PalliOnc Symposium

BOSTON – Medical residents in the United States appear to understand that cardiopulmonary resuscitation or intubation is highly unlikely to benefit patients with advanced cancers at the end of life, but the majority of residents surveyed said that they do not discuss code-status options or potentially beneficial palliative care with their dying patients.

“This was primarily due to residents’ perceptions of patient autonomy: Residents wanted patients to make their own decisions, without any influence from the doctor, which misses the concept of informed decision making. These incomplete discussions can cause at minimum improper documentation of patients’ wishes, and at most psychological harm, damage to the physician-patient relationship, and the potential for unwanted attempts at resuscitation,” said Dr. David J. Einstein, a resident at Beth Israel Deaconess Medical Center and Tufts Medical Center, both in Boston.

Despite their reluctance to have the discussion, however, the majority of residents said they preferred to discuss code status with patients themselves rather than hand it off to the attending physician, primarily out of a sense that it is their responsibility as physicians. Yet these physicians in training did not seem to feel that they were also responsible for providing guidance to patients, Dr. Einstein said at the Palliative Care in Oncology Symposium.

“We felt that this represented an unmet need in training and practice. Residents and attendings should be providing guidance on all medical interventions, including CPR, and if they aren’t sure what to recommend, then they themselves should be seeking guidance from other experts, before asking a patient to falsely choose between an intervention and death,” he said.

The first discussion of code status – do not resuscitate (DNR) or do not intubate (DNI) – may occur in the hospital, and is often left to a resident physician. Ideally, the physician and patient should discuss the patient’s prognosis, goals for care, evaluation of CPR as a means of meeting those goals, and a recommendation. But many residents lack training in the end-of-life discussion, which can have a significant impact on the quality of the patients’ remaining weeks or months of life.

Nationwide survey

Dr. Einstein and his colleagues conducted a nationwide survey to measure the likelihood that residents would discuss prognostic information and offer recommendations to patients with limited life expectancy. They also sought to determine why residents might be reluctant to provide discussion, and to evaluate their satisfaction with code-status discussions that both they and their attending physicians have conducted.

The survey presented respondents with a hypothetical case of a patient with stage IV adenocarcinoma of the lung metastatic to the brain. The patient, who has disease progression despite receiving first- and second-line therapy, presents to the emergency department with dyspnea and is slightly hypoxemic, but is not in distress. The patient has not previously established a code-status preference.

The investigators contacted 387 residency program directors by mail, 19 of whom agreed to participate and responded. They sent surveys to a total of 1,627 residents, 358 of which were completed and included.

The investigators found that slightly less than half of the respondents said they would share information with the patient about his/her prognosis and the relative benefit of CPR, and more than two-thirds said they would be unlikely to offer a specific recommendation.

“So even in the situation with a clearly declining patient, residents were as likely as not to provide the information needed to make an informed decision, and were far less likely to provide guidance on this decision,” Dr. Einstein said.

Asked the reason for their decisions, 69% of the residents who would not offer a recommendation said that the patient should make his/her own decision without any influence, and 26.5% said that the attending would not want them to offer a recommendation. Nonetheless, only 1.3% of this group said they believed that CPR would offer the patient a reasonable chance of resuscitation.

The majority of respondents who would offer a recommendation (93.5%) said they would recommend DNR and DNI.

Code-status talk a ‘responsibility’

When they were asked whether they would prefer the attending to discuss code status, nearly 70% of respondents disagreed.

Of those residents who said they preferred to retain the code-status discussion, 93.4% said they thought it was part of their responsibility as a physician, and 65.8% said they thought they had sufficient training and knowledge to do it. A minority in this group (2.5%) said that they would be likely to disagree with the attending’s estimate of prognosis, and 4.9% said they thought the attending would not share his/her estimate honestly.

When the authors asked about the residents’ general satisfaction with discussion of code status, “we learned two things: one, the residents are significantly more satisfied with their own discussions than their attendings’ discussions; and two, there is a substantial minority that is dissatisfied with all discussions, and a small number who are actually very satisfied,” Dr. Einstein said.

In a linear regression analysis testing for hypothesized correlations, the investigators found that more-senior residents were more likely to share prognostic information and make recommendations (P = .002). Residents who expressed an interest in hematology/oncology or palliative care specialization were also more likely to offer prognostic information, but not to make a recommendation about code status.

More-senior year of training correlated negatively with satisfaction with both the resident’s own and the attending’s discussion of code status.

“We found substantial dissatisfaction with code-status discussions in general, and we hypothesize that this is due to an internal conflict. When a resident knows that an intervention may be more harmful than beneficial, but thinks that the patient should make their own decision alone, then one may experience substantial frustration, and this would increase as training goes on and one becomes more sure of the outcomes of interventions like CPR,” Dr. Einstein said.

Generation gap

Evoking a potential generation gap between old-school doctors and the up-and-coming young physicians who by statute work fewer hours than their mentors had to, “I’m struck that [residents] don’t trust the attendings. When I was a resident, you didn’t do anything without asking the attending,” said Dr. Michael H. Levy, an invited discussant who is vice chair of medical oncology and director of the pain and palliative care program at Fox Chase Cancer Center, Philadelphia.

“I’m glad that the residents want to do it, but they have the same arrogance/ignorance that they don’t know how, so if we want them to do it, we have to train them,” he said.

The symposium was cosponsored by AAHPM, ASCO, ASTRO, and MASCC. The study was supported in part by the Conquer Cancer Foundation. Dr. Einstein and Dr. Levy reported having no relevant disclosures.

BOSTON – Medical residents in the United States appear to understand that cardiopulmonary resuscitation or intubation is highly unlikely to benefit patients with advanced cancers at the end of life, but the majority of residents surveyed said that they do not discuss code-status options or potentially beneficial palliative care with their dying patients.

