins from 0.1 to 0.9 mm, and 33% for those with margins from 1.0 to 1.9 mm.

In univariate analysis, the presence of residual disease on re-excision did not show any significant association with age, race, menopausal status, width of the closest final margin, hormone receptor status, tumor histology, triple-negative disease, or the presence of angiolymphatic invasion. There was a trend, albeit nonsignificant, toward an association between residual disease and more than one margin narrower than 2 mm, Dr. Garvey noted.

At a median follow-up of 43 months (range, 0-140 months), 5-year local recurrence rates were 1.9% for patients who had breast-conserving therapy, and 1.1% for those who had mastectomy.

Patients who underwent breast-conserving therapy without re-excision had a 5-year local recurrence rate of 1.8%, compared with 4.3% for those who required re-excision, and 0% for those whose procedures were converted to mastectomy.

There was a nonsignificant trend toward higher local recurrence rates for breast-conserving therapy in patients who had re-excisions, which became significant when those patients who had conversion to mastectomy were excluded, with a hazard ratio compared with no re-excision of 2.56 (P = .04).

Narrower margins, larger risk

Dr. FitzSullivan and her M.D. Anderson colleagues reviewed the records of 810 women treated with mastectomy for DCIS from 1996 to 2009. They looked at the final width of histologic margins, defining disease-free margins as those of 3 mm or greater.

In all, 4 patients had positive margins, 59 had margins of 1 mm or smaller, and 35 had margins from 1.1 to 2.9 mm.

In multivariate analysis, independent predictors of close or positive margins were pathologic tumor size of 1.5 cm or greater (odds ratio, 5.11; P = .001), multicentric disease (OR, 5.44; P = .026), and the presence of necrosis (OR, 2.5; P = .003). Neither age, postmenopausal status, skin-sparing mastectomy, nor immediate breast reconstruction were significantly associated with close or positive margins, however.

None of seven patients who underwent postmastectomy radiotherapy had local-regional recurrences. Of the 803 patients who did not receive postsurgery radiation, 10-year LRR rates were 1%, consisting of 7 cases of invasive disease and 1 of DICS. Five patients had surgical management, and the remaining 3 had no further treatment.

When the researchers stratified the local recurrence rates by margin status, they saw that 5% of patients with margins of 1 mm or smaller had LRRs within 10 years, as did 3.6% of those with margins from 1.1 to 2.9 mm, compared with just 0.07% of those with disease-free margins (P less than .001). There was no difference in LRR between the two narrow-margin groups.

Among 546 patients with an intact contralateral breast, the 10-year rate of contralateral breast disease was 6.4%.

On univariate analysis, significant predictors of LRR included margin status (P = .002), multicentric disease (P = .005), and necrosis (P = .005). On multivariate analysis, however, only margin status remained significant, with an HR of 8.0 (P = .006).

Dr. FitzSullivan said that the low rate of LRR of DCIS treated with mastectomy and close surgical margins, compared with the rate of contralateral breast cancer, suggests that routine postmastectomy radiation therapy is not warranted, and should be reserved only for those patients with close or positive surgical margins that cannot be surgically excised.

Each study was internally funded. Dr. Garvey and Dr. FitzSullivan reported having no financial disclosures.

[email protected]

NATIONAL HARBOR, MD. – There are no hard and fast rules when it comes to selecting neoadjuvant or adjuvant therapy for patients with gastric cancer, according to Dr. Daniel G. Coit.

"Clinical trials of adjuvant therapy describe the impact of treatment on one group relative to another group. They do not predict the impact of treatment on an individual patient," said Dr. Coit, a surgical oncologist at Memorial Sloan-Kettering Cancer Center, New York.

Patient selection is a multi-step process that must incorporate an understanding of the risk of recurrence, the absolute benefit of treatment, strategies to improve the odds for achieving a benefit, and individual patient comorbidities and preferences, Dr. Coit said at the annual Society of Surgical Oncology Cancer Symposium.

Where oncologists often run into trouble is in the interpretation of data on benefits from clinical trials, and on the question of neoadjuvant vs. adjuvant therapy, both, or neither, he said.

Therapy delivered prior to surgery is likely to be better tolerated and improve the chance of R0 or "clean" resections. Neoadjuvant therapy can also help to identify, before surgery, those patients unlikely to respond metabolically or histologically to therapy, which can serve as both a prognostic factor for poor outcomes, and as a guide for additional pre- or postoperative therapy, Dr. Coit said.

Adjuvant therapy, on the other hand, allows treatment decisions to be based on an accurate estimate of recurrence risk using postoperative pathologic staging, he added.

"If we’re going to even consider preoperative therapy, neoadjuvant chemotherapy, we need to have some sense for how good our preoperative or pretreatment risk-assessment tools are. They’re not all that good," he said.

He noted that in a 2001 study (J. Clin. Gastroenterol. 2001;32:41-4) of 549 patients who were tested for carcinoembryonic antigen (CEA) and cancer antigen 19-9 (CA-19-9) before gastrectomy for gastric cancer, elevated levels of both markers were associated with "increased depth of invasion, more extensive nodal metastases, liver metastases, and less ‘curability.’ "

However, the authors found on multivariate analysis that a doubling of the normal threshold CEA was the only preoperative factor predictive of prognosis.

Endoscopic ultrasound (EUS), which is about 75% accurate at identifying node-positive vs. node-negative disease and at distinguishing early- from late-stage disease, is "very poorly predictive of outcome."

"The same thing happens with N stage: the endoscopic observation of node positivity is a terrible predictor of outcome," Dr. Coit said.

There are a few published nomograms that are fairly good at predicting disease-specific survival after R0 resections for gastric cancer, he noted, but these are all based on postoperative pathology findings, and "as yet we do not have a good nomogram for preoperative risk assessment."

Preventing recurrence

The goal of current therapy is to prevent recurrence at all costs, because few patients who experience recurrence will be salvageable, Dr. Coit said.

Studies of adjuvant therapy for gastric cancer with either postoperative 5-fluorauracil (5FU) and radiation (N. Engl. J. Med. 2001;345:725-30), postoperative epirubicin, cisplatin, and continuous 5FU infusion (ECF) (N. Engl. J. Med. 2006;355:11-20), or the oral fluoropyrimidine agent S-1, which is not available in the United States (N. Engl. J. Med. 2007:357:1810-20) all show about a 10% benefit, meaning that 10 patients would need to be treated for 1 to benefit.

More recently, in the CLASSIC trial (Lancet 2012;379:315-21) 1,035 patients with stage II-III gastric cancer who underwent curative resections with D2 extended lymph node dissection who were randomized to adjuvant capecitabine (Xeloda) and oxaliplatin had a 74% 3-year disease-free survival rate, compared with 59% for patients managed with postoperative observation only (hazard ratio 0.56, P less than .0001).

This trial showed that if patients are determined to have high-risk disease on postoperative pathology, "you have a very legitimate option available to you for postoperative treatment," Dr. Coit said.

In the ARTIST trial (J. Clin. Oncol. 2011;30:268-73), 458 patients underwent curative surgery with D2 lymph node dissection and were randomized to postoperative Xeloda/cisplatin (XP) or XP plus concurrent capecitabine radio therapy. The researchers found no statistically significant difference in 3-year disease-free survival among all patients, although there was some evidence of a benefit for XP plus radiation in patients with node-positive disease (P = .04).

Enriching the population

"It’s very clear that clinical trials do show a modest improvement in survival among the group of patients receiving adjuvant chemotherapy. The adjuvant therapy has no impact whatsoever in virtually 90% of the patients who are being treated, and we need to enrich this population that’s likely to benefit," he stated.

One strategy available for better selection of patients for adjuvant therapy is genetic tumor profiling to indentify targetable mutations such as HER-2 which is overexpressed in some gastric cancers.

A promising approach for identifying patients who might benefit from neoadjuvant therapy involves tumor uptake of fluorodeoxyglucose on positron-emission tomography (FDG-PET). In the MUNICON phase II trial (Lancet Oncol 2007;8:797-805), patients deemed to be responders, defined by decreases in tumor glucose standard uptake values (SUVs), had a median event-free survival of 29.7 months compared with 14.1 months in nonresponders (hazard ratio 2.18, P = .002), and 29 of 29 of 50 metabolic responders (58% ) had major histologic remissions, whereas none of the metabolic nonresponders did.

Dr. Coit reported having no financial disclosures.

NATIONAL HARBOR, MD. – There are no hard and fast rules when it comes to selecting neoadjuvant or adjuvant therapy for patients with gastric cancer, according to Dr. Daniel G. Coit.

"Clinical trials of adjuvant therapy describe the impact of treatment on one group relative to another group. They do not predict the impact of treatment on an individual patient," said Dr. Coit, a surgical oncologist at Memorial Sloan-Kettering Cancer Center, New York.

Patient selection is a multi-step process that must incorporate an understanding of the risk of recurrence, the absolute benefit of treatment, strategies to improve the odds for achieving a benefit, and individual patient comorbidities and preferences, Dr. Coit said at the annual Society of Surgical Oncology Cancer Symposium.

Where oncologists often run into trouble is in the interpretation of data on benefits from clinical trials, and on the question of neoadjuvant vs. adjuvant therapy, both, or neither, he said.

Therapy delivered prior to surgery is likely to be better tolerated and improve the chance of R0 or "clean" resections. Neoadjuvant therapy can also help to identify, before surgery, those patients unlikely to respond metabolically or histologically to therapy, which can serve as both a prognostic factor for poor outcomes, and as a guide for additional pre- or postoperative therapy, Dr. Coit said.

Adjuvant therapy, on the other hand, allows treatment decisions to be based on an accurate estimate of recurrence risk using postoperative pathologic staging, he added.

"If we’re going to even consider preoperative therapy, neoadjuvant chemotherapy, we need to have some sense for how good our preoperative or pretreatment risk-assessment tools are. They’re not all that good," he said.

He noted that in a 2001 study (J. Clin. Gastroenterol. 2001;32:41-4) of 549 patients who were tested for carcinoembryonic antigen (CEA) and cancer antigen 19-9 (CA-19-9) before gastrectomy for gastric cancer, elevated levels of both markers were associated with "increased depth of invasion, more extensive nodal metastases, liver metastases, and less ‘curability.’ "

However, the authors found on multivariate analysis that a doubling of the normal threshold CEA was the only preoperative factor predictive of prognosis.

Endoscopic ultrasound (EUS), which is about 75% accurate at identifying node-positive vs. node-negative disease and at distinguishing early- from late-stage disease, is "very poorly predictive of outcome."

"The same thing happens with N stage: the endoscopic observation of node positivity is a terrible predictor of outcome," Dr. Coit said.

There are a few published nomograms that are fairly good at predicting disease-specific survival after R0 resections for gastric cancer, he noted, but these are all based on postoperative pathology findings, and "as yet we do not have a good nomogram for preoperative risk assessment."

Preventing recurrence

The goal of current therapy is to prevent recurrence at all costs, because few patients who experience recurrence will be salvageable, Dr. Coit said.

