Kate M. O'Rourke


August 12, 2015

When an elderly patient is treated with a cancer drug, it is very likely that the drug has not been tested in a patient who is similar to them.

"Cancer clinical trials have been conducted primarily in middle aged patients, 50-60 years of age, 10-15 years younger than the incidence of the disease in the population," said Stuart Lichtman, MD, a medical oncologist at Memorial Sloan Kettering Cancer Center in New York City. "We know that 60% of cancer is in people over the age of 65, and 79% of cancer mortality is in people greater than 65 years of age. We treat older patients, particularly over the age of 80, with no evidence base whatsoever." During a presentation at the annual meeting of the American Society of Clinical Oncology (ASCO), Dr Lichtman said that changes are needed to the clinical trials infrastructure to improve cancer care for the elderly.

We treat older patients, particularly over the age of 80, with no evidence base whatsoever.

"Is it ethical to treat older patients with regimens based on data in which they were not adequately represented?" asked Dr Lichtman. "Is it ethical to design clinical trials in which older patients are ineligible, despite that they are the majority of patients with the disorder being studied? Is it ethical to design clinical trials that do not adequately assess older patients?"

According to 2007-2011 data from the Surveillance, Epidemiology, and End Results program, the average age of a patient diagnosed with colorectal cancer is 68 years, but in the MOSAIC trial, a practice-changing colorectal cancer trial, the median age of participants was 60 years.[1] "I can give you about a million other examples [like MOSAIC]," said Dr Lichtman.

The problem is likely to get worse, given longer life expectancies and an aging cancer population. A 75-year-old woman has a 25% chance of living into her 90s.[2]

The oncology community is aware of the problem, but little has been done to correct it. In 2002, a study analyzing National Cancer Institute-sponsored trials concluded that older adults were underrepresented in clinical trials and that there were essentially no data for patients over the age of 80 years.[3] Ten years later, another study found that little had changed.[4]

Elderly patients are often excluded from clinical trials because of their comorbidities or prior malignancies, but in many cases, the elderly are not even invited to join studies. A multivariate analysis of data from institutions involved in the Cancer and Leukemia Group B clinical trials group found that age and stage were the only predictors of whether a patient was invited to join a trial.[5] This analysis, which controlled for comorbid conditions, found that patients who were younger than 65 years were twice as likely to be invited to join a clinical trial than patients aged 65 years or older (68% vs 34%).[5] The greatest impediment to enrolling older individuals was physician perceptions about age and tolerance of toxicity.

The typical patient in a cancer clinical trial is young and fit, with a performance status of 0, so little is known about the toxicities of drugs in the elderly, said Dr Lichtman. There are very few pharmacokinetic studies in elderly patients, in part because pharmacokinetic studies have involved complicated schemes and sampling strategies that have made participation difficult for older patients. "Polypharmacy has also limited participation," said Dr Lichtman. "Polypharmacy is an extremely important component of a geriatric assessment. It is associated with adverse drug events and poor outcome." A recent study found that the median number of medications taken by seniors is 9.2.[6]

In the handful of pharmacokinetic studies conducted in older individuals, comorbidity, rather than age, has been the factor that causes changes to pharmacokinetics. Dr Lichtman pointed out that dose-limiting toxicity of cancer drugs is often due to nonhematologic toxicities, which are not related to significant differences in pharmacokinetics. One example is neuropathy from oxaliplatin (Eloxatin®). "The pharmacodynamics are altered by age and comorbidities and possibly pharmacogenomics," said Dr Lichtman. Many elderly patients on medications for comorbidities and cancer may experience cytochrome P450 drug interactions.

According to Heidi Klepin, MD, associate professor of medicine at the Wake Forest University Comprehensive Cancer Center, who also spoke at the annual ASCO meeting, dose reductions in older adults are commonly needed because changes in physiology and increased comorbidities can decrease treatment tolerance. No routine adjustments are needed for the decreased intestinal absorption, decreased hepatic metabolism, and body composition changes that frequently occur in older patients. However, dose adjustments based on creatinine clearance are routinely needed for impaired renal excretion.

Dr Klepin said that oncologists treating older patients should conduct a thorough medication review and discontinue medications that do not have a clear indication. Clinicians should also try to anticipate adverse events by using available drug interaction software and facilitating communication with primary care providers. She said that incorporation of geriatric assessment strategies is practical in oncology and can help individualize treatment and management decisions.