“This was primarily due to residents’ perceptions of patient autonomy: Residents wanted patients to make their own decisions, without any influence from the doctor, which misses the concept of informed decision making. These incomplete discussions can cause at minimum improper documentation of patients’ wishes, and at most psychological harm, damage to the physician-patient relationship, and the potential for unwanted attempts at resuscitation,” said Dr. David J. Einstein, a resident at Beth Israel Deaconess Medical Center and Tufts Medical Center, both in Boston.

Despite their reluctance to have the discussion, however, the majority of residents said they preferred to discuss code status with patients themselves rather than hand it off to the attending physician, primarily out of a sense that it is their responsibility as physicians. Yet these physicians in training did not seem to feel that they were also responsible for providing guidance to patients, Dr. Einstein said at the Palliative Care in Oncology Symposium.

“We felt that this represented an unmet need in training and practice. Residents and attendings should be providing guidance on all medical interventions, including CPR, and if they aren’t sure what to recommend, then they themselves should be seeking guidance from other experts, before asking a patient to falsely choose between an intervention and death,” he said.

The first discussion of code status – do not resuscitate (DNR) or do not intubate (DNI) – may occur in the hospital, and is often left to a resident physician. Ideally, the physician and patient should discuss the patient’s prognosis, goals for care, evaluation of CPR as a means of meeting those goals, and a recommendation. But many residents lack training in the end-of-life discussion, which can have a significant impact on the quality of the patients’ remaining weeks or months of life.

Nationwide survey

Dr. Einstein and his colleagues conducted a nationwide survey to measure the likelihood that residents would discuss prognostic information and offer recommendations to patients with limited life expectancy. They also sought to determine why residents might be reluctant to provide discussion, and to evaluate their satisfaction with code-status discussions that both they and their attending physicians have conducted.

The survey presented respondents with a hypothetical case of a patient with stage IV adenocarcinoma of the lung metastatic to the brain. The patient, who has disease progression despite receiving first- and second-line therapy, presents to the emergency department with dyspnea and is slightly hypoxemic, but is not in distress. The patient has not previously established a code-status preference.

The investigators contacted 387 residency program directors by mail, 19 of whom agreed to participate and responded. They sent surveys to a total of 1,627 residents, 358 of which were completed and included.

The investigators found that slightly less than half of the respondents said they would share information with the patient about his/her prognosis and the relative benefit of CPR, and more than two-thirds said they would be unlikely to offer a specific recommendation.

“So even in the situation with a clearly declining patient, residents were as likely as not to provide the information needed to make an informed decision, and were far less likely to provide guidance on this decision,” Dr. Einstein said.

Asked the reason for their decisions, 69% of the residents who would not offer a recommendation said that the patient should make his/her own decision without any influence, and 26.5% said that the attending would not want them to offer a recommendation. Nonetheless, only 1.3% of this group said they believed that CPR would offer the patient a reasonable chance of resuscitation.

The majority of respondents who would offer a recommendation (93.5%) said they would recommend DNR and DNI.

Code-status talk a ‘responsibility’

When they were asked whether they would prefer the attending to discuss code status, nearly 70% of respondents disagreed.

Of those residents who said they preferred to retain the code-status discussion, 93.4% said they thought it was part of their responsibility as a physician, and 65.8% said they thought they had sufficient training and knowledge to do it. A minority in this group (2.5%) said that they would be likely to disagree with the attending’s estimate of prognosis, and 4.9% said they thought the attending would not share his/her estimate honestly.

When the authors asked about the residents’ general satisfaction with discussion of code status, “we learned two things: one, the residents are significantly more satisfied with their own discussions than their attendings’ discussions; and two, there is a substantial minority that is dissatisfied with all discussions, and a small number who are actually very satisfied,” Dr. Einstein said.

In a linear regression analysis testing for hypothesized correlations, the investigators found that more-senior residents were more likely to share prognostic information and make recommendations (P = .002). Residents who expressed an interest in hematology/oncology or palliative care specialization were also more likely to offer prognostic information, but not to make a recommendation about code status.

More-senior year of training correlated negatively with satisfaction with both the resident’s own and the attending’s discussion of code status.

“We found substantial dissatisfaction with code-status discussions in general, and we hypothesize that this is due to an internal conflict. When a resident knows that an intervention may be more harmful than beneficial, but thinks that the patient should make their own decision alone, then one may experience substantial frustration, and this would increase as training goes on and one becomes more sure of the outcomes of interventions like CPR,” Dr. Einstein said.

Generation gap

Evoking a potential generation gap between old-school doctors and the up-and-coming young physicians who by statute work fewer hours than their mentors had to, “I’m struck that [residents] don’t trust the attendings. When I was a resident, you didn’t do anything without asking the attending,” said Dr. Michael H. Levy, an invited discussant who is vice chair of medical oncology and director of the pain and palliative care program at Fox Chase Cancer Center, Philadelphia.

“I’m glad that the residents want to do it, but they have the same arrogance/ignorance that they don’t know how, so if we want them to do it, we have to train them,” he said.

The symposium was cosponsored by AAHPM, ASCO, ASTRO, and MASCC. The study was supported in part by the Conquer Cancer Foundation. Dr. Einstein and Dr. Levy reported having no relevant disclosures.

BOSTON – Medical residents in the United States appear to understand that cardiopulmonary resuscitation or intubation is highly unlikely to benefit patients with advanced cancers at the end of life, but the majority of residents surveyed said that they do not discuss code-status options or potentially beneficial palliative care with their dying patients.