Studies of adjuvant therapy for gastric cancer with either postoperative 5-fluorauracil (5FU) and radiation (N. Engl. J. Med. 2001;345:725-30), postoperative epirubicin, cisplatin, and continuous 5FU infusion (ECF) (N. Engl. J. Med. 2006;355:11-20), or the oral fluoropyrimidine agent S-1, which is not available in the United States (N. Engl. J. Med. 2007:357:1810-20) all show about a 10% benefit, meaning that 10 patients would need to be treated for 1 to benefit.

More recently, in the CLASSIC trial (Lancet 2012;379:315-21) 1,035 patients with stage II-III gastric cancer who underwent curative resections with D2 extended lymph node dissection who were randomized to adjuvant capecitabine (Xeloda) and oxaliplatin had a 74% 3-year disease-free survival rate, compared with 59% for patients managed with postoperative observation only (hazard ratio 0.56, P less than .0001).

This trial showed that if patients are determined to have high-risk disease on postoperative pathology, "you have a very legitimate option available to you for postoperative treatment," Dr. Coit said.

In the ARTIST trial (J. Clin. Oncol. 2011;30:268-73), 458 patients underwent curative surgery with D2 lymph node dissection and were randomized to postoperative Xeloda/cisplatin (XP) or XP plus concurrent capecitabine radio therapy. The researchers found no statistically significant difference in 3-year disease-free survival among all patients, although there was some evidence of a benefit for XP plus radiation in patients with node-positive disease (P = .04).

Enriching the population

"It’s very clear that clinical trials do show a modest improvement in survival among the group of patients receiving adjuvant chemotherapy. The adjuvant therapy has no impact whatsoever in virtually 90% of the patients who are being treated, and we need to enrich this population that’s likely to benefit," he stated.

One strategy available for better selection of patients for adjuvant therapy is genetic tumor profiling to indentify targetable mutations such as HER-2 which is overexpressed in some gastric cancers.

A promising approach for identifying patients who might benefit from neoadjuvant therapy involves tumor uptake of fluorodeoxyglucose on positron-emission tomography (FDG-PET). In the MUNICON phase II trial (Lancet Oncol 2007;8:797-805), patients deemed to be responders, defined by decreases in tumor glucose standard uptake values (SUVs), had a median event-free survival of 29.7 months compared with 14.1 months in nonresponders (hazard ratio 2.18, P = .002), and 29 of 29 of 50 metabolic responders (58% ) had major histologic remissions, whereas none of the metabolic nonresponders did.

Dr. Coit reported having no financial disclosures.

NATIONAL HARBOR, MD. – There are no hard and fast rules when it comes to selecting neoadjuvant or adjuvant therapy for patients with gastric cancer, according to Dr. Daniel G. Coit.

"Clinical trials of adjuvant therapy describe the impact of treatment on one group relative to another group. They do not predict the impact of treatment on an individual patient," said Dr. Coit, a surgical oncologist at Memorial Sloan-Kettering Cancer Center, New York.

Patient selection is a multi-step process that must incorporate an understanding of the risk of recurrence, the absolute benefit of treatment, strategies to improve the odds for achieving a benefit, and individual patient comorbidities and preferences, Dr. Coit said at the annual Society of Surgical Oncology Cancer Symposium.

Where oncologists often run into trouble is in the interpretation of data on benefits from clinical trials, and on the question of neoadjuvant vs. adjuvant therapy, both, or neither, he said.

Therapy delivered prior to surgery is likely to be better tolerated and improve the chance of R0 or "clean" resections. Neoadjuvant therapy can also help to identify, before surgery, those patients unlikely to respond metabolically or histologically to therapy, which can serve as both a prognostic factor for poor outcomes, and as a guide for additional pre- or postoperative therapy, Dr. Coit said.

Adjuvant therapy, on the other hand, allows treatment decisions to be based on an accurate estimate of recurrence risk using postoperative pathologic staging, he added.

"If we’re going to even consider preoperative therapy, neoadjuvant chemotherapy, we need to have some sense for how good our preoperative or pretreatment risk-assessment tools are. They’re not all that good," he said.

He noted that in a 2001 study (J. Clin. Gastroenterol. 2001;32:41-4) of 549 patients who were tested for carcinoembryonic antigen (CEA) and cancer antigen 19-9 (CA-19-9) before gastrectomy for gastric cancer, elevated levels of both markers were associated with "increased depth of invasion, more extensive nodal metastases, liver metastases, and less ‘curability.’ "

However, the authors found on multivariate analysis that a doubling of the normal threshold CEA was the only preoperative factor predictive of prognosis.

Endoscopic ultrasound (EUS), which is about 75% accurate at identifying node-positive vs. node-negative disease and at distinguishing early- from late-stage disease, is "very poorly predictive of outcome."

"The same thing happens with N stage: the endoscopic observation of node positivity is a terrible predictor of outcome," Dr. Coit said.

There are a few published nomograms that are fairly good at predicting disease-specific survival after R0 resections for gastric cancer, he noted, but these are all based on postoperative pathology findings, and "as yet we do not have a good nomogram for preoperative risk assessment."

Preventing recurrence

The goal of current therapy is to prevent recurrence at all costs, because few patients who experience recurrence will be salvageable, Dr. Coit said.

Studies of adjuvant therapy for gastric cancer with either postoperative 5-fluorauracil (5FU) and radiation (N. Engl. J. Med. 2001;345:725-30), postoperative epirubicin, cisplatin, and continuous 5FU infusion (ECF) (N. Engl. J. Med. 2006;355:11-20), or the oral fluoropyrimidine agent S-1, which is not available in the United States (N. Engl. J. Med. 2007:357:1810-20) all show about a 10% benefit, meaning that 10 patients would need to be treated for 1 to benefit.

More recently, in the CLASSIC trial (Lancet 2012;379:315-21) 1,035 patients with stage II-III gastric cancer who underwent curative resections with D2 extended lymph node dissection who were randomized to adjuvant capecitabine (Xeloda) and oxaliplatin had a 74% 3-year disease-free survival rate, compared with 59% for patients managed with postoperative observation only (hazard ratio 0.56, P less than .0001).

This trial showed that if patients are determined to have high-risk disease on postoperative pathology, "you have a very legitimate option available to you for postoperative treatment," Dr. Coit said.

In the ARTIST trial (J. Clin. Oncol. 2011;30:268-73), 458 patients underwent curative surgery with D2 lymph node dissection and were randomized to postoperative Xeloda/cisplatin (XP) or XP plus concurrent capecitabine radio therapy. The researchers found no statistically significant difference in 3-year disease-free survival among all patients, although there was some evidence of a benefit for XP plus radiation in patients with node-positive disease (P = .04).

Enriching the population

"It’s very clear that clinical trials do show a modest improvement in survival among the group of patients receiving adjuvant chemotherapy. The adjuvant therapy has no impact whatsoever in virtually 90% of the patients who are being treated, and we need to enrich this population that’s likely to benefit," he stated.

One strategy available for better selection of patients for adjuvant therapy is genetic tumor profiling to indentify targetable mutations such as HER-2 which is overexpressed in some gastric cancers.

A promising approach for identifying patients who might benefit from neoadjuvant therapy involves tumor uptake of fluorodeoxyglucose on positron-emission tomography (FDG-PET). In the MUNICON phase II trial (Lancet Oncol 2007;8:797-805), patients deemed to be responders, defined by decreases in tumor glucose standard uptake values (SUVs), had a median event-free survival of 29.7 months compared with 14.1 months in nonresponders (hazard ratio 2.18, P = .002), and 29 of 29 of 50 metabolic responders (58% ) had major histologic remissions, whereas none of the metabolic nonresponders did.

Dr. Coit reported having no financial disclosures.

NATIONAL HARBOR, MD – Patients with stage IV melanoma treated with a combination of ipilimumab and surgical resection had a high rate of melanoma-specific and overall survival, a retrospective study of a single-center case series has shown.

"To our knowledge, this is the first report of 5-year melanoma-specific survival data on patients who have undergone surgical resection and ipilimumab treatment, and the data suggests that surgical resection and ipilimumab treatment may result in long-term survival in select metastatic melanoma patients," Dr. Junko Ozao-Choy said at the annual Society of Surgical Oncology Cancer Symposium.

Among 44 patients treated with the CTLA-4 (cytotoxic T-lymphocyte antigen 4) inhibitor ipilimumab (Yervoy) and surgical resection, the 5-year melanoma-specific survival (MSS) rate was 51% and the median overall survival duration was 60 months, reported Dr. Ozao-Choy of the John Wayne Cancer Institute at Saint John’s Health Center in Santa Monica, Calif.

For 24 patients who received ipilimumab before resection, the 5-year MSS was 61% at a median of 60 months, and for 18 of 20 patients treated with ipilimumab after surgery, the 5-year MSS was 42% at a median of 47 months, but this difference was not significant (data were incomplete for 2 patients in the latter group), she noted.

In a recent study of retrospective data on patients with metastatic melanoma treated at her center, the 4-year survival of patients who underwent resection of metastatic lesions with or without systemic medical therapy was 20.8%, compared with 7% for those who underwent systemic medical therapy alone. The study investigators concluded that more than half of patients with metastatic melanoma were eligible for metastasectomy (Ann. Surg. Oncol. 2012;19:2547-55).

Dr. Ozao-Choy and her colleagues reviewed the center’s records on patients with metastatic melanoma who underwent resection and had received ipilimumab, looking at disease-specific survival from the date of diagnosis of stage IV disease.

The groups were well balanced in terms of age, sex, mean Breslow thickness scores, and nodal status. However, significantly more patients who received ipilimumab before surgery had brain metastases (13 of 24 vs. 3 of 18, P = .001). In a univariate analysis, patients with brain metastases had a significantly worse 5-year MSS (31% vs. 60%, P = .049).

The only other significant variables associated in the univariate analysis with better survival were prior immunotherapy, with 69% of patients who had received any immunotherapy having a 5-year MSS of 69%, compared with 29% for those with no immunotherapy (P = .01), and previous number of resections, with more resections being associated with better survival (P = .01).

Neither previous treatment with Bacillus Calmette-Guérin vaccine, previous chemotherapy, T stage, N stage, or timing of ipilimumab were significantly associated with MSS.

In a multivariate analysis (which controlled for demographic and disease factors), only the previous number of resections remained a significant predictor of MSS (P = .01).

In the audience response segment following the presentation, Dr. Daniel G. Coit, a surgical oncologist at Memorial Sloan-Kettering Cancer Center in New York City, pointed out that the previous number of resections is a not an adequate independent predictor for survival. "One of the inescapable truths is that if you have to have more than one operation, you have to still be alive. ... Of necessity, older people live longer than younger people; people who die at an older age live longer than people who die at a younger age," he said.

The study was internally funded. Dr. Ozao-Choy and Dr. Coit reported having no relevant financial disclosures.

NATIONAL HARBOR, MD – Patients with stage IV melanoma treated with a combination of ipilimumab and surgical resection had a high rate of melanoma-specific and overall survival, a retrospective study of a single-center case series has shown.

"To our knowledge, this is the first report of 5-year melanoma-specific survival data on patients who have undergone surgical resection and ipilimumab treatment, and the data suggests that surgical resection and ipilimumab treatment may result in long-term survival in select metastatic melanoma patients," Dr. Junko Ozao-Choy said at the annual Society of Surgical Oncology Cancer Symposium.

