Although older adults account for the majority of cancer incidence and mortality, few clinical trials have focused on this patient population (Figure 1). At the same time, the US population is aging, and the average life expectancy is increasing. (Over the 20th century, US life expectancy increased from 47.3 years in 1900 to 76.9 years in 2000.) Projections show that the US population age 65 and older will double by 2030, when 1 in 5 adults will be over the age of 65. The population age 85 and older (known as the "oldest old") will also double from 2000 to 2030, then double again from 2030 to 2050. Since approximately 60% of cancer incidence and 70% of cancer mortality occur in adults age 65 and older, it is projected that the number of older cancer patients and cancer survivors seeking treatment will also go up exponentially in the near future.
Representation of Older Adults in Clinical Trials
The growing worldwide population of older adults has been underrepresented in clinical trials that set the standards for oncology care.[3-7] From 1993 to 1996, only 25% of patients enrolled in Southwest Oncology Group (SWOG) trials were age 65 or older, but approximately 63% of cancers were diagnosed in this age group. During this same period, a similar underrepresentation of older patients was reported in Canadian trials, where only 22% of patients enrolled in clinical trials performed by the National Cancer Institute of Canada Clinical Trials Group (NCIC CTG) were age 65 and older. However, more recent data from SWOG and the Cancer and Leukemia Group B (CALGB) demonstrate some improvement in accrual of older adults to clinical trials. From 2001 to 2003, 38% of patients enrolled in SWOG trials and 36% of patients in CALGB trials were age 65 or older.[8,9] This accrual increase may result in part from a policy change in the year 2000 that required Medicare to reimburse providers for routine costs of patient care in clinical trials.
While improvement in the enrollment of older adults in clinical trials is encouraging, the representation of older adults on trials is still far from ideal, and the enrollment fraction of patients 75 and older in clinical trials (ie, the number enrolled divided by estimated cancer cases) worsens with increasing age. For example, in the National Cancer Institute (NCI) trials for breast, colorectal, lung, and prostate cancer from 2000 to 2002, the enrollment fraction was 3% for patients aged 30 to 64, 1.3% for patients aged 65 to 74, and 0.5% for patients aged 75 and older. Perhaps the most concerning statistics are that older adults are underrepresented in registration trials of new drugs approved by the US Food and Drug Administration (FDA), despite the FDA's recommendation to include older adults in clinical trials (Table 1). As a result, the standard drug dosing and expected side-effect profile is primarily derived from a younger cohort of patients.
In this article, we discuss the barriers to enrollment of older adults in clinical cancer trials, and propose research priorities to address these barriers and to overcome this knowledge gap in the treatment of older adults with cancer. We review age-related changes in physiology and pharmacokinetics that can affect treatment tolerance and efficacy in older adults, and discuss the role of a geriatric assessment in pinpointing factors other than chronologic age that are independent predictors of morbidity and mortality.
Barriers to Clinical Trial Enrollment of Older Patients
In a retrospective case-controlled study of patients with breast cancer, Kemeny et al sought to determine whether older patients (age ≥ 65 years) were as likely as younger patients (age < 65 years) to be offered and to accept treatment on a clinical trial. The pairs of older and younger patients were matched by disease stage and treating physician. The study results demonstrated that older patients were significantly less likely to be offered a clinical trial (35% vs 51%; P=.006); however, if offered, they were as likely to accept enrollment (50% ≥ age 65 vs 56% < age 65; P = .67). In a multivariate analysis, age and stage were significant predictors for whether a clinical trial was offered.
Older and younger patients declined to participate in a clinical trial for similar reasons: They wanted to choose their own treatment, wanted a treatment that was not offered, and they did not want to be an "experiment." The most common reason physicians did not offer a clinical trial to older and younger patients differed. For an older patient, the most common physician concern was that the treatment was too toxic, which was cited in 33% of the cases. Additional reasons for not offering a clinical trial included the doctor's perception that the trial did not offer the best treatment option, the doctor was unaware that a trial was available, the doctor thought the patient was not eligible, or the doctor was concerned about comorbid conditions affecting tolerance to therapy.
At the Canadian NCIC CTG meeting in 2000, 40 physicians were surveyed regarding the enrollment of older adults in clinical trials. Only 10% of the physicians reported that they enrolled more than 15% of their older patients on clinical trials. Concern regarding increased toxicity was the barrier cited in 50% of cases. Other reasons included comorbid medical conditions, patient or family members' preference, the older patient's lack of social support, time required for patient enrollment and follow-up, and physician preference for conventional treatment.
A total of 156 physicians from 10 CALGB institutions were surveyed regarding their perception of the most important barriers to accrual of older adults. The most commonly cited barriers included concern regarding the effect of comorbid conditions on the ability to tolerate treatment, concern regarding the patient's ability to comprehend a complicated treatment trial resulting in poor compliance, concern regarding treatment toxicity in the older patient, and the older patient not meeting the specified eligibility criteria. The most frequent suggestions for interventions to improve the accrual of older adults to clinical trials included making personnel available in clinic to explain the trial and providing educational materials to the physician concerning treatment toxicity in older patients.
Based on these suggestions, investigators from the CALGB Cancer in the Elderly Committee tested the impact of an educational intervention to improve the accrual of older adults to clinical trials. In this study by Kimmick et al, CALGB main member institutions were randomized to receive either standard information or an educational intervention. The standard information consisted of access to the CALGB website and periodic notification about protocols. The educational intervention consisted of a geriatric oncology educational seminar, educational materials, a list of protocols available to older patients placed in the patient's chart, and monthly e-mail and mail reminders. Despite this intervention, study results demonstrated that accrual of older patients to clinical trials decreased in comparison to baseline, and there was no accrual improvement in the intervention arm compared to the control arm in the first year (36% vs 32%; P = .35) or second year (31% vs 31%; P = .83). The authors postulated that a more intense, multifaceted intervention was needed to improve the accrual of older adults in clinical trials.
While older patients are underrepresented in clinical cancer trials for multifaceted and complex reasons, among the most commonly cited reasons by physicians for not enrolling older adults on clinical trial is a concern about treatment toxicity.[4,11] Previous studies disagree on whether older patients are truly at increased risk; however, these studies likely enrolled a select group of older patients who were deemed fit for participation in a clinical trial, making it difficult to extrapolate study results to the population as a whole.[13-17]
Future studies are needed to address ways for improving clinical trial accrual of older patients. A significant barrier to enrollment is the physician's concern regarding the age-related risk of inordinate toxicity. In order to address this concern, further trials are needed to evaluate the toxicity and efficacy of cancer therapies in older adults (taking into account age-related changes in physiology and pharmacokinetics that may affect treatment tolerance), to evaluate the role of supportive care measures that can improve treatment tolerance, and to pinpoint factors other than chronologic age that may be independent predictors of morbidity and mortality.