“This was primarily due to residents’ perceptions of patient autonomy: Residents wanted patients to make their own decisions, without any influence from the doctor, which misses the concept of informed decision making. These incomplete discussions can cause at minimum improper documentation of patients’ wishes, and at most psychological harm, damage to the physician-patient relationship, and the potential for unwanted attempts at resuscitation,” said Dr. David J. Einstein, a resident at Beth Israel Deaconess Medical Center and Tufts Medical Center, both in Boston.

Despite their reluctance to have the discussion, however, the majority of residents said they preferred to discuss code status with patients themselves rather than hand it off to the attending physician, primarily out of a sense that it is their responsibility as physicians. Yet these physicians in training did not seem to feel that they were also responsible for providing guidance to patients, Dr. Einstein said at the Palliative Care in Oncology Symposium.

“We felt that this represented an unmet need in training and practice. Residents and attendings should be providing guidance on all medical interventions, including CPR, and if they aren’t sure what to recommend, then they themselves should be seeking guidance from other experts, before asking a patient to falsely choose between an intervention and death,” he said.

The first discussion of code status – do not resuscitate (DNR) or do not intubate (DNI) – may occur in the hospital, and is often left to a resident physician. Ideally, the physician and patient should discuss the patient’s prognosis, goals for care, evaluation of CPR as a means of meeting those goals, and a recommendation. But many residents lack training in the end-of-life discussion, which can have a significant impact on the quality of the patients’ remaining weeks or months of life.

Nationwide survey

Dr. Einstein and his colleagues conducted a nationwide survey to measure the likelihood that residents would discuss prognostic information and offer recommendations to patients with limited life expectancy. They also sought to determine why residents might be reluctant to provide discussion, and to evaluate their satisfaction with code-status discussions that both they and their attending physicians have conducted.

The survey presented respondents with a hypothetical case of a patient with stage IV adenocarcinoma of the lung metastatic to the brain. The patient, who has disease progression despite receiving first- and second-line therapy, presents to the emergency department with dyspnea and is slightly hypoxemic, but is not in distress. The patient has not previously established a code-status preference.

The investigators contacted 387 residency program directors by mail, 19 of whom agreed to participate and responded. They sent surveys to a total of 1,627 residents, 358 of which were completed and included.

The investigators found that slightly less than half of the respondents said they would share information with the patient about his/her prognosis and the relative benefit of CPR, and more than two-thirds said they would be unlikely to offer a specific recommendation.

“So even in the situation with a clearly declining patient, residents were as likely as not to provide the information needed to make an informed decision, and were far less likely to provide guidance on this decision,” Dr. Einstein said.

Asked the reason for their decisions, 69% of the residents who would not offer a recommendation said that the patient should make his/her own decision without any influence, and 26.5% said that the attending would not want them to offer a recommendation. Nonetheless, only 1.3% of this group said they believed that CPR would offer the patient a reasonable chance of resuscitation.

The majority of respondents who would offer a recommendation (93.5%) said they would recommend DNR and DNI.

Code-status talk a ‘responsibility’

When they were asked whether they would prefer the attending to discuss code status, nearly 70% of respondents disagreed.

Of those residents who said they preferred to retain the code-status discussion, 93.4% said they thought it was part of their responsibility as a physician, and 65.8% said they thought they had sufficient training and knowledge to do it. A minority in this group (2.5%) said that they would be likely to disagree with the attending’s estimate of prognosis, and 4.9% said they thought the attending would not share his/her estimate honestly.

When the authors asked about the residents’ general satisfaction with discussion of code status, “we learned two things: one, the residents are significantly more satisfied with their own discussions than their attendings’ discussions; and two, there is a substantial minority that is dissatisfied with all discussions, and a small number who are actually very satisfied,” Dr. Einstein said.

In a linear regression analysis testing for hypothesized correlations, the investigators found that more-senior residents were more likely to share prognostic information and make recommendations (P = .002). Residents who expressed an interest in hematology/oncology or palliative care specialization were also more likely to offer prognostic information, but not to make a recommendation about code status.

More-senior year of training correlated negatively with satisfaction with both the resident’s own and the attending’s discussion of code status.

“We found substantial dissatisfaction with code-status discussions in general, and we hypothesize that this is due to an internal conflict. When a resident knows that an intervention may be more harmful than beneficial, but thinks that the patient should make their own decision alone, then one may experience substantial frustration, and this would increase as training goes on and one becomes more sure of the outcomes of interventions like CPR,” Dr. Einstein said.

Generation gap

Evoking a potential generation gap between old-school doctors and the up-and-coming young physicians who by statute work fewer hours than their mentors had to, “I’m struck that [residents] don’t trust the attendings. When I was a resident, you didn’t do anything without asking the attending,” said Dr. Michael H. Levy, an invited discussant who is vice chair of medical oncology and director of the pain and palliative care program at Fox Chase Cancer Center, Philadelphia.

“I’m glad that the residents want to do it, but they have the same arrogance/ignorance that they don’t know how, so if we want them to do it, we have to train them,” he said.

The symposium was cosponsored by AAHPM, ASCO, ASTRO, and MASCC. The study was supported in part by the Conquer Cancer Foundation. Dr. Einstein and Dr. Levy reported having no relevant disclosures.

BOSTON – The risk for cognitive decline following cancer treatment varies by both cancer and therapy types, and can range from subtle changes to severe deficits, according to a researcher.

Patients who are older, have lower cognitive reserves, or have comorbidities such as cardiovascular disease or diabetes are at risk for cognitive problems following cancer treatment, said Tim A. Ahles, Ph.D., director of the Neurocognitive Research Laboratory, Memorial Sloan-Kettering Cancer Center, New York.

“I think it’s important that we identify these modifiable risk factors for intervention, and some of the nonmodifiable risk factors that inform decision making,” he said at the Palliative Care in Oncology Symposium.