Among 44 patients treated with the CTLA-4 (cytotoxic T-lymphocyte antigen 4) inhibitor ipilimumab (Yervoy) and surgical resection, the 5-year melanoma-specific survival (MSS) rate was 51% and the median overall survival duration was 60 months, reported Dr. Ozao-Choy of the John Wayne Cancer Institute at Saint John’s Health Center in Santa Monica, Calif.

For 24 patients who received ipilimumab before resection, the 5-year MSS was 61% at a median of 60 months, and for 18 of 20 patients treated with ipilimumab after surgery, the 5-year MSS was 42% at a median of 47 months, but this difference was not significant (data were incomplete for 2 patients in the latter group), she noted.

In a recent study of retrospective data on patients with metastatic melanoma treated at her center, the 4-year survival of patients who underwent resection of metastatic lesions with or without systemic medical therapy was 20.8%, compared with 7% for those who underwent systemic medical therapy alone. The study investigators concluded that more than half of patients with metastatic melanoma were eligible for metastasectomy (Ann. Surg. Oncol. 2012;19:2547-55).

Dr. Ozao-Choy and her colleagues reviewed the center’s records on patients with metastatic melanoma who underwent resection and had received ipilimumab, looking at disease-specific survival from the date of diagnosis of stage IV disease.

The groups were well balanced in terms of age, sex, mean Breslow thickness scores, and nodal status. However, significantly more patients who received ipilimumab before surgery had brain metastases (13 of 24 vs. 3 of 18, P = .001). In a univariate analysis, patients with brain metastases had a significantly worse 5-year MSS (31% vs. 60%, P = .049).

The only other significant variables associated in the univariate analysis with better survival were prior immunotherapy, with 69% of patients who had received any immunotherapy having a 5-year MSS of 69%, compared with 29% for those with no immunotherapy (P = .01), and previous number of resections, with more resections being associated with better survival (P = .01).

Neither previous treatment with Bacillus Calmette-Guérin vaccine, previous chemotherapy, T stage, N stage, or timing of ipilimumab were significantly associated with MSS.

In a multivariate analysis (which controlled for demographic and disease factors), only the previous number of resections remained a significant predictor of MSS (P = .01).

In the audience response segment following the presentation, Dr. Daniel G. Coit, a surgical oncologist at Memorial Sloan-Kettering Cancer Center in New York City, pointed out that the previous number of resections is a not an adequate independent predictor for survival. "One of the inescapable truths is that if you have to have more than one operation, you have to still be alive. ... Of necessity, older people live longer than younger people; people who die at an older age live longer than people who die at a younger age," he said.

The study was internally funded. Dr. Ozao-Choy and Dr. Coit reported having no relevant financial disclosures.

NATIONAL HARBOR, MD – Patients with stage IV melanoma treated with a combination of ipilimumab and surgical resection had a high rate of melanoma-specific and overall survival, a retrospective study of a single-center case series has shown.

"To our knowledge, this is the first report of 5-year melanoma-specific survival data on patients who have undergone surgical resection and ipilimumab treatment, and the data suggests that surgical resection and ipilimumab treatment may result in long-term survival in select metastatic melanoma patients," Dr. Junko Ozao-Choy said at the annual Society of Surgical Oncology Cancer Symposium.

Among 44 patients treated with the CTLA-4 (cytotoxic T-lymphocyte antigen 4) inhibitor ipilimumab (Yervoy) and surgical resection, the 5-year melanoma-specific survival (MSS) rate was 51% and the median overall survival duration was 60 months, reported Dr. Ozao-Choy of the John Wayne Cancer Institute at Saint John’s Health Center in Santa Monica, Calif.

For 24 patients who received ipilimumab before resection, the 5-year MSS was 61% at a median of 60 months, and for 18 of 20 patients treated with ipilimumab after surgery, the 5-year MSS was 42% at a median of 47 months, but this difference was not significant (data were incomplete for 2 patients in the latter group), she noted.

In a recent study of retrospective data on patients with metastatic melanoma treated at her center, the 4-year survival of patients who underwent resection of metastatic lesions with or without systemic medical therapy was 20.8%, compared with 7% for those who underwent systemic medical therapy alone. The study investigators concluded that more than half of patients with metastatic melanoma were eligible for metastasectomy (Ann. Surg. Oncol. 2012;19:2547-55).

Dr. Ozao-Choy and her colleagues reviewed the center’s records on patients with metastatic melanoma who underwent resection and had received ipilimumab, looking at disease-specific survival from the date of diagnosis of stage IV disease.

The groups were well balanced in terms of age, sex, mean Breslow thickness scores, and nodal status. However, significantly more patients who received ipilimumab before surgery had brain metastases (13 of 24 vs. 3 of 18, P = .001). In a univariate analysis, patients with brain metastases had a significantly worse 5-year MSS (31% vs. 60%, P = .049).

The only other significant variables associated in the univariate analysis with better survival were prior immunotherapy, with 69% of patients who had received any immunotherapy having a 5-year MSS of 69%, compared with 29% for those with no immunotherapy (P = .01), and previous number of resections, with more resections being associated with better survival (P = .01).

Neither previous treatment with Bacillus Calmette-Guérin vaccine, previous chemotherapy, T stage, N stage, or timing of ipilimumab were significantly associated with MSS.

In a multivariate analysis (which controlled for demographic and disease factors), only the previous number of resections remained a significant predictor of MSS (P = .01).

In the audience response segment following the presentation, Dr. Daniel G. Coit, a surgical oncologist at Memorial Sloan-Kettering Cancer Center in New York City, pointed out that the previous number of resections is a not an adequate independent predictor for survival. "One of the inescapable truths is that if you have to have more than one operation, you have to still be alive. ... Of necessity, older people live longer than younger people; people who die at an older age live longer than people who die at a younger age," he said.

The study was internally funded. Dr. Ozao-Choy and Dr. Coit reported having no relevant financial disclosures.

NATIONAL HARBOR, MD. – Elevated levels of C-reactive protein may be a marker for poor response to palliative surgery, Dr. Andrew M. Blakely said at the annual Society of Surgical Oncology Cancer Symposium.

In a study of 50 patients who underwent elective palliative procedures for various cancer-related symptoms, elevated C-reactive protein (CRP), but not fatigue scores or performance status, was associated with a nearly fourfold risk for worse overall survival, reported Dr. Blakely, a surgery resident at Alpert Medical School, Brown University, Providence, R.I.

Elevated CRP, defined as levels greater than 8 mg/L, was also associated with increased overall and tumor-grade specific complications.

"This is important, because these complications are poorly tolerated by advanced cancer patients, and they serve to greatly diminish the potential benefit of palliative surgery," Dr. Blakely said.

He and colleagues at Brown and the Rhode Island Hospital, also in Providence, looked at potential prognostic factors in patients who chose surgery to relieve symptoms of advanced cancers such as gastrointestinal obstruction, or local-regional control of tumor-related pain or bleeding.

Malignancies treated included gastric, pancreatic, colon, melanoma, sarcomas, breast cancer, hepatobiliary and ovarian cancer, and squamous cell carcinomas.

They found that 44 patients (88%) stated that the surgery "was worth it," and 37 patients (74%) reported symptom resolution or improvement following surgery.

At 1-month follow-up, the complication rate was 44%, and 10% of patients had died. Although there was no association between CRP levels and morbidity or mortality among all patients, 11 of 27 patients with low-grade (grade 1 or 2) complications had elevated CRP (P = .001), as did all 6 patients with high-grade (grade 3 or 4) malignancies (P = .010), and 4 of 5 patients who died (P =.007).

In univariate analysis, factors significantly associated with major complications included elevated CRP (P = .005) and National Cancer Institute (NCI) fatigue score of 1 or greater (P = .001). Factors associated with overall survival were elevated CRP (P less than .0001), and NCI fatigue score (P = .04).

But in multivariate analysis controlling for those factors and for male sex, low albumin levels, significant weight loss, no previous cancer therapy, and hemoglobin less than 10.5 g/dL, only elevated CRP was associated with poor overall survival, with a hazard ratio of 3.6 (P = .0003).

The authors speculate that elevated CRP is a marker for systemic inflammation that may be associated with a higher risk for postoperative complications and for poorer overall survival among patients with advanced cancer.

"Even in this highly selected population, there was 44% morbidity and 10% mortality, illustrating the difficulty in patient selection," Dr. Blakely said.

CRP testing is a simple, low-cost, and objective test that can be easily added on to preoperative evaluation of candidates for palliative surgery, and it may be a useful element in patient selection and counseling, he said.

The study was internally funded. Dr. Blakely reported having no financial disclosures.

NATIONAL HARBOR, MD. – Elevated levels of C-reactive protein may be a marker for poor response to palliative surgery, Dr. Andrew M. Blakely said at the annual Society of Surgical Oncology Cancer Symposium.

In a study of 50 patients who underwent elective palliative procedures for various cancer-related symptoms, elevated C-reactive protein (CRP), but not fatigue scores or performance status, was associated with a nearly fourfold risk for worse overall survival, reported Dr. Blakely, a surgery resident at Alpert Medical School, Brown University, Providence, R.I.

Elevated CRP, defined as levels greater than 8 mg/L, was also associated with increased overall and tumor-grade specific complications.

"This is important, because these complications are poorly tolerated by advanced cancer patients, and they serve to greatly diminish the potential benefit of palliative surgery," Dr. Blakely said.

He and colleagues at Brown and the Rhode Island Hospital, also in Providence, looked at potential prognostic factors in patients who chose surgery to relieve symptoms of advanced cancers such as gastrointestinal obstruction, or local-regional control of tumor-related pain or bleeding.

Malignancies treated included gastric, pancreatic, colon, melanoma, sarcomas, breast cancer, hepatobiliary and ovarian cancer, and squamous cell carcinomas.

They found that 44 patients (88%) stated that the surgery "was worth it," and 37 patients (74%) reported symptom resolution or improvement following surgery.

At 1-month follow-up, the complication rate was 44%, and 10% of patients had died. Although there was no association between CRP levels and morbidity or mortality among all patients, 11 of 27 patients with low-grade (grade 1 or 2) complications had elevated CRP (P = .001), as did all 6 patients with high-grade (grade 3 or 4) malignancies (P = .010), and 4 of 5 patients who died (P =.007).

In univariate analysis, factors significantly associated with major complications included elevated CRP (P = .005) and National Cancer Institute (NCI) fatigue score of 1 or greater (P = .001). Factors associated with overall survival were elevated CRP (P less than .0001), and NCI fatigue score (P = .04).

But in multivariate analysis controlling for those factors and for male sex, low albumin levels, significant weight loss, no previous cancer therapy, and hemoglobin less than 10.5 g/dL, only elevated CRP was associated with poor overall survival, with a hazard ratio of 3.6 (P = .0003).

The authors speculate that elevated CRP is a marker for systemic inflammation that may be associated with a higher risk for postoperative complications and for poorer overall survival among patients with advanced cancer.

"Even in this highly selected population, there was 44% morbidity and 10% mortality, illustrating the difficulty in patient selection," Dr. Blakely said.