"Caring for older adults with cancer presents unique challenges, which are often not addressed in clinical trials," said Dr Klepin. "Elderly-specific trials that reflect our vulnerable patient population are necessary, and collection of patient-centric outcomes are really going to help us when we say, 'how do we manage you upfront and during the course of your therapy?'"

When analyzing clinical trial data, experts agree that lumping "the elderly" into one group is misleading. The elderly population is a much more heterogeneous population than a group of 40-year-olds. "When we design clinical trials, we need to determine which older patient we are talking about: the healthy older patient with no comorbidity and a PS of 0, the medically vulnerable patient who may be vulnerable to toxicity issues, or the frail elderly patient who may not be appropriate to be studied," said Dr Lichtman. He pointed out that when clinical trials include a large percentage of older patients, investigators often fail to describe the distribution of toxicities in this subpopulation when they publish their data.[7]

In 2013, the European Organisation for Research and Treatment of Cancer, the Alliance for Clinical Trials in Oncology, and the International Society of Geriatric Oncology made several recommendations regarding endpoints and trial design in geriatric oncology research.[8] They recommended doing away with the upper age limit for enrollment in cancer trials and encouraged the launch of trials specifically for older cancer patients that integrate meaningful measures of outcome. They also recommended the obligatory integration of a comparable form of a geriatric assessment in clinical trials and the establishment of large observational cohort studies/registries in the community setting to identify treatment differences in subsets of patients. "The European Medicines Agency and US Food and Drug Administration should require adequate collection of data on efficacy and toxicity of new drugs in fit and frail elderly subpopulations," the report's authors wrote.

Arti Hurria, MD, director of the Cancer and Aging Research Program at City of Hope in Duarte, California, also believes that geriatric oncology components need to be added to the standard oncology study design. In addition to focusing on cancer-specific outcomes such as disease-free survival and overall survival, clinical trials should focus on patient-centered outcomes such as function, cognition, toxicity, cost, and caregiver needs. Oncology outcomes need to be melded with geriatric outcomes, according to Dr Hurria.

Dr Lichtman agrees. "For a 70- or 80-year-old patient, should the endpoint of a clinical trial be survival? Probably not; the issue is whether the patient is going to die of or with cancer," said Dr Lichtman. "More appropriate endpoints may be time without symptoms, quality of life, functional benefit, and maintenance of independence."

Another area that needs reform is toxicity scales, according to geriatric oncology experts. Many of the scales have been developed in younger patients who do not have comorbidities, whereas elderly patients are more likely to experience severe toxicities from cancer treatments, including treatment-related mortality.[9,10] A study of 12,480 community-dwelling elders, roughly 20% of whom had a diagnosis of a non-skin cancer, found that after adjusting for confounders, a cancer diagnosis was associated with low self-rated health (adjusted odds ratio [OR], 1.46; relative risk [RR], 1.33), limitations in activities of daily living (OR, 1.19; RR, 1.13), limitations in instrumental activities of daily living (OR, 1.25; RR, 1.13), a geriatric syndrome (OR, 1.27; RR, 1.11), and frailty (OR, 1.46; RR, 1.09).[11] Other studies have also shown that evaluation of activities of daily living is an extremely powerful predictor of adverse events and survival.[12]

Toxicity scales should include more functional measures and patient-reported outcomes, say researchers. "Cancer increases the likelihood of having vulnerability and frailty," said Dr Lichtman. "We have to incorporate these issues into clinical trials. We have a number of geriatric scales we can use, such as activities of daily living, geriatric syndromes, physical functioning, mini-mental status evaluation, and geriatric depression scales."

A few years ago, Dr Hurria and colleagues conducted a study in 500 cancer patients over the age of 65 years and found that geriatric assessment variables independently predicted the risk for toxicity.[13] They developed a risk stratification schema that can establish the risk for chemotherapy toxicity in older adults. The 20-minute assessment, which is mostly completed by the patient, is undergoing additional validation but has now been integrated into a number of clinical trials. The assessment includes factors such as sociodemographics, laboratory test results, tumor/treatment variables, and geriatric assessment variables including function, falls, cognition, and comorbidities. The tool is one step forward in better serving elderly cancer patients, but more efforts are needed.

"We have to figure out a way to add appropriate geriatric oncology components to clinical trials because we have tremendous gaps in knowledge," said Dr Lichtman.

Drs Lichtman, Hurria, and Klepin have disclosed no relevant financial relationships.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.