“When we talk about cognitive function, we’re really talking about how does cancer and cancer treatment impact on memory, concentration, executive function, or ability to multitask, the speed at which we process information,” he said.

Oncologists have known for decades that brain tumors and their treatment have a negative effect on cognitive function, particularly among children under 5 years of age, whose developing brains are sensitive to treatments such as chemotherapy, surgery, radiation, and high-dose steroids.

There is also a dose-response effect, with high-dose chemotherapy such as ablative regimens used for bone-marrow transplantation being associated with higher probability of cognitive problems.

Dr. Ahles noted that about two-thirds of adult survivors of childhood cancers develop chronic illnesses within 30 years of diagnosis, including cardiac and pulmonary disease, and diabetes and endocrine dysfunction.

“It turns out they’re also at higher risk for cognitive issues,” including white matter abnormalities and microvascular stroke, he said.

The population of survivors of brain tumors and childhood cancers is dwarfed, however, by the large and growing population of breast, colorectal, lung, and prostate cancer patients who are diagnosed every year and exposed to adjuvant therapies, he added.

Aging and cognitive reserve

Evidence from breast cancer studies has shown that about 20%-25% of patients have lower than expected cognitive functioning – based on age, education, occupation, and other factors – before they embark on adjuvant therapy.

“That’s actually a risk factor for posttreatment cognitive decline, so there’s something that’s already going on that’s disrupting the cognitive information-processing system before we even start adjuvant treatment that may be critically important in terms of their outcomes as survivors,” Dr. Ahles said.

A significant subset of women in longitudinal studies of breast cancer survivors – about 15%-30% – experience long-term posttreatment cognitive problems, making it imperative for researchers and clinicians to identify risk factors for persistent cognitive decline, he said.

There is evidence to suggest that cancer treatments may interplay with biologic factors at the cellular level to increase the risk for cognitive loss. For example, aging is associated with reduction in brain volume, decrease in white matter integrity, and decreases in vascularization and neurotransmitter activity.

The effects of age on the brain are attenuated, however, among patients with higher cognitive reserves, defined as a combination of innate and developed cognitive capacity. Cognitive reserve is influenced by a number of factors, including genetics, education, occupational attainment, and lifestyle.

High cognitive reserve has been associated with later onset of Alzheimer’s disease symptoms and smaller changes in cognitive function with normal aging or following a brain injury, Dr. Ahles noted.

In a longitudinal study of cognitive changes associated with adjuvant therapy for breast cancer, Dr. Ahles and colleagues found that both age and pretreatment cognitive reserve were related to posttreatment decline in processing speed in women exposed to chemotherapy, compared with those who did not have chemotherapy or with healthy controls. In addition, chemotherapy had a short-term impact on verbal ability. The authors found evidence to suggest the patterns they saw may be related to the combination of chemotherapy and tamoxifen.

Part of the difficulty of studying cognitive decline among older patients is the higher prevalence of changes associated with aging. Dr. Ahles pointed to a French longitudinal study of women over 65 being treated for breast cancer, in which investigators found that 41% of the study population had cognitive impairment before starting on adjuvant therapy.

Older adults may be more frail, with diminished biological reserves and lower resistance to stressors caused by “cumulative declines across a variety of physiological systems making you more vulnerable to adverse events,” Dr. Ahles said.

Aging and genetics

Genes associated with cognitive aging are also risk factors for posttreatment cognitive decline, notably the genetic variants of APOE, including the epsilon 4 (APOE-e4) allele linked to increased risk for early-onset Alzheimer’s disease.

Dr. Ahles and colleagues had previously shown that APOE-e4 may be a biomarker for increased risk for chemotherapy-induced cognitive decline. The adverse effects of APOE-e4 appear to be mitigated somewhat by smoking, because it may correct for a deficit in nicotinic receptor density and dopamine levels in carriers.

Another genetic factor linked to postchemotherapy cognitive decline is the Val158Met polymorphism of the gene encoding for Catechol-O-methyltransferase (COMT), an enzyme that degrades neurotransmitters such as dopamine. Patients with this polymorphism have rapid dopamine metabolism, resulting in reduced dopamine activity.

These findings point to potential molecular mechanisms for cognitive changes associated with chemotherapy, and suggest that therapies targeted at neurotransmitter systems may ameliorate the effect, Dr. Ahles said.

He noted that animal studies have shown that fluoxetine (Prozac) prevents deficits in behavior and hippocampal function associated with 5-fluourauracil (5-FU), and that nicotine patches have been shown to improve cognitive functioning in patients with mild cognitive impairment.

The symposium was cosponsored by AAHPM, ASCO, ASTRO, and MASCC.

BOSTON – The risk for cognitive decline following cancer treatment varies by both cancer and therapy types, and can range from subtle changes to severe deficits, according to a researcher.

Patients who are older, have lower cognitive reserves, or have comorbidities such as cardiovascular disease or diabetes are at risk for cognitive problems following cancer treatment, said Tim A. Ahles, Ph.D., director of the Neurocognitive Research Laboratory, Memorial Sloan-Kettering Cancer Center, New York.

“I think it’s important that we identify these modifiable risk factors for intervention, and some of the nonmodifiable risk factors that inform decision making,” he said at the Palliative Care in Oncology Symposium.

“When we talk about cognitive function, we’re really talking about how does cancer and cancer treatment impact on memory, concentration, executive function, or ability to multitask, the speed at which we process information,” he said.

Oncologists have known for decades that brain tumors and their treatment have a negative effect on cognitive function, particularly among children under 5 years of age, whose developing brains are sensitive to treatments such as chemotherapy, surgery, radiation, and high-dose steroids.

There is also a dose-response effect, with high-dose chemotherapy such as ablative regimens used for bone-marrow transplantation being associated with higher probability of cognitive problems.