CRP testing is a simple, low-cost, and objective test that can be easily added on to preoperative evaluation of candidates for palliative surgery, and it may be a useful element in patient selection and counseling, he said.

The study was internally funded. Dr. Blakely reported having no financial disclosures.

NATIONAL HARBOR, MD. – Elevated levels of C-reactive protein may be a marker for poor response to palliative surgery, Dr. Andrew M. Blakely said at the annual Society of Surgical Oncology Cancer Symposium.

In a study of 50 patients who underwent elective palliative procedures for various cancer-related symptoms, elevated C-reactive protein (CRP), but not fatigue scores or performance status, was associated with a nearly fourfold risk for worse overall survival, reported Dr. Blakely, a surgery resident at Alpert Medical School, Brown University, Providence, R.I.

Elevated CRP, defined as levels greater than 8 mg/L, was also associated with increased overall and tumor-grade specific complications.

"This is important, because these complications are poorly tolerated by advanced cancer patients, and they serve to greatly diminish the potential benefit of palliative surgery," Dr. Blakely said.

He and colleagues at Brown and the Rhode Island Hospital, also in Providence, looked at potential prognostic factors in patients who chose surgery to relieve symptoms of advanced cancers such as gastrointestinal obstruction, or local-regional control of tumor-related pain or bleeding.

Malignancies treated included gastric, pancreatic, colon, melanoma, sarcomas, breast cancer, hepatobiliary and ovarian cancer, and squamous cell carcinomas.

They found that 44 patients (88%) stated that the surgery "was worth it," and 37 patients (74%) reported symptom resolution or improvement following surgery.

At 1-month follow-up, the complication rate was 44%, and 10% of patients had died. Although there was no association between CRP levels and morbidity or mortality among all patients, 11 of 27 patients with low-grade (grade 1 or 2) complications had elevated CRP (P = .001), as did all 6 patients with high-grade (grade 3 or 4) malignancies (P = .010), and 4 of 5 patients who died (P =.007).

In univariate analysis, factors significantly associated with major complications included elevated CRP (P = .005) and National Cancer Institute (NCI) fatigue score of 1 or greater (P = .001). Factors associated with overall survival were elevated CRP (P less than .0001), and NCI fatigue score (P = .04).

But in multivariate analysis controlling for those factors and for male sex, low albumin levels, significant weight loss, no previous cancer therapy, and hemoglobin less than 10.5 g/dL, only elevated CRP was associated with poor overall survival, with a hazard ratio of 3.6 (P = .0003).

The authors speculate that elevated CRP is a marker for systemic inflammation that may be associated with a higher risk for postoperative complications and for poorer overall survival among patients with advanced cancer.

"Even in this highly selected population, there was 44% morbidity and 10% mortality, illustrating the difficulty in patient selection," Dr. Blakely said.

CRP testing is a simple, low-cost, and objective test that can be easily added on to preoperative evaluation of candidates for palliative surgery, and it may be a useful element in patient selection and counseling, he said.

The study was internally funded. Dr. Blakely reported having no financial disclosures.

NATIONAL HARBOR, MD – Designer T cells used in immunotherapy for the treatment of liver metastases are well tolerated and reduced tumor burden in a small phase I study, reported an investigator at the annual Society of Surgical Oncology Cancer Symposium.

But the biggest barrier to the success of the therapy may lie in the liver itself, said Dr. Steven C. Katz, a surgical oncologist at the Roger Williams Cancer Center in Providence, R.I.

"Thinking long term, we’re going to have to address the suppressive forces that exist in the liver. I don’t think this is really going to succeed clinically until we find a way to do that," he said.

Genetically modified, designer T-cell technology allows translational science researchers to manufacture and deliver highly potent, specific agents to invoke a desired antitumor response, rather than trying to boost a general host immune response through, say, a vaccine, Dr. Katz said.

The technology has the potential to act against a variety of tumor types, and has been demonstrated to have activity against chronic lymphocytic leukemia, synovial sarcoma, and melanoma, he said.

The technique involves harvesting T cells from patients, activating them in culture, and then transfecting them with a retrovirus to get them to express a highly specific chimeric antigen receptor (CAR).

Dr. Katz and his colleagues are developing a CAR with an immunoglobulin fragment directed against carcinoembryonic antigen (CEA).

"A T cell that expresses this anti-CEA CAR, when it encounters a CEA-positive tumor, ... is activated to divide [and] produce cytokines and is poised to lyse and kill tumor cells in a highly specific fashion, and we demonstrated this in vitro, as have several other groups," he said.

But getting the antigen to the target – CEA-positive liver metastases – is just part of the challenge, he added. Delivering the CAR into the hepatic arterial circulation appears to be the optimal approach for confining the therapy to the target area, thus avoiding possible complications such as colitis that can occur when anti-CEA is delivered systemically.

A second barrier to the therapy is that the liver contains high numbers of suppressive immune cells such as regulatory T cells and myeloid-derived suppressor cells. Coinfusion of cytokines such as interleukin-2 (IL-2) appears to be a necessary step for preventing intrahepatic suppression of therapy, Dr. Katz said.

The investigators have initiated a phase I safety and intrapatient dose-escalation trial of the technology. Patients with CEA-positive liver metastases are given percutaneous infusions of their modified T cells through a femoral artery puncture directly into the hepatic artery. Embolization of the gastroduodenal and right gastric arteries is performed to prevent or limit extrahepatic infusion of the modified cells.

The patients all have unresectable, histologically confirmed CEA-positive liver metastases from colorectal, hepatobiliary, ampullary, or gastric primary tumors; have no extrahepatic disease beyond the lungs; and have had failure of one or more lines of standard systemic chemotherapy.

The first three patients have been treated in a dose-escalation fashion without cytokines. A second cohort of three patients is receiving both designer T cells and a continuous systemic infusion of IL-2 at a relatively low dose of 75,000 IU/kg per day to suppress the liver’s immune response.

"We know that IL-2 is going to be essential for these T cells to survive and do what we want them to do in vivo," Dr. Katz said.

Of the first six patients enrolled, two have been withdrawn from the study: one who had disease progression within the lungs following a second infusion of designer T cells, and one whose designer T cells were discovered to be infected with West Nile virus.

There have been no serious grade 3 or 4 adverse events to date. Adverse events have included grade 2 fevers, a grade 1 fever, and a rash. There has been no significant elevation over baseline in liver enzymes or bilirubin, Dr. Katz said.

Flow cytometry in early testing has shown that the modified T cells appear to be confined to the normal liver and metastatic sites, and have not been detected in significant numbers in peripheral circulation.

To date, only one patient has been treated with designer T cells and IL-2 support; this patient has thus far had a "very favorable" course, with a 40% decrease in tumor burden, although these results are still too early for clinical conclusions to be drawn.

The investigators are exploring additional strategies to circumvent intrahepatic suppression of the therapy by blocking the action of myeloid-derived suppressor cells.

The study was supported by a Society of Surgical Oncology Clinical Investigator Award, an education grant from Genentech, the National Institutes of Health, and the Rhode Island Foundation. Dr. Katz reported having no relevant financial disclosures.

NATIONAL HARBOR, MD – Designer T cells used in immunotherapy for the treatment of liver metastases are well tolerated and reduced tumor burden in a small phase I study, reported an investigator at the annual Society of Surgical Oncology Cancer Symposium.

But the biggest barrier to the success of the therapy may lie in the liver itself, said Dr. Steven C. Katz, a surgical oncologist at the Roger Williams Cancer Center in Providence, R.I.

"Thinking long term, we’re going to have to address the suppressive forces that exist in the liver. I don’t think this is really going to succeed clinically until we find a way to do that," he said.

Genetically modified, designer T-cell technology allows translational science researchers to manufacture and deliver highly potent, specific agents to invoke a desired antitumor response, rather than trying to boost a general host immune response through, say, a vaccine, Dr. Katz said.

The technology has the potential to act against a variety of tumor types, and has been demonstrated to have activity against chronic lymphocytic leukemia, synovial sarcoma, and melanoma, he said.

The technique involves harvesting T cells from patients, activating them in culture, and then transfecting them with a retrovirus to get them to express a highly specific chimeric antigen receptor (CAR).

Dr. Katz and his colleagues are developing a CAR with an immunoglobulin fragment directed against carcinoembryonic antigen (CEA).

"A T cell that expresses this anti-CEA CAR, when it encounters a CEA-positive tumor, ... is activated to divide [and] produce cytokines and is poised to lyse and kill tumor cells in a highly specific fashion, and we demonstrated this in vitro, as have several other groups," he said.

But getting the antigen to the target – CEA-positive liver metastases – is just part of the challenge, he added. Delivering the CAR into the hepatic arterial circulation appears to be the optimal approach for confining the therapy to the target area, thus avoiding possible complications such as colitis that can occur when anti-CEA is delivered systemically.

A second barrier to the therapy is that the liver contains high numbers of suppressive immune cells such as regulatory T cells and myeloid-derived suppressor cells. Coinfusion of cytokines such as interleukin-2 (IL-2) appears to be a necessary step for preventing intrahepatic suppression of therapy, Dr. Katz said.

The investigators have initiated a phase I safety and intrapatient dose-escalation trial of the technology. Patients with CEA-positive liver metastases are given percutaneous infusions of their modified T cells through a femoral artery puncture directly into the hepatic artery. Embolization of the gastroduodenal and right gastric arteries is performed to prevent or limit extrahepatic infusion of the modified cells.

The patients all have unresectable, histologically confirmed CEA-positive liver metastases from colorectal, hepatobiliary, ampullary, or gastric primary tumors; have no extrahepatic disease beyond the lungs; and have had failure of one or more lines of standard systemic chemotherapy.

The first three patients have been treated in a dose-escalation fashion without cytokines. A second cohort of three patients is receiving both designer T cells and a continuous systemic infusion of IL-2 at a relatively low dose of 75,000 IU/kg per day to suppress the liver’s immune response.

"We know that IL-2 is going to be essential for these T cells to survive and do what we want them to do in vivo," Dr. Katz said.

Of the first six patients enrolled, two have been withdrawn from the study: one who had disease progression within the lungs following a second infusion of designer T cells, and one whose designer T cells were discovered to be infected with West Nile virus.

There have been no serious grade 3 or 4 adverse events to date. Adverse events have included grade 2 fevers, a grade 1 fever, and a rash. There has been no significant elevation over baseline in liver enzymes or bilirubin, Dr. Katz said.

Flow cytometry in early testing has shown that the modified T cells appear to be confined to the normal liver and metastatic sites, and have not been detected in significant numbers in peripheral circulation.

To date, only one patient has been treated with designer T cells and IL-2 support; this patient has thus far had a "very favorable" course, with a 40% decrease in tumor burden, although these results are still too early for clinical conclusions to be drawn.

The investigators are exploring additional strategies to circumvent intrahepatic suppression of the therapy by blocking the action of myeloid-derived suppressor cells.

The study was supported by a Society of Surgical Oncology Clinical Investigator Award, an education grant from Genentech, the National Institutes of Health, and the Rhode Island Foundation. Dr. Katz reported having no relevant financial disclosures.

NATIONAL HARBOR, MD – Designer T cells used in immunotherapy for the treatment of liver metastases are well tolerated and reduced tumor burden in a small phase I study, reported an investigator at the annual Society of Surgical Oncology Cancer Symposium.

