Age-based disparities in clinical trials supporting FDA approval of therapies for solid tumors.

Abstract

e18534 Background: Cancer of any site is most frequently diagnosed among adults age 65 years and older, however older adults have historically been under-represented in clinical trials for oncology drugs (difference in median age [DMA] −6.49 years among clinical trials for breast, prostate, lung and colorectal cancer from 1994-2015 [Ludmir et\xa0al, JAMA Oncology 2019]). Representation in registrational trials, which are the basis for drug approval and inform the prescribing information, is of particular importance as age‐related differences may affect response and toxicity. We conducted an analysis of the trials supporting recent Food and Drug Administration approvals of new therapies for the treatment of solid tumors to assess the median age of participants and compare it to the median age of patients in the real world for each malignancy. Methods: Prescribing information for novel therapies that received primary approval for solid tumor indications between 2015 and 2019 was reviewed, and the median age in the evaluable population of the supporting clinical trials was abstracted. Median age estimates among adults for each indication in the general population were obtained from the Surveillance, Epidemiology and End Results database or the published literature. The DMA was calculated for each trial by subtracting the median age in the real-world population from the median age in the trial. Characteristics of trial protocols were obtained from approval packages accessed in the Drugs@FDA database. Data are presented using descriptive statistics. Results: A total of 35 solid tumor drugs were approved based on 38 trials, with 15859 patients in the evaluable populations. Less than half of the trials were phase 3 (16; 42%) or were randomized with a control group (18; 47%). On average, the median age of the trial participants was 2.6 years younger than the median age of the disease populations (95% CI: -1.4, -3.9 years; p < 0.01). For indications with > 1 trial, those with the greatest disparities were in melanoma (mean DMA -6.5), urothelial carcinoma (-5.8), and lung cancer (-4.1). The DMA among phase 3 trials was -1.8, vs. -3.2 among phase 1 and 2 trials. None of the protocols for the trials specified upper age limits, however 23 (61%) required an ECOG performance status (PS) ≤1, and all trials had eligibility restrictions related to organ function or comorbidities. Conclusions: In this contemporary analysis of registrational trials across all solid tumor indications, we found that the median age of clinical trial participants was nearly 3 years younger than their real-world counterparts. No trial protocols had upper age limits, however most restricted PS, organ impairment or the presence of comorbidities, which may have impacted older patients’ eligibility to participate. Although inclusion of older adults in clinical trials appears to have improved in recent years, changes to trial design may help to ensure adequate representation.