Dr. Ahles noted that about two-thirds of adult survivors of childhood cancers develop chronic illnesses within 30 years of diagnosis, including cardiac and pulmonary disease, and diabetes and endocrine dysfunction.

“It turns out they’re also at higher risk for cognitive issues,” including white matter abnormalities and microvascular stroke, he said.

The population of survivors of brain tumors and childhood cancers is dwarfed, however, by the large and growing population of breast, colorectal, lung, and prostate cancer patients who are diagnosed every year and exposed to adjuvant therapies, he added.

Aging and cognitive reserve

Evidence from breast cancer studies has shown that about 20%-25% of patients have lower than expected cognitive functioning – based on age, education, occupation, and other factors – before they embark on adjuvant therapy.

“That’s actually a risk factor for posttreatment cognitive decline, so there’s something that’s already going on that’s disrupting the cognitive information-processing system before we even start adjuvant treatment that may be critically important in terms of their outcomes as survivors,” Dr. Ahles said.

A significant subset of women in longitudinal studies of breast cancer survivors – about 15%-30% – experience long-term posttreatment cognitive problems, making it imperative for researchers and clinicians to identify risk factors for persistent cognitive decline, he said.

There is evidence to suggest that cancer treatments may interplay with biologic factors at the cellular level to increase the risk for cognitive loss. For example, aging is associated with reduction in brain volume, decrease in white matter integrity, and decreases in vascularization and neurotransmitter activity.

The effects of age on the brain are attenuated, however, among patients with higher cognitive reserves, defined as a combination of innate and developed cognitive capacity. Cognitive reserve is influenced by a number of factors, including genetics, education, occupational attainment, and lifestyle.

High cognitive reserve has been associated with later onset of Alzheimer’s disease symptoms and smaller changes in cognitive function with normal aging or following a brain injury, Dr. Ahles noted.

In a longitudinal study of cognitive changes associated with adjuvant therapy for breast cancer, Dr. Ahles and colleagues found that both age and pretreatment cognitive reserve were related to posttreatment decline in processing speed in women exposed to chemotherapy, compared with those who did not have chemotherapy or with healthy controls. In addition, chemotherapy had a short-term impact on verbal ability. The authors found evidence to suggest the patterns they saw may be related to the combination of chemotherapy and tamoxifen.

Part of the difficulty of studying cognitive decline among older patients is the higher prevalence of changes associated with aging. Dr. Ahles pointed to a French longitudinal study of women over 65 being treated for breast cancer, in which investigators found that 41% of the study population had cognitive impairment before starting on adjuvant therapy.

Older adults may be more frail, with diminished biological reserves and lower resistance to stressors caused by “cumulative declines across a variety of physiological systems making you more vulnerable to adverse events,” Dr. Ahles said.

Aging and genetics

Genes associated with cognitive aging are also risk factors for posttreatment cognitive decline, notably the genetic variants of APOE, including the epsilon 4 (APOE-e4) allele linked to increased risk for early-onset Alzheimer’s disease.

Dr. Ahles and colleagues had previously shown that APOE-e4 may be a biomarker for increased risk for chemotherapy-induced cognitive decline. The adverse effects of APOE-e4 appear to be mitigated somewhat by smoking, because it may correct for a deficit in nicotinic receptor density and dopamine levels in carriers.

Another genetic factor linked to postchemotherapy cognitive decline is the Val158Met polymorphism of the gene encoding for Catechol-O-methyltransferase (COMT), an enzyme that degrades neurotransmitters such as dopamine. Patients with this polymorphism have rapid dopamine metabolism, resulting in reduced dopamine activity.

These findings point to potential molecular mechanisms for cognitive changes associated with chemotherapy, and suggest that therapies targeted at neurotransmitter systems may ameliorate the effect, Dr. Ahles said.

He noted that animal studies have shown that fluoxetine (Prozac) prevents deficits in behavior and hippocampal function associated with 5-fluourauracil (5-FU), and that nicotine patches have been shown to improve cognitive functioning in patients with mild cognitive impairment.

The symposium was cosponsored by AAHPM, ASCO, ASTRO, and MASCC.

BOSTON – The risk for cognitive decline following cancer treatment varies by both cancer and therapy types, and can range from subtle changes to severe deficits, according to a researcher.

Patients who are older, have lower cognitive reserves, or have comorbidities such as cardiovascular disease or diabetes are at risk for cognitive problems following cancer treatment, said Tim A. Ahles, Ph.D., director of the Neurocognitive Research Laboratory, Memorial Sloan-Kettering Cancer Center, New York.

“I think it’s important that we identify these modifiable risk factors for intervention, and some of the nonmodifiable risk factors that inform decision making,” he said at the Palliative Care in Oncology Symposium.

“When we talk about cognitive function, we’re really talking about how does cancer and cancer treatment impact on memory, concentration, executive function, or ability to multitask, the speed at which we process information,” he said.

Oncologists have known for decades that brain tumors and their treatment have a negative effect on cognitive function, particularly among children under 5 years of age, whose developing brains are sensitive to treatments such as chemotherapy, surgery, radiation, and high-dose steroids.

There is also a dose-response effect, with high-dose chemotherapy such as ablative regimens used for bone-marrow transplantation being associated with higher probability of cognitive problems.

Dr. Ahles noted that about two-thirds of adult survivors of childhood cancers develop chronic illnesses within 30 years of diagnosis, including cardiac and pulmonary disease, and diabetes and endocrine dysfunction.

“It turns out they’re also at higher risk for cognitive issues,” including white matter abnormalities and microvascular stroke, he said.

The population of survivors of brain tumors and childhood cancers is dwarfed, however, by the large and growing population of breast, colorectal, lung, and prostate cancer patients who are diagnosed every year and exposed to adjuvant therapies, he added.