But the biggest barrier to the success of the therapy may lie in the liver itself, said Dr. Steven C. Katz, a surgical oncologist at the Roger Williams Cancer Center in Providence, R.I.

"Thinking long term, we’re going to have to address the suppressive forces that exist in the liver. I don’t think this is really going to succeed clinically until we find a way to do that," he said.

Genetically modified, designer T-cell technology allows translational science researchers to manufacture and deliver highly potent, specific agents to invoke a desired antitumor response, rather than trying to boost a general host immune response through, say, a vaccine, Dr. Katz said.

The technology has the potential to act against a variety of tumor types, and has been demonstrated to have activity against chronic lymphocytic leukemia, synovial sarcoma, and melanoma, he said.

The technique involves harvesting T cells from patients, activating them in culture, and then transfecting them with a retrovirus to get them to express a highly specific chimeric antigen receptor (CAR).

Dr. Katz and his colleagues are developing a CAR with an immunoglobulin fragment directed against carcinoembryonic antigen (CEA).

"A T cell that expresses this anti-CEA CAR, when it encounters a CEA-positive tumor, ... is activated to divide [and] produce cytokines and is poised to lyse and kill tumor cells in a highly specific fashion, and we demonstrated this in vitro, as have several other groups," he said.

But getting the antigen to the target – CEA-positive liver metastases – is just part of the challenge, he added. Delivering the CAR into the hepatic arterial circulation appears to be the optimal approach for confining the therapy to the target area, thus avoiding possible complications such as colitis that can occur when anti-CEA is delivered systemically.

A second barrier to the therapy is that the liver contains high numbers of suppressive immune cells such as regulatory T cells and myeloid-derived suppressor cells. Coinfusion of cytokines such as interleukin-2 (IL-2) appears to be a necessary step for preventing intrahepatic suppression of therapy, Dr. Katz said.

The investigators have initiated a phase I safety and intrapatient dose-escalation trial of the technology. Patients with CEA-positive liver metastases are given percutaneous infusions of their modified T cells through a femoral artery puncture directly into the hepatic artery. Embolization of the gastroduodenal and right gastric arteries is performed to prevent or limit extrahepatic infusion of the modified cells.

The patients all have unresectable, histologically confirmed CEA-positive liver metastases from colorectal, hepatobiliary, ampullary, or gastric primary tumors; have no extrahepatic disease beyond the lungs; and have had failure of one or more lines of standard systemic chemotherapy.

The first three patients have been treated in a dose-escalation fashion without cytokines. A second cohort of three patients is receiving both designer T cells and a continuous systemic infusion of IL-2 at a relatively low dose of 75,000 IU/kg per day to suppress the liver’s immune response.

"We know that IL-2 is going to be essential for these T cells to survive and do what we want them to do in vivo," Dr. Katz said.

Of the first six patients enrolled, two have been withdrawn from the study: one who had disease progression within the lungs following a second infusion of designer T cells, and one whose designer T cells were discovered to be infected with West Nile virus.

There have been no serious grade 3 or 4 adverse events to date. Adverse events have included grade 2 fevers, a grade 1 fever, and a rash. There has been no significant elevation over baseline in liver enzymes or bilirubin, Dr. Katz said.

Flow cytometry in early testing has shown that the modified T cells appear to be confined to the normal liver and metastatic sites, and have not been detected in significant numbers in peripheral circulation.

To date, only one patient has been treated with designer T cells and IL-2 support; this patient has thus far had a "very favorable" course, with a 40% decrease in tumor burden, although these results are still too early for clinical conclusions to be drawn.

The investigators are exploring additional strategies to circumvent intrahepatic suppression of the therapy by blocking the action of myeloid-derived suppressor cells.

The study was supported by a Society of Surgical Oncology Clinical Investigator Award, an education grant from Genentech, the National Institutes of Health, and the Rhode Island Foundation. Dr. Katz reported having no relevant financial disclosures.

NATIONAL HARBOR, MD. – It may sound like heresy, but select patients with locally advanced rectal cancer may be spared surgery and its associated complications, a cancer surgeon suggested at the annual Society of Surgical Oncology Cancer Symposium.

Approximately 10%-25% of patients with locally advanced rectal cancer will have clinical complete responses (cCR) to neoadjuvant chemotherapy and radiation, said Dr. Philip B. Paty, an attending surgeon in the colorectal surgery service at Memorial Sloan-Kettering Cancer Center in New York City.

"The vast majority of these patients will avoid rectal resection, at least within the first 5 years," Dr. Paty added.

Although local failure occurs in 10%-25% of patients, most of the failures occur within the first 18 months, and most of these cases can be salvaged with R0 resections. Patients treated with nonoperative management appear to have rates of distant recurrence and survival similar to those of patients with pathologic complete responses (pCR) treated with total mesorectal resection, he said.

If surgery is required, local excision may be sufficient for some patients with stage T1 lesions and a select few with T2 lesions, said Dr. Heidi Nelson, professor of surgery in the department of colon and rectal surgery at the Mayo Clinic in Rochester, Minn.

If a patient has a favorable T1 lesion and would otherwise face a life-altering procedure such as abdominal perineal resection (APR) and colostomy, the surgeon should at least show the patient the data and discuss local excision as a safe and effective alternative with results comparable to more extensive resections, she said.

T2 lesions are more problematic, but a select few patients with this tumor type might be spared the morbidity of standard rectal resection, she added.

Hold the surgery?

Dr. Paty noted that, with standard management of stage T3 or T4 rectal cancers, the combination of neoadjuvant chemoradiotherapy, surgery, and adjuvant chemotherapy resulted in a 76% overall survival rate with less than 0.3% local recurrence after 5 years (Ann. Surg. 2005;241:829-36).

"What we have not dwelt on much is the morbidity of surgery, which is very significant. Having a rectal resection is a life-changing event for every patient that has one. Surgeons know that, and patients know that even better than surgeons," he said.

Rectal resections are associated with significant perioperative morbidity, colostomy, altered bowel function, sexual dysfunction, and infertility, he noted.

Pathologic complete responses to neoadjuvant therapy occur in 10%-44% of patients, and patients who have a pCR have markedly better oncologic outcomes than patients with less robust responses.

Of course, pCR can only be determined after surgery, raising the question of whether a clinical CR is sufficient for determining whether a patient might be spared rectal resection.

There are currently more data on pCR in rectal cancer than cCR, "probably because clinical CR criteria right now are quite stringent; we don’t want to not operate on patients who have disease," Dr. Paty said.

His criteria for clinical complete response include a flat mucosa with no nodularity or mass on digital rectal examination. Smooth induration or minor scarring without nodularity is acceptable, "but it has to have a benign feel to it," he said.

In addition, on proctoscopy the mucosa must appear normal and flat, and if a scar is present it should be pale or white in appearance. Alternatively, there can be telangiectasias, he said.

"What’s not clear is whether ulceration is an exclusion criterion [for nonoperative management]. For me it is. Any time I see ulceration I feel there is something ongoing in that tumor that is not resolved and I don’t feel comfortable calling it a complete response," Dr. Paty commented.

Take local route

When surgery is required, local excision rather than total mesorectal resection may suffice, Dr. Nelson said. Suitable patients may be those who are frail or elderly or have limited life expectancy or serious medical conditions that might preclude more extensive surgery.

Tumors that may be good candidates for local excision include smaller lesions (less than 2-3 cm) below the peritoneal reflection that are not amenable to lower anterior resection. The tumors should be subject to full thickness excision, and the team should be able to confirm negative margins, she noted.

Favorable pathologic findings include well-differentiated tumors with the absence of lymphovascular invasion, mucinous features, or signet ring features, she said.

"Local excision really just takes care of the primary, of course. It doesn’t deal with the lymphatics, which is always the hidden game," Dr. Nelson said.

She noted that a 1989 study showed that the likelihood of untreated lymph node disease in patients who had undergone local excision was 0% for patients with T1 tumors, 28% for those with T2 tumors, 36% for T3, and 53% for T4 lesions, showing a significant increase in risk associated with tumor depth (Cancer 1989;63:1421-9).

"If you start tackling anything above a T2 lesion, you’re probably going to be missing lymphatic disease. It’s of relevance because it will form the site of recurrent disease," she said.

For patients with T1 tumors, overall survival is the same, but disease-free survival and local recurrence rates favor standard resection over local excision. "Selection criteria must be much more restrictive when it comes to a T2 lesion," Dr. Nelson said. "I’m pretty reticent to use it in my own practice. I have to really choose the tumor well and choose the patient well to want to do that with some assurance that it’s the right decision."

She pointed to a study published in 2000 comparing patients who underwent either local excision or standard resection for rectal cancer (Dis. Colon Rectum 2000;43:1064-71). Over about 4.5 years of follow-up, local recurrence for patients with T2 lesions was 47% if they had received local excision, compared with 6% for those who had standard resections. Respective overall survival rates were 65% and 81%.

Dr. Paty and Dr. Nelson reported having no financial disclosures.

NATIONAL HARBOR, MD. – It may sound like heresy, but select patients with locally advanced rectal cancer may be spared surgery and its associated complications, a cancer surgeon suggested at the annual Society of Surgical Oncology Cancer Symposium.

Approximately 10%-25% of patients with locally advanced rectal cancer will have clinical complete responses (cCR) to neoadjuvant chemotherapy and radiation, said Dr. Philip B. Paty, an attending surgeon in the colorectal surgery service at Memorial Sloan-Kettering Cancer Center in New York City.

"The vast majority of these patients will avoid rectal resection, at least within the first 5 years," Dr. Paty added.

Although local failure occurs in 10%-25% of patients, most of the failures occur within the first 18 months, and most of these cases can be salvaged with R0 resections. Patients treated with nonoperative management appear to have rates of distant recurrence and survival similar to those of patients with pathologic complete responses (pCR) treated with total mesorectal resection, he said.

If surgery is required, local excision may be sufficient for some patients with stage T1 lesions and a select few with T2 lesions, said Dr. Heidi Nelson, professor of surgery in the department of colon and rectal surgery at the Mayo Clinic in Rochester, Minn.

If a patient has a favorable T1 lesion and would otherwise face a life-altering procedure such as abdominal perineal resection (APR) and colostomy, the surgeon should at least show the patient the data and discuss local excision as a safe and effective alternative with results comparable to more extensive resections, she said.

T2 lesions are more problematic, but a select few patients with this tumor type might be spared the morbidity of standard rectal resection, she added.

Hold the surgery?

Dr. Paty noted that, with standard management of stage T3 or T4 rectal cancers, the combination of neoadjuvant chemoradiotherapy, surgery, and adjuvant chemotherapy resulted in a 76% overall survival rate with less than 0.3% local recurrence after 5 years (Ann. Surg. 2005;241:829-36).

"What we have not dwelt on much is the morbidity of surgery, which is very significant. Having a rectal resection is a life-changing event for every patient that has one. Surgeons know that, and patients know that even better than surgeons," he said.

Rectal resections are associated with significant perioperative morbidity, colostomy, altered bowel function, sexual dysfunction, and infertility, he noted.

