With an ageing population, where 60% of cancer diagnoses are made in patients over the age of 65, the inclusion of older patients in clinical trials is a priority [7, 11]. In our evaluation of phase III trials assessing systemic therapy for advanced gastroesophageal malignancies, 32% of the studies excluded patients based on older age alone. Furthermore, the median age of patients included was just 62, and those aged over 65 made up only 36% of the total study population. Consequently, it is difficult to apply this evidence to our everyday clinical practice, where we frequently encounter older, frailer and more complex patients than the individuals included in these trials. This may lead to both suboptimal treatment of some ‘fitter’ patients and over-treatment of those who may be more frail than their biological age, resulting in detrimental patient outcomes.
Reasons for poor representation of older adults in clinical trials are complex and multifactorial. They relate to a mix of patient, physician and system factors. There have been a few studies examining patient perspectives and attitudes towards clinical trial participation. Townsley et al. conducted a study focusing on understanding the attitudes of elderly patients with cancer towards clinical trial enrolment [5]. Over 80% of respondents were between the ages of 70 and 79 and the majority of patients stated they would participate in clinical trials to prevent or screen for cancer, to compare a new drug to a 'standard' drug, and 70% would participate in clinical trials to test a new drug in situations where there is no 'standard' drug [5]. However, while most were willing to consider participation when offered, few elderly patients actively sought clinical trials and overall were less well informed regarding the availability of relevant clinical trials [5]. A study by Yellen et al. looked at age and clinical decision-making in oncology patients using clinical vignettes in an interview situation. They found that older patients were as likely as the younger cohort to agree to chemotherapy for both curative and disease control purposes [12]. Ayodele et al. carried out a similar study to compare the attitudes of younger and older patients to clinical trials and found that older patients were as willing as younger patients to participate in clinical trials, yet significantly less were enrolled [13]. In reality, the inclusion of elderly patients is not always straightforward and may not be feasible, due to comorbidities, cognitive issues or social circumstances. Multiple clinic visits, paperwork and travel to medical appointments are well-documented barriers to trial accrual, and reducing trial participation burden is an area where further progress is needed [11].
Misconceptions among physicians can also act as a barrier to trial enrollment. In a survey of American oncologists, 50% indicated that they declare patients unsuitable for clinical trials based on age alone [14]. Another study examining barriers to clinical trial participation in older women with breast cancer, found that the physicians’ perceptions about age and tolerance of toxicity were the greatest obstacle to enrolling older women onto trials [15]. Sedrek et al. carried out semi-structured interviews with 44 medical oncologists (24 academic-based and 20 community-based) in an attempt to explore oncologists’ perceptions of barriers to clinical trial enrollment of older adults with cancer. The most common barriers identified by oncologists were stringent eligibility criteria and concerns for treatment toxicities [16]. A better awareness of clinical trials must be promoted amongst healthcare providers in order to increase older adult participation.
Efforts to increase the representation of older adults has been recognised as a priority by a number of international oncological organisations. The International Society of Geriatric Oncology (SIOG) published updated guidance in January 2021 relating to ‘Priorities for global advancement of care for older adults with cancer’ [17]. This policy document discusses priorities relating to education, clinical practice, research, and collaborations in an attempt to improve healthcare for this rapidly growing patient cohort.
In February 2021, the American Society of Clinical Oncology (ASCO) and the Friends of Cancer Research (Friends) group issued new recommendations to further broaden eligibility criteria for clinical trials, with the aim of expanding patient access [18]. Although the underlying rationale for eligibility criteria is to protect the safety of trial participants and to exclude patients who may have an unacceptably high risk of toxicity, this must be balanced with need to include a representative group of individuals [18, 19]. Interestingly, an analysis conducted from CancerLinQ Discovery® (CancerLinQ’s deidentified real-world data product for researchers) found that the number of lung cancer patients potentially suitable for clinical trials almost doubled, when three common eligibility criteria (renal function, presence of brain metastases, history of prior malignancy) were relaxed [20].
In addition to the broadening of eligibility criteria, other suggestions to promote inclusion of older patients in trials focus on addressing the study design, statistical analysis and reporting of trial results. These include assessing more ‘age-relevant’ endpoints such as functional status and quality of life, as well as the traditional outcomes of progression free survival (PFS) and overall survival (OS). Similarly, expanding treatment-related toxicities to include adverse effects relevant to elderly patients, such as incontinence and falls has been recommended [21, 22, 11]. In terms of reporting of results, it is advised that age-specific subgroup analyses should be powered to detect any age related differences, and in situations where sub-group analysis is not pre-specified, any conclusions should be described as exploratory [23]. Tackling the above issues will help to strengthen and develop our evidence base and allow better decision making for complex, older patients.
Clinical trials can also specifically focus on older adults with cancer and indeed there have been a number of ‘elderly-specific’ trials in the last few years, highlighting that it is possible to conduct phase III trials in this patient population [24–26]. In the context of gastroesophageal cancer, the GO2 trial was a large phase III study which included 514 older patients with advanced gastroesophageal cancer who were unfit for full dose chemotherapy and aimed to find the optimal dosing strategy. Patients were randomised (1:1:1) to oxaliplatin and capecitabine (xelox) on 3 different dose schedules. The lowest dose demonstrated decreased rates of toxicity and improved quality of life, without shortening survival [26]. These studies illustrate that large randomized studies on older patients are feasible and contribute to creating a body of evidence that guides clinical decision making in this setting.
Overall, there is a slow shift towards including older and multimorbid patients in clinical trials, and certainly the creation of ‘elderly-specific’ trials as mentioned above is important and encouraging. Interestingly, in our study, when we analysed the patterns of enrollment over the time, we found that age limits were much more common pre-2003. Perhaps this was influenced by the publication by Hutchins et al. in the New England Journal of Medicine in 1999, ‘Underrepresentation of patients ≥ 65 in cancer treatment trials.’ In this study, data on 16,396 patients enrolled in clinical trials between 1993 and 1996 particularly focusing on sex, race and age, were analysed and compared with rates of cancer in the general population, according to the US Census and the National Cancer Institute [3]. Efforts had been made previously to address the under-representation of women and minority ethnic groups, and the overall proportions of these cohorts were found to be similar. In contrast, patients 65 years or older were dramatically under represented (25% versus 63%, p < 0.001) [3]. Since then a number of policies have been developed in an attempt to remedy this. As part of our study, we accessed clinicaltrials.gov and we found that none of the currently active and enrolling phase III clinical trials in advanced gastroesophageal cancer specified an upper age limit for inclusion. It is encouraging that trials have reduced the explicit upper age limits and now we must move to focusing on the other barriers that disproportionately exclude older individuals.
In conclusion, recent years have seen modest improvements in clinical trial protocols, with many no longer specifying restrictive age criteria. With an ageing population there is a growing need to include older, frailer patients who are more reflective of the ‘real world’ oncology patient in clinical trials. We must advocate for a more inclusive culture and strive to generate the best data and evidence that will allow us to make informed treatment decisions for our patients.