Aging and cognitive reserve

Evidence from breast cancer studies has shown that about 20%-25% of patients have lower than expected cognitive functioning – based on age, education, occupation, and other factors – before they embark on adjuvant therapy.

“That’s actually a risk factor for posttreatment cognitive decline, so there’s something that’s already going on that’s disrupting the cognitive information-processing system before we even start adjuvant treatment that may be critically important in terms of their outcomes as survivors,” Dr. Ahles said.

A significant subset of women in longitudinal studies of breast cancer survivors – about 15%-30% – experience long-term posttreatment cognitive problems, making it imperative for researchers and clinicians to identify risk factors for persistent cognitive decline, he said.

There is evidence to suggest that cancer treatments may interplay with biologic factors at the cellular level to increase the risk for cognitive loss. For example, aging is associated with reduction in brain volume, decrease in white matter integrity, and decreases in vascularization and neurotransmitter activity.

The effects of age on the brain are attenuated, however, among patients with higher cognitive reserves, defined as a combination of innate and developed cognitive capacity. Cognitive reserve is influenced by a number of factors, including genetics, education, occupational attainment, and lifestyle.

High cognitive reserve has been associated with later onset of Alzheimer’s disease symptoms and smaller changes in cognitive function with normal aging or following a brain injury, Dr. Ahles noted.

In a longitudinal study of cognitive changes associated with adjuvant therapy for breast cancer, Dr. Ahles and colleagues found that both age and pretreatment cognitive reserve were related to posttreatment decline in processing speed in women exposed to chemotherapy, compared with those who did not have chemotherapy or with healthy controls. In addition, chemotherapy had a short-term impact on verbal ability. The authors found evidence to suggest the patterns they saw may be related to the combination of chemotherapy and tamoxifen.

Part of the difficulty of studying cognitive decline among older patients is the higher prevalence of changes associated with aging. Dr. Ahles pointed to a French longitudinal study of women over 65 being treated for breast cancer, in which investigators found that 41% of the study population had cognitive impairment before starting on adjuvant therapy.

Older adults may be more frail, with diminished biological reserves and lower resistance to stressors caused by “cumulative declines across a variety of physiological systems making you more vulnerable to adverse events,” Dr. Ahles said.

Aging and genetics

Genes associated with cognitive aging are also risk factors for posttreatment cognitive decline, notably the genetic variants of APOE, including the epsilon 4 (APOE-e4) allele linked to increased risk for early-onset Alzheimer’s disease.

Dr. Ahles and colleagues had previously shown that APOE-e4 may be a biomarker for increased risk for chemotherapy-induced cognitive decline. The adverse effects of APOE-e4 appear to be mitigated somewhat by smoking, because it may correct for a deficit in nicotinic receptor density and dopamine levels in carriers.

Another genetic factor linked to postchemotherapy cognitive decline is the Val158Met polymorphism of the gene encoding for Catechol-O-methyltransferase (COMT), an enzyme that degrades neurotransmitters such as dopamine. Patients with this polymorphism have rapid dopamine metabolism, resulting in reduced dopamine activity.

These findings point to potential molecular mechanisms for cognitive changes associated with chemotherapy, and suggest that therapies targeted at neurotransmitter systems may ameliorate the effect, Dr. Ahles said.

He noted that animal studies have shown that fluoxetine (Prozac) prevents deficits in behavior and hippocampal function associated with 5-fluourauracil (5-FU), and that nicotine patches have been shown to improve cognitive functioning in patients with mild cognitive impairment.

The symposium was cosponsored by AAHPM, ASCO, ASTRO, and MASCC.

BOSTON – An automated system for monitoring the symptoms of patients in home hospice and supporting their caregivers with coaching can both improve patient comfort and relieve at least some of the stress on the caregiver, said investigators in a pilot program.

In a prospective, randomized controlled trial, caregivers assigned to symptom care by phone, in which they received automated coaching tailored to the specific situation, had better overall vitality, and the patients they cared for had less fatigue and anxiety than caregiver-patient pairs assigned to usual care, reported Dr. Kathi Mooney, professor of nursing at the University of Utah College of Nursing, Salt Lake City.

“The family inclusion in hospice care, the philosophy around that, offers the opportunity to extend PROs [patient-reported outcomes] to FCROs, family caregiver reported outcomes. We call that an opportunity to use electronic monitoring to monitor a family,” she said at the Palliative Care in Oncology Symposium.

Dr. Mooney and her colleagues have previously reported on the use of automated symptom monitoring for support of patients undergoing ambulatory chemotherapy. In the current study, she described its application in home-based end-of-life care to support both the patient and the caregiver.

The investigators recruited 319 cancer patient/caregiver dyads from 12 hospices in Illinois, Massachusetts, Oregon, and Utah. The mean age of the patients was 72 years, and the mean age of caregivers was 59 years.

The caregivers were randomly assigned either to the symptom care by phone (SCP) system or to usual care. The SCP system is an automated system in which caregivers phone in to report the patient’s symptoms and their own levels of stress, anxiety, etc., and receive automated responses based on the severity of symptoms and number of days reported without relief. The system offers both patient care and self-care strategies, and identifies and reinforces issues that should be addressed by contacting the hospice team. At the end of the call, the data are sent to the nurse, including alerts to matters that require prompt attention.

Caregivers in each group made daily automated monitoring calls to report 1 or more of 11 patient symptoms and 5 caregiver symptoms on a 0-10 scale, and to report the patient’s and their own distress about the symptoms.

“We have developed algorithms to provide just-in-time, tailored suggestions related to the pattern that they have reported,” Dr. Mooney said.

Data reported by caregivers in the control (usual care) group were recorded but not acted upon, whereas data reported by those in the SCP intervention group triggered the automated coaching and hospice nurse alerts for moderate to severe symptoms.