Pathologic complete responses to neoadjuvant therapy occur in 10%-44% of patients, and patients who have a pCR have markedly better oncologic outcomes than patients with less robust responses.

Of course, pCR can only be determined after surgery, raising the question of whether a clinical CR is sufficient for determining whether a patient might be spared rectal resection.

There are currently more data on pCR in rectal cancer than cCR, "probably because clinical CR criteria right now are quite stringent; we don’t want to not operate on patients who have disease," Dr. Paty said.

His criteria for clinical complete response include a flat mucosa with no nodularity or mass on digital rectal examination. Smooth induration or minor scarring without nodularity is acceptable, "but it has to have a benign feel to it," he said.

In addition, on proctoscopy the mucosa must appear normal and flat, and if a scar is present it should be pale or white in appearance. Alternatively, there can be telangiectasias, he said.

"What’s not clear is whether ulceration is an exclusion criterion [for nonoperative management]. For me it is. Any time I see ulceration I feel there is something ongoing in that tumor that is not resolved and I don’t feel comfortable calling it a complete response," Dr. Paty commented.

Take local route

When surgery is required, local excision rather than total mesorectal resection may suffice, Dr. Nelson said. Suitable patients may be those who are frail or elderly or have limited life expectancy or serious medical conditions that might preclude more extensive surgery.

Tumors that may be good candidates for local excision include smaller lesions (less than 2-3 cm) below the peritoneal reflection that are not amenable to lower anterior resection. The tumors should be subject to full thickness excision, and the team should be able to confirm negative margins, she noted.

Favorable pathologic findings include well-differentiated tumors with the absence of lymphovascular invasion, mucinous features, or signet ring features, she said.

"Local excision really just takes care of the primary, of course. It doesn’t deal with the lymphatics, which is always the hidden game," Dr. Nelson said.

She noted that a 1989 study showed that the likelihood of untreated lymph node disease in patients who had undergone local excision was 0% for patients with T1 tumors, 28% for those with T2 tumors, 36% for T3, and 53% for T4 lesions, showing a significant increase in risk associated with tumor depth (Cancer 1989;63:1421-9).

"If you start tackling anything above a T2 lesion, you’re probably going to be missing lymphatic disease. It’s of relevance because it will form the site of recurrent disease," she said.

For patients with T1 tumors, overall survival is the same, but disease-free survival and local recurrence rates favor standard resection over local excision. "Selection criteria must be much more restrictive when it comes to a T2 lesion," Dr. Nelson said. "I’m pretty reticent to use it in my own practice. I have to really choose the tumor well and choose the patient well to want to do that with some assurance that it’s the right decision."

She pointed to a study published in 2000 comparing patients who underwent either local excision or standard resection for rectal cancer (Dis. Colon Rectum 2000;43:1064-71). Over about 4.5 years of follow-up, local recurrence for patients with T2 lesions was 47% if they had received local excision, compared with 6% for those who had standard resections. Respective overall survival rates were 65% and 81%.

Dr. Paty and Dr. Nelson reported having no financial disclosures.

NATIONAL HARBOR, MD. – It may sound like heresy, but select patients with locally advanced rectal cancer may be spared surgery and its associated complications, a cancer surgeon suggested at the annual Society of Surgical Oncology Cancer Symposium.

Approximately 10%-25% of patients with locally advanced rectal cancer will have clinical complete responses (cCR) to neoadjuvant chemotherapy and radiation, said Dr. Philip B. Paty, an attending surgeon in the colorectal surgery service at Memorial Sloan-Kettering Cancer Center in New York City.

"The vast majority of these patients will avoid rectal resection, at least within the first 5 years," Dr. Paty added.

Although local failure occurs in 10%-25% of patients, most of the failures occur within the first 18 months, and most of these cases can be salvaged with R0 resections. Patients treated with nonoperative management appear to have rates of distant recurrence and survival similar to those of patients with pathologic complete responses (pCR) treated with total mesorectal resection, he said.

If surgery is required, local excision may be sufficient for some patients with stage T1 lesions and a select few with T2 lesions, said Dr. Heidi Nelson, professor of surgery in the department of colon and rectal surgery at the Mayo Clinic in Rochester, Minn.

If a patient has a favorable T1 lesion and would otherwise face a life-altering procedure such as abdominal perineal resection (APR) and colostomy, the surgeon should at least show the patient the data and discuss local excision as a safe and effective alternative with results comparable to more extensive resections, she said.

T2 lesions are more problematic, but a select few patients with this tumor type might be spared the morbidity of standard rectal resection, she added.

Hold the surgery?

Dr. Paty noted that, with standard management of stage T3 or T4 rectal cancers, the combination of neoadjuvant chemoradiotherapy, surgery, and adjuvant chemotherapy resulted in a 76% overall survival rate with less than 0.3% local recurrence after 5 years (Ann. Surg. 2005;241:829-36).

"What we have not dwelt on much is the morbidity of surgery, which is very significant. Having a rectal resection is a life-changing event for every patient that has one. Surgeons know that, and patients know that even better than surgeons," he said.

Rectal resections are associated with significant perioperative morbidity, colostomy, altered bowel function, sexual dysfunction, and infertility, he noted.

Pathologic complete responses to neoadjuvant therapy occur in 10%-44% of patients, and patients who have a pCR have markedly better oncologic outcomes than patients with less robust responses.

Of course, pCR can only be determined after surgery, raising the question of whether a clinical CR is sufficient for determining whether a patient might be spared rectal resection.

There are currently more data on pCR in rectal cancer than cCR, "probably because clinical CR criteria right now are quite stringent; we don’t want to not operate on patients who have disease," Dr. Paty said.

His criteria for clinical complete response include a flat mucosa with no nodularity or mass on digital rectal examination. Smooth induration or minor scarring without nodularity is acceptable, "but it has to have a benign feel to it," he said.

In addition, on proctoscopy the mucosa must appear normal and flat, and if a scar is present it should be pale or white in appearance. Alternatively, there can be telangiectasias, he said.

"What’s not clear is whether ulceration is an exclusion criterion [for nonoperative management]. For me it is. Any time I see ulceration I feel there is something ongoing in that tumor that is not resolved and I don’t feel comfortable calling it a complete response," Dr. Paty commented.

Take local route

When surgery is required, local excision rather than total mesorectal resection may suffice, Dr. Nelson said. Suitable patients may be those who are frail or elderly or have limited life expectancy or serious medical conditions that might preclude more extensive surgery.

Tumors that may be good candidates for local excision include smaller lesions (less than 2-3 cm) below the peritoneal reflection that are not amenable to lower anterior resection. The tumors should be subject to full thickness excision, and the team should be able to confirm negative margins, she noted.

Favorable pathologic findings include well-differentiated tumors with the absence of lymphovascular invasion, mucinous features, or signet ring features, she said.

"Local excision really just takes care of the primary, of course. It doesn’t deal with the lymphatics, which is always the hidden game," Dr. Nelson said.

She noted that a 1989 study showed that the likelihood of untreated lymph node disease in patients who had undergone local excision was 0% for patients with T1 tumors, 28% for those with T2 tumors, 36% for T3, and 53% for T4 lesions, showing a significant increase in risk associated with tumor depth (Cancer 1989;63:1421-9).

"If you start tackling anything above a T2 lesion, you’re probably going to be missing lymphatic disease. It’s of relevance because it will form the site of recurrent disease," she said.

For patients with T1 tumors, overall survival is the same, but disease-free survival and local recurrence rates favor standard resection over local excision. "Selection criteria must be much more restrictive when it comes to a T2 lesion," Dr. Nelson said. "I’m pretty reticent to use it in my own practice. I have to really choose the tumor well and choose the patient well to want to do that with some assurance that it’s the right decision."

She pointed to a study published in 2000 comparing patients who underwent either local excision or standard resection for rectal cancer (Dis. Colon Rectum 2000;43:1064-71). Over about 4.5 years of follow-up, local recurrence for patients with T2 lesions was 47% if they had received local excision, compared with 6% for those who had standard resections. Respective overall survival rates were 65% and 81%.

Dr. Paty and Dr. Nelson reported having no financial disclosures.

NATIONAL HARBOR, MD – Seeing is believing, especially when enhanced visualization of tumors during surgery helps improve chances for complete resection, investigators said at the annual Society of Surgical Oncology Cancer Symposium.

With near-infrared (NIR) fluorescence imaging and a portable camera that can be used in an operating room, surgeons can eliminate some of the guesswork involved in identifying involved surgical margins or lymph nodes, researchers from the U.S. and the Netherlands reported in oral and poster sessions.

Courtesy Dr. Sunil Singhal

In early human trials, a small, portable infrared camera has been successful at identifying dye-impregnated tumors – including noncontiguous pockets of malignancy – during surgery to resect squamous cell carcinomas and adenocarcinomas of the lung, reported Dr. Sunil Singhal, of the department of surgery at the University of Pennsylvania, Philadelphia.

"Even in this day and age, surgeons leave behind disease in 40% of the cases, and in about a quarter of those cases the tumor was within two centimeters of where the surgeon was working," Dr. Singhal said.

To improve the odds, he and his colleagues have been investigating optical contrast agents that can be delivered safely to tumors and cause them to fluoresce under light in the near-infrared portion of the spectrum. In preclinical studies with dogs, they found that indocyanine green had the right combination of toxicity, photostability, pharmacokinetics, and cost. The dye, currently used in retinal angiography, has an emission profile that makes it easy for observers to discriminate between the fluorescing dye and blood or tissues, Dr. Singhal said.

They also developed an intraoperative device, dubbed the "FloCam," which consists of a light source and near-infrared (NIR) camera that sits above the patient and sends images to a computer monitor showing the operation in NIR.

In animal studies, the system found evidence of residual disease that was not visible to the naked eye or on x-ray microtomography. On pathologic examination, they saw that the dye was "remarkably precise in delineating margins from normal surrounding tissues," particularly in tumors with neovascular features.

Dr. Singhal said that the imaging technique has been effective at identifying tumor sites during surgery in 36 of 38 patients in early human trials, failing only for 1 patient with melanoma, and for 1 with a sarcoma.

One patient was a 64-year-old nonsmoking man who presented with a cough and was found to have a 2.5 cm right upper lobe lung tumor. Evaluation of the mediastinum with imaging and pathology samples was negative for malignancy, but during surgery, the dye highlighted previously undetected tumor hotspots in the right lower lobe.

"This is another patient who would have gone home [with a diagnosis of] stage 1A. I would have walked down to the recovery room, say ‘I cured you,’ and he would come back 1 year later with metastatic disease and die. This patient, who had minimal disease when we discovered it, got chemotherapy and is still alive at the 1-year mark," Dr. Singhal said.

Green hybrid

In a separate study, investigators in the Netherlands reported on improved intraoperative sentinel node identification and harvesting using a novel hybrid radiopharmaceutical tracer combining indocyanine green with technetium-99m in a nanocolloid suspension.

They found that in 96 patients with malignant melanomas of the head and neck, trunk, or extremities, the hybrid tracer, facilitated both preoperative SPECT/CT imaging and intraoperative radio- and fluorescence-guide sentinel node biopsy in all patients.