“People in the usual care group understood that they were just contributing symptom information, and they were told on every call and at consent that if they had any concerns to call their hospice nurse,” Dr. Mooney said.

For clinicians, the automated system triggers an alert at preset thresholds of severity (4-10), as well as trend alerts. Hospice nurses have online and mobile access to the alert website, where they can view a report of symptoms over the previous 24 hours, and review graphs of symptoms over time to see trends. The nurses were instructed to log that they had seen the alert and what their planned action was, which was left to their professional judgment.

The average length of the calls was 7 minutes, 4 seconds for the usual care group, and 7 minutes, 59 seconds for the intervention group. Control group caregivers completed 59% of expected calls, and those in the intervention group completed 64%.

Reported symptoms present in more than 50% of patients included fatigue (88.8%), pain (60.2%), eating/drinking problems (76.6%), difficulty thinking (69.6%), anxiety (67.6%), negative mood (67.3%), bowel problems (61%), and trouble sleeping (58.3%).

Among caregivers, 73.3% reported fatigue, 66.7% anxiety, 61% trouble sleeping, and 57.1% negative mood.

In a mixed-effects model looking at caregiver vitality – a composite of caregiver symptoms – the authors found that the severity of symptoms in the usual care group increased steadily over the course of hospice care, out to at least 91 days. In contrast, the severity of symptoms among caregivers in the SCP group rose only slightly. The between-group difference was significant (P < .001).

Similarly, among patients the overall symptom severity scores were lower for those in the intervention group than for controls (P =.03). Additionally, the onset of benefit, defined as time to the first symptom-free day, was significantly earlier among patients in the SCP group (P < .02)

Patterns of patient and fatigue during the last 8 weeks of life also favored the symptom-care intervention, Dr. Mooney noted.`

Invited discussant Dr. Michael H. Levy, vice chair of medical oncology and director of the pain and palliative care program at Fox Chase Cancer Center in Philadelphia, commented that the SCP system “is interesting, but it’s very expensive, the impact is limited, and it’s challenging to export.”

He noted that for a palliative care support system to be implemented successfully, it would need to be reproducible and applicable to a wide range of health care systems.

The symposium was cosponsored by AAHPM, ASCO, ASTRO, and MASCC.

The study was supported by the National Cancer Institute. Dr. Mooney and Dr. Levy reported having no relevant disclosures.

BOSTON – An automated system for monitoring the symptoms of patients in home hospice and supporting their caregivers with coaching can both improve patient comfort and relieve at least some of the stress on the caregiver, said investigators in a pilot program.

In a prospective, randomized controlled trial, caregivers assigned to symptom care by phone, in which they received automated coaching tailored to the specific situation, had better overall vitality, and the patients they cared for had less fatigue and anxiety than caregiver-patient pairs assigned to usual care, reported Dr. Kathi Mooney, professor of nursing at the University of Utah College of Nursing, Salt Lake City.

“The family inclusion in hospice care, the philosophy around that, offers the opportunity to extend PROs [patient-reported outcomes] to FCROs, family caregiver reported outcomes. We call that an opportunity to use electronic monitoring to monitor a family,” she said at the Palliative Care in Oncology Symposium.

Dr. Mooney and her colleagues have previously reported on the use of automated symptom monitoring for support of patients undergoing ambulatory chemotherapy. In the current study, she described its application in home-based end-of-life care to support both the patient and the caregiver.

The investigators recruited 319 cancer patient/caregiver dyads from 12 hospices in Illinois, Massachusetts, Oregon, and Utah. The mean age of the patients was 72 years, and the mean age of caregivers was 59 years.

The caregivers were randomly assigned either to the symptom care by phone (SCP) system or to usual care. The SCP system is an automated system in which caregivers phone in to report the patient’s symptoms and their own levels of stress, anxiety, etc., and receive automated responses based on the severity of symptoms and number of days reported without relief. The system offers both patient care and self-care strategies, and identifies and reinforces issues that should be addressed by contacting the hospice team. At the end of the call, the data are sent to the nurse, including alerts to matters that require prompt attention.

Caregivers in each group made daily automated monitoring calls to report 1 or more of 11 patient symptoms and 5 caregiver symptoms on a 0-10 scale, and to report the patient’s and their own distress about the symptoms.

“We have developed algorithms to provide just-in-time, tailored suggestions related to the pattern that they have reported,” Dr. Mooney said.

Data reported by caregivers in the control (usual care) group were recorded but not acted upon, whereas data reported by those in the SCP intervention group triggered the automated coaching and hospice nurse alerts for moderate to severe symptoms.

“People in the usual care group understood that they were just contributing symptom information, and they were told on every call and at consent that if they had any concerns to call their hospice nurse,” Dr. Mooney said.

For clinicians, the automated system triggers an alert at preset thresholds of severity (4-10), as well as trend alerts. Hospice nurses have online and mobile access to the alert website, where they can view a report of symptoms over the previous 24 hours, and review graphs of symptoms over time to see trends. The nurses were instructed to log that they had seen the alert and what their planned action was, which was left to their professional judgment.

The average length of the calls was 7 minutes, 4 seconds for the usual care group, and 7 minutes, 59 seconds for the intervention group. Control group caregivers completed 59% of expected calls, and those in the intervention group completed 64%.

Reported symptoms present in more than 50% of patients included fatigue (88.8%), pain (60.2%), eating/drinking problems (76.6%), difficulty thinking (69.6%), anxiety (67.6%), negative mood (67.3%), bowel problems (61%), and trouble sleeping (58.3%).

Among caregivers, 73.3% reported fatigue, 66.7% anxiety, 61% trouble sleeping, and 57.1% negative mood.