"The hybrid tracer was found to be particularly useful for the detection of sentinel nodes in the neck, and for sentinel nodes that failed to accumulate patent blue dye," wrote Dr. Oscar R. Brouwer from the division of nuclear medicine at the Netherlands Cancer Institute in Amsterdam, and colleagues in a scientific poster.

Dr. Singhal’s studies were supported by the Society of Surgical Oncology and the University of Pennsylvania. He reported having no financial disclosures. Dr. Brouwer’s study was supported by the Netherlands Cancer Institute. He reported having no financial disclosures.

NATIONAL HARBOR, MD – Seeing is believing, especially when enhanced visualization of tumors during surgery helps improve chances for complete resection, investigators said at the annual Society of Surgical Oncology Cancer Symposium.

With near-infrared (NIR) fluorescence imaging and a portable camera that can be used in an operating room, surgeons can eliminate some of the guesswork involved in identifying involved surgical margins or lymph nodes, researchers from the U.S. and the Netherlands reported in oral and poster sessions.

Courtesy Dr. Sunil Singhal

In early human trials, a small, portable infrared camera has been successful at identifying dye-impregnated tumors – including noncontiguous pockets of malignancy – during surgery to resect squamous cell carcinomas and adenocarcinomas of the lung, reported Dr. Sunil Singhal, of the department of surgery at the University of Pennsylvania, Philadelphia.

"Even in this day and age, surgeons leave behind disease in 40% of the cases, and in about a quarter of those cases the tumor was within two centimeters of where the surgeon was working," Dr. Singhal said.

To improve the odds, he and his colleagues have been investigating optical contrast agents that can be delivered safely to tumors and cause them to fluoresce under light in the near-infrared portion of the spectrum. In preclinical studies with dogs, they found that indocyanine green had the right combination of toxicity, photostability, pharmacokinetics, and cost. The dye, currently used in retinal angiography, has an emission profile that makes it easy for observers to discriminate between the fluorescing dye and blood or tissues, Dr. Singhal said.

They also developed an intraoperative device, dubbed the "FloCam," which consists of a light source and near-infrared (NIR) camera that sits above the patient and sends images to a computer monitor showing the operation in NIR.

In animal studies, the system found evidence of residual disease that was not visible to the naked eye or on x-ray microtomography. On pathologic examination, they saw that the dye was "remarkably precise in delineating margins from normal surrounding tissues," particularly in tumors with neovascular features.

Dr. Singhal said that the imaging technique has been effective at identifying tumor sites during surgery in 36 of 38 patients in early human trials, failing only for 1 patient with melanoma, and for 1 with a sarcoma.

One patient was a 64-year-old nonsmoking man who presented with a cough and was found to have a 2.5 cm right upper lobe lung tumor. Evaluation of the mediastinum with imaging and pathology samples was negative for malignancy, but during surgery, the dye highlighted previously undetected tumor hotspots in the right lower lobe.

"This is another patient who would have gone home [with a diagnosis of] stage 1A. I would have walked down to the recovery room, say ‘I cured you,’ and he would come back 1 year later with metastatic disease and die. This patient, who had minimal disease when we discovered it, got chemotherapy and is still alive at the 1-year mark," Dr. Singhal said.

Green hybrid

In a separate study, investigators in the Netherlands reported on improved intraoperative sentinel node identification and harvesting using a novel hybrid radiopharmaceutical tracer combining indocyanine green with technetium-99m in a nanocolloid suspension.

They found that in 96 patients with malignant melanomas of the head and neck, trunk, or extremities, the hybrid tracer, facilitated both preoperative SPECT/CT imaging and intraoperative radio- and fluorescence-guide sentinel node biopsy in all patients.

"The hybrid tracer was found to be particularly useful for the detection of sentinel nodes in the neck, and for sentinel nodes that failed to accumulate patent blue dye," wrote Dr. Oscar R. Brouwer from the division of nuclear medicine at the Netherlands Cancer Institute in Amsterdam, and colleagues in a scientific poster.

Dr. Singhal’s studies were supported by the Society of Surgical Oncology and the University of Pennsylvania. He reported having no financial disclosures. Dr. Brouwer’s study was supported by the Netherlands Cancer Institute. He reported having no financial disclosures.

NATIONAL HARBOR, MD – Seeing is believing, especially when enhanced visualization of tumors during surgery helps improve chances for complete resection, investigators said at the annual Society of Surgical Oncology Cancer Symposium.

With near-infrared (NIR) fluorescence imaging and a portable camera that can be used in an operating room, surgeons can eliminate some of the guesswork involved in identifying involved surgical margins or lymph nodes, researchers from the U.S. and the Netherlands reported in oral and poster sessions.

Courtesy Dr. Sunil Singhal

In early human trials, a small, portable infrared camera has been successful at identifying dye-impregnated tumors – including noncontiguous pockets of malignancy – during surgery to resect squamous cell carcinomas and adenocarcinomas of the lung, reported Dr. Sunil Singhal, of the department of surgery at the University of Pennsylvania, Philadelphia.

"Even in this day and age, surgeons leave behind disease in 40% of the cases, and in about a quarter of those cases the tumor was within two centimeters of where the surgeon was working," Dr. Singhal said.

To improve the odds, he and his colleagues have been investigating optical contrast agents that can be delivered safely to tumors and cause them to fluoresce under light in the near-infrared portion of the spectrum. In preclinical studies with dogs, they found that indocyanine green had the right combination of toxicity, photostability, pharmacokinetics, and cost. The dye, currently used in retinal angiography, has an emission profile that makes it easy for observers to discriminate between the fluorescing dye and blood or tissues, Dr. Singhal said.

They also developed an intraoperative device, dubbed the "FloCam," which consists of a light source and near-infrared (NIR) camera that sits above the patient and sends images to a computer monitor showing the operation in NIR.

In animal studies, the system found evidence of residual disease that was not visible to the naked eye or on x-ray microtomography. On pathologic examination, they saw that the dye was "remarkably precise in delineating margins from normal surrounding tissues," particularly in tumors with neovascular features.

Dr. Singhal said that the imaging technique has been effective at identifying tumor sites during surgery in 36 of 38 patients in early human trials, failing only for 1 patient with melanoma, and for 1 with a sarcoma.

One patient was a 64-year-old nonsmoking man who presented with a cough and was found to have a 2.5 cm right upper lobe lung tumor. Evaluation of the mediastinum with imaging and pathology samples was negative for malignancy, but during surgery, the dye highlighted previously undetected tumor hotspots in the right lower lobe.

"This is another patient who would have gone home [with a diagnosis of] stage 1A. I would have walked down to the recovery room, say ‘I cured you,’ and he would come back 1 year later with metastatic disease and die. This patient, who had minimal disease when we discovered it, got chemotherapy and is still alive at the 1-year mark," Dr. Singhal said.

Green hybrid

In a separate study, investigators in the Netherlands reported on improved intraoperative sentinel node identification and harvesting using a novel hybrid radiopharmaceutical tracer combining indocyanine green with technetium-99m in a nanocolloid suspension.

They found that in 96 patients with malignant melanomas of the head and neck, trunk, or extremities, the hybrid tracer, facilitated both preoperative SPECT/CT imaging and intraoperative radio- and fluorescence-guide sentinel node biopsy in all patients.

"The hybrid tracer was found to be particularly useful for the detection of sentinel nodes in the neck, and for sentinel nodes that failed to accumulate patent blue dye," wrote Dr. Oscar R. Brouwer from the division of nuclear medicine at the Netherlands Cancer Institute in Amsterdam, and colleagues in a scientific poster.

Dr. Singhal’s studies were supported by the Society of Surgical Oncology and the University of Pennsylvania. He reported having no financial disclosures. Dr. Brouwer’s study was supported by the Netherlands Cancer Institute. He reported having no financial disclosures.

NATIONAL HARBOR, MD. – The rate of contralateral prophylactic mastectomies is rising, even though there is no evidence for a survival benefit.

From 1998 through 2007, contralateral prophylactic mastectomies (CPM) were performed within 1 year of unilateral mastectomies in 21% of those with ductal carcinoma in situ (DCIS) and in 17% of those with stage I-III breast cancer who were treated at one of 10 National Comprehensive Cancer Network (NCCN) centers.

But in an analysis of overall survival for patients with stages I-III invasive breast cancer, there was so significant difference in overall survival for patients who underwent a CPM, compared with those who underwent only unilateral mastectomy, regardless of whether they had received neoadjuvant chemotherapy, reported Dr. William E. Carson III, professor of surgery at the Ohio State University Comprehensive Cancer Center in Columbus.

In addition, CPMs are associated with a significantly greater risk of complications than unilateral mastectomies, including increased risk for major complications requiring reoperation and rehospitalization, said Dr. Megan Miller, a surgery resident at the University of Chicago, Illinois.

"CPM patients are 1.5 times more likely to have any complication, and 2.6 times more likely to have a major complication than unilateral mastectomy patients," she said at the annual Society of Surgical Oncology Cancer Symposium.

Among patients who underwent CPM, almost 40% of the complications occurred on the side of the body without cancer, she noted.

SSO position statement

A 2007 position statement from the Society of Surgical Oncology (SSO) states that in patients with a current or prior diagnosis, CPM may be indicated for risk reduction in cases where surveillance is difficult or for reconstructive issues such as symmetry and balance, Dr. Carson noted.

Two studies using Surveillance, Epidemiology, and End Results (SEER) data (Tuttle et al. [J. Clin. Oncology 2009;27:1362-7]); Bedrosian et al. [J. Natl. Cancer Inst. 2010;102:401-9]) and one from his own center (Jones et al. [Ann. Surg. Oncol. 2009;16:2691-6]) showed about a 10% increase in the rate of CPM over a decade. Younger women with higher levels of education were more likely to seek CPM.

To see whether this trend extended to NCCN centers, Dr. Carson and his colleagues reviewed data on 1,309 women with DCIS, and 7,044 with stage I-III breast cancer who underwent unilateral mastectomy from 1998 through 2007 at one of 10 designated centers.

In all, 273 of the women diagnosed with DCIS (21%) had a contralateral prophylactic mastectomy, as did 1,199 (17%) of the women with a diagnosis of stage I-III invasive disease. Median follow-up was more than 4 years for both groups.

In a multivariate analysis, factors that significantly predicted the likelihood of CPM included age younger than 50 years, Caucasian race, MRI as the method of detection, and tumor size of 1 cm or smaller. In women with invasive disease, years of education, node-negative status, and no immediate reconstruction were also significant predictors of CPM (P less than .0001 for all variables).

Use of CPM varied widely by institution from 8.2%-34.7% of women with DCIS, and from 3.6%-30.8% of patients with stage I-III disease. As other studies have shown, the use of CPM increased over time, from 15% for DCIS in 1998 to 27% in 2007. For patients with invasive breast cancer, the respective increase was from 8% to 26%. The most pronounced increases were among patients younger than 50 years, Dr. Carson noted.

When they looked at overall survival in a multivariate Cox regression model adjusted for age, race, tumor size, nodal status, tumor grade, histology, and treatment, they found that there was no significant survival advantage for unilateral mastectomy plus CPM, compared with unilateral mastectomy alone.

Complications, complications

Dr. Miller and her colleagues retrospectively reviewed 600 patients who underwent either unilateral mastectomy (391) or CPM (209) at their center from January 2009 through March 2012. They looked at major complications such as seroma or hematoma requiring reoperations, infections requiring hospital admission, total nipple or flap necrosis, and bleeding requiring transfusion; and minor complications such as seromas and hematomas requiring aspiration, infections requiring oral antibiotics, partial nipple or flap necrosis, minor bleeding, and delayed wound healing.

The percentage of patients experiencing any complications was 29% for patients who had a unilateral mastectomy, compared with 42% of those who underwent CPM (P less than .001). Major complications occurred in 4.1% and 14%, respectively (P less than .001). Rates of minor complications were identical between the groups, at 15% each.

Multiple major complications were seen in 4.9% of unilateral patients, compared with 9.1% of CPM patients (P = .043).

Among the CPM patients, 40% of complications occurred on the CPM side.

In a multivariate analysis controlling for age, body mass index, diabetes, previous radiation, smoking history and reconstruction type, CPM was associated with an odds ratio for any complication of 1.5 (P = .029) and 2.6 for major complications (P = .007).

"We believe that patients considering CPM should be made aware of these risks, and certainly more research is needed on patient decision pathways and shared decision making," Dr. Miller said.

Both Dr. Carson’s and Dr. Miller’s studies were internally funded. Dr. Carson disclosed serving on the NCCN Board of Directors. Dr. Miller reported having no financial disclosures.

NATIONAL HARBOR, MD. – The rate of contralateral prophylactic mastectomies is rising, even though there is no evidence for a survival benefit.

From 1998 through 2007, contralateral prophylactic mastectomies (CPM) were performed within 1 year of unilateral mastectomies in 21% of those with ductal carcinoma in situ (DCIS) and in 17% of those with stage I-III breast cancer who were treated at one of 10 National Comprehensive Cancer Network (NCCN) centers.

But in an analysis of overall survival for patients with stages I-III invasive breast cancer, there was so significant difference in overall survival for patients who underwent a CPM, compared with those who underwent only unilateral mastectomy, regardless of whether they had received neoadjuvant chemotherapy, reported Dr. William E. Carson III, professor of surgery at the Ohio State University Comprehensive Cancer Center in Columbus.

In addition, CPMs are associated with a significantly greater risk of complications than unilateral mastectomies, including increased risk for major complications requiring reoperation and rehospitalization, said Dr. Megan Miller, a surgery resident at the University of Chicago, Illinois.

"CPM patients are 1.5 times more likely to have any complication, and 2.6 times more likely to have a major complication than unilateral mastectomy patients," she said at the annual Society of Surgical Oncology Cancer Symposium.

Among patients who underwent CPM, almost 40% of the complications occurred on the side of the body without cancer, she noted.

SSO position statement

A 2007 position statement from the Society of Surgical Oncology (SSO) states that in patients with a current or prior diagnosis, CPM may be indicated for risk reduction in cases where surveillance is difficult or for reconstructive issues such as symmetry and balance, Dr. Carson noted.

Two studies using Surveillance, Epidemiology, and End Results (SEER) data (Tuttle et al. [J. Clin. Oncology 2009;27:1362-7]); Bedrosian et al. [J. Natl. Cancer Inst. 2010;102:401-9]) and one from his own center (Jones et al. [Ann. Surg. Oncol. 2009;16:2691-6]) showed about a 10% increase in the rate of CPM over a decade. Younger women with higher levels of education were more likely to seek CPM.

To see whether this trend extended to NCCN centers, Dr. Carson and his colleagues reviewed data on 1,309 women with DCIS, and 7,044 with stage I-III breast cancer who underwent unilateral mastectomy from 1998 through 2007 at one of 10 designated centers.

In all, 273 of the women diagnosed with DCIS (21%) had a contralateral prophylactic mastectomy, as did 1,199 (17%) of the women with a diagnosis of stage I-III invasive disease. Median follow-up was more than 4 years for both groups.

In a multivariate analysis, factors that significantly predicted the likelihood of CPM included age younger than 50 years, Caucasian race, MRI as the method of detection, and tumor size of 1 cm or smaller. In women with invasive disease, years of education, node-negative status, and no immediate reconstruction were also significant predictors of CPM (P less than .0001 for all variables).

Use of CPM varied widely by institution from 8.2%-34.7% of women with DCIS, and from 3.6%-30.8% of patients with stage I-III disease. As other studies have shown, the use of CPM increased over time, from 15% for DCIS in 1998 to 27% in 2007. For patients with invasive breast cancer, the respective increase was from 8% to 26%. The most pronounced increases were among patients younger than 50 years, Dr. Carson noted.

When they looked at overall survival in a multivariate Cox regression model adjusted for age, race, tumor size, nodal status, tumor grade, histology, and treatment, they found that there was no significant survival advantage for unilateral mastectomy plus CPM, compared with unilateral mastectomy alone.

Complications, complications

Dr. Miller and her colleagues retrospectively reviewed 600 patients who underwent either unilateral mastectomy (391) or CPM (209) at their center from January 2009 through March 2012. They looked at major complications such as seroma or hematoma requiring reoperations, infections requiring hospital admission, total nipple or flap necrosis, and bleeding requiring transfusion; and minor complications such as seromas and hematomas requiring aspiration, infections requiring oral antibiotics, partial nipple or flap necrosis, minor bleeding, and delayed wound healing.

The percentage of patients experiencing any complications was 29% for patients who had a unilateral mastectomy, compared with 42% of those who underwent CPM (P less than .001). Major complications occurred in 4.1% and 14%, respectively (P less than .001). Rates of minor complications were identical between the groups, at 15% each.

Multiple major complications were seen in 4.9% of unilateral patients, compared with 9.1% of CPM patients (P = .043).

Among the CPM patients, 40% of complications occurred on the CPM side.

In a multivariate analysis controlling for age, body mass index, diabetes, previous radiation, smoking history and reconstruction type, CPM was associated with an odds ratio for any complication of 1.5 (P = .029) and 2.6 for major complications (P = .007).

"We believe that patients considering CPM should be made aware of these risks, and certainly more research is needed on patient decision pathways and shared decision making," Dr. Miller said.

Both Dr. Carson’s and Dr. Miller’s studies were internally funded. Dr. Carson disclosed serving on the NCCN Board of Directors. Dr. Miller reported having no financial disclosures.

NATIONAL HARBOR, MD. – The rate of contralateral prophylactic mastectomies is rising, even though there is no evidence for a survival benefit.

From 1998 through 2007, contralateral prophylactic mastectomies (CPM) were performed within 1 year of unilateral mastectomies in 21% of those with ductal carcinoma in situ (DCIS) and in 17% of those with stage I-III breast cancer who were treated at one of 10 National Comprehensive Cancer Network (NCCN) centers.

But in an analysis of overall survival for patients with stages I-III invasive breast cancer, there was so significant difference in overall survival for patients who underwent a CPM, compared with those who underwent only unilateral mastectomy, regardless of whether they had received neoadjuvant chemotherapy, reported Dr. William E. Carson III, professor of surgery at the Ohio State University Comprehensive Cancer Center in Columbus.

In addition, CPMs are associated with a significantly greater risk of complications than unilateral mastectomies, including increased risk for major complications requiring reoperation and rehospitalization, said Dr. Megan Miller, a surgery resident at the University of Chicago, Illinois.

"CPM patients are 1.5 times more likely to have any complication, and 2.6 times more likely to have a major complication than unilateral mastectomy patients," she said at the annual Society of Surgical Oncology Cancer Symposium.

Among patients who underwent CPM, almost 40% of the complications occurred on the side of the body without cancer, she noted.

SSO position statement

A 2007 position statement from the Society of Surgical Oncology (SSO) states that in patients with a current or prior diagnosis, CPM may be indicated for risk reduction in cases where surveillance is difficult or for reconstructive issues such as symmetry and balance, Dr. Carson noted.

Two studies using Surveillance, Epidemiology, and End Results (SEER) data (Tuttle et al. [J. Clin. Oncology 2009;27:1362-7]); Bedrosian et al. [J. Natl. Cancer Inst. 2010;102:401-9]) and one from his own center (Jones et al. [Ann. Surg. Oncol. 2009;16:2691-6]) showed about a 10% increase in the rate of CPM over a decade. Younger women with higher levels of education were more likely to seek CPM.

To see whether this trend extended to NCCN centers, Dr. Carson and his colleagues reviewed data on 1,309 women with DCIS, and 7,044 with stage I-III breast cancer who underwent unilateral mastectomy from 1998 through 2007 at one of 10 designated centers.

In all, 273 of the women diagnosed with DCIS (21%) had a contralateral prophylactic mastectomy, as did 1,199 (17%) of the women with a diagnosis of stage I-III invasive disease. Median follow-up was more than 4 years for both groups.

In a multivariate analysis, factors that significantly predicted the likelihood of CPM included age younger than 50 years, Caucasian race, MRI as the method of detection, and tumor size of 1 cm or smaller. In women with invasive disease, years of education, node-negative status, and no immediate reconstruction were also significant predictors of CPM (P less than .0001 for all variables).

Use of CPM varied widely by institution from 8.2%-34.7% of women with DCIS, and from 3.6%-30.8% of patients with stage I-III disease. As other studies have shown, the use of CPM increased over time, from 15% for DCIS in 1998 to 27% in 2007. For patients with invasive breast cancer, the respective increase was from 8% to 26%. The most pronounced increases were among patients younger than 50 years, Dr. Carson noted.

When they looked at overall survival in a multivariate Cox regression model adjusted for age, race, tumor size, nodal status, tumor grade, histology, and treatment, they found that there was no significant survival advantage for unilateral mastectomy plus CPM, compared with unilateral mastectomy alone.

Complications, complications

Dr. Miller and her colleagues retrospectively reviewed 600 patients who underwent either unilateral mastectomy (391) or CPM (209) at their center from January 2009 through March 2012. They looked at major complications such as seroma or hematoma requiring reoperations, infections requiring hospital admission, total nipple or flap necrosis, and bleeding requiring transfusion; and minor complications such as seromas and hematomas requiring aspiration, infections requiring oral antibiotics, partial nipple or flap necrosis, minor bleeding, and delayed wound healing.

The percentage of patients experiencing any complications was 29% for patients who had a unilateral mastectomy, compared with 42% of those who underwent CPM (P less than .001). Major complications occurred in 4.1% and 14%, respectively (P less than .001). Rates of minor complications were identical between the groups, at 15% each.

Multiple major complications were seen in 4.9% of unilateral patients, compared with 9.1% of CPM patients (P = .043).

Among the CPM patients, 40% of complications occurred on the CPM side.

In a multivariate analysis controlling for age, body mass index, diabetes, previous radiation, smoking history and reconstruction type, CPM was associated with an odds ratio for any complication of 1.5 (P = .029) and 2.6 for major complications (P = .007).

"We believe that patients considering CPM should be made aware of these risks, and certainly more research is needed on patient decision pathways and shared decision making," Dr. Miller said.

Both Dr. Carson’s and Dr. Miller’s studies were internally funded. Dr. Carson disclosed serving on the NCCN Board of Directors. Dr. Miller reported having no financial disclosures.