In a mixed-effects model looking at caregiver vitality – a composite of caregiver symptoms – the authors found that the severity of symptoms in the usual care group increased steadily over the course of hospice care, out to at least 91 days. In contrast, the severity of symptoms among caregivers in the SCP group rose only slightly. The between-group difference was significant (P < .001).

Similarly, among patients the overall symptom severity scores were lower for those in the intervention group than for controls (P =.03). Additionally, the onset of benefit, defined as time to the first symptom-free day, was significantly earlier among patients in the SCP group (P < .02)

Patterns of patient and fatigue during the last 8 weeks of life also favored the symptom-care intervention, Dr. Mooney noted.`

Invited discussant Dr. Michael H. Levy, vice chair of medical oncology and director of the pain and palliative care program at Fox Chase Cancer Center in Philadelphia, commented that the SCP system “is interesting, but it’s very expensive, the impact is limited, and it’s challenging to export.”

He noted that for a palliative care support system to be implemented successfully, it would need to be reproducible and applicable to a wide range of health care systems.

The symposium was cosponsored by AAHPM, ASCO, ASTRO, and MASCC.

The study was supported by the National Cancer Institute. Dr. Mooney and Dr. Levy reported having no relevant disclosures.

BOSTON – An automated system for monitoring the symptoms of patients in home hospice and supporting their caregivers with coaching can both improve patient comfort and relieve at least some of the stress on the caregiver, said investigators in a pilot program.

In a prospective, randomized controlled trial, caregivers assigned to symptom care by phone, in which they received automated coaching tailored to the specific situation, had better overall vitality, and the patients they cared for had less fatigue and anxiety than caregiver-patient pairs assigned to usual care, reported Dr. Kathi Mooney, professor of nursing at the University of Utah College of Nursing, Salt Lake City.

“The family inclusion in hospice care, the philosophy around that, offers the opportunity to extend PROs [patient-reported outcomes] to FCROs, family caregiver reported outcomes. We call that an opportunity to use electronic monitoring to monitor a family,” she said at the Palliative Care in Oncology Symposium.

Dr. Mooney and her colleagues have previously reported on the use of automated symptom monitoring for support of patients undergoing ambulatory chemotherapy. In the current study, she described its application in home-based end-of-life care to support both the patient and the caregiver.

The investigators recruited 319 cancer patient/caregiver dyads from 12 hospices in Illinois, Massachusetts, Oregon, and Utah. The mean age of the patients was 72 years, and the mean age of caregivers was 59 years.

The caregivers were randomly assigned either to the symptom care by phone (SCP) system or to usual care. The SCP system is an automated system in which caregivers phone in to report the patient’s symptoms and their own levels of stress, anxiety, etc., and receive automated responses based on the severity of symptoms and number of days reported without relief. The system offers both patient care and self-care strategies, and identifies and reinforces issues that should be addressed by contacting the hospice team. At the end of the call, the data are sent to the nurse, including alerts to matters that require prompt attention.

Caregivers in each group made daily automated monitoring calls to report 1 or more of 11 patient symptoms and 5 caregiver symptoms on a 0-10 scale, and to report the patient’s and their own distress about the symptoms.

“We have developed algorithms to provide just-in-time, tailored suggestions related to the pattern that they have reported,” Dr. Mooney said.

Data reported by caregivers in the control (usual care) group were recorded but not acted upon, whereas data reported by those in the SCP intervention group triggered the automated coaching and hospice nurse alerts for moderate to severe symptoms.

“People in the usual care group understood that they were just contributing symptom information, and they were told on every call and at consent that if they had any concerns to call their hospice nurse,” Dr. Mooney said.

For clinicians, the automated system triggers an alert at preset thresholds of severity (4-10), as well as trend alerts. Hospice nurses have online and mobile access to the alert website, where they can view a report of symptoms over the previous 24 hours, and review graphs of symptoms over time to see trends. The nurses were instructed to log that they had seen the alert and what their planned action was, which was left to their professional judgment.

The average length of the calls was 7 minutes, 4 seconds for the usual care group, and 7 minutes, 59 seconds for the intervention group. Control group caregivers completed 59% of expected calls, and those in the intervention group completed 64%.

Reported symptoms present in more than 50% of patients included fatigue (88.8%), pain (60.2%), eating/drinking problems (76.6%), difficulty thinking (69.6%), anxiety (67.6%), negative mood (67.3%), bowel problems (61%), and trouble sleeping (58.3%).

Among caregivers, 73.3% reported fatigue, 66.7% anxiety, 61% trouble sleeping, and 57.1% negative mood.

In a mixed-effects model looking at caregiver vitality – a composite of caregiver symptoms – the authors found that the severity of symptoms in the usual care group increased steadily over the course of hospice care, out to at least 91 days. In contrast, the severity of symptoms among caregivers in the SCP group rose only slightly. The between-group difference was significant (P < .001).

Similarly, among patients the overall symptom severity scores were lower for those in the intervention group than for controls (P =.03). Additionally, the onset of benefit, defined as time to the first symptom-free day, was significantly earlier among patients in the SCP group (P < .02)

Patterns of patient and fatigue during the last 8 weeks of life also favored the symptom-care intervention, Dr. Mooney noted.`

Invited discussant Dr. Michael H. Levy, vice chair of medical oncology and director of the pain and palliative care program at Fox Chase Cancer Center in Philadelphia, commented that the SCP system “is interesting, but it’s very expensive, the impact is limited, and it’s challenging to export.”

He noted that for a palliative care support system to be implemented successfully, it would need to be reproducible and applicable to a wide range of health care systems.

The symposium was cosponsored by AAHPM, ASCO, ASTRO, and MASCC.

The study was supported by the National Cancer Institute. Dr. Mooney and Dr. Levy reported having no relevant disclosures.