Equity and representation in oncology clinical evidence for FDA-approved treatments and the impact of COVID-19.

Background: Pivotal clinical trials form the evidentiary foundation for FDA oncology drug approvals, yet persistent disparities in participant demographics may limit the generalizability and equity of cancer care. This study aimed to quantify demographic disparities in pivotal trials that supported FDA oncology drug approvals, and to identify trial-level characteristics associated with the underrepresentation of women, racial and ethnic minorities, and older adults. We also assessed the impact of COVID-19 on participant representation. Methods: We systematically identified pivotal trials supporting FDA oncology drug approvals from January 2018 through November 2024 using the FDA Hematology/Oncology Approvals database. Demographic data were extracted for gender, race, ethnicity, and age. SEER 21 data were used to derive population-based estimates. Enrollment incidence ratios (EIRs) were calculated for gender, race, ethnicity, and older adults (EIR <1 indicating underrepresentation). Meta-regression was used to evaluate temporal trends and associations between trial characteristics and representation disparities. Subgroup analyses were conducted for pediatric trials, accelerated approvals, and other trial-level factors. Results: Among 329 included trials enrolling over 164,000 participants, women, older adults, Black, and Hispanic individuals were consistently underrepresented. Pooled EIRs indicated significant underrepresentation of female (EIR, 0.83; 95% CI, 0.78-0.88), Black (EIR, 0.26; 95% CI, 0.22-0.30), Hispanic (EIR, 0.50; 95% CI, 0.44-0.57), and older participants (EIR, 0.86; 95% CI, 0.83-0.89). Meta-regression showed modest gains in female representation over time (β=0.02, p=0.03), but no significant improvement for Black or Hispanic participants. Older adult representation declined slightly. Trial-level factors associated with underrepresentation included industry sponsorship, use of overall survival as the primary endpoint, and accelerated approval designation. Pandemic-period EIRs were largely similar to non-pandemic EIRs, with lower representation for older adults and White participants. Conclusions: Despite recent regulatory efforts, demographic disparities persist in pivotal oncology trials supporting FDA approvals. Female, Black, Hispanic, and older adults remain underrepresented, limiting the applicability of trial findings to real-world cancer populations. Greater efforts are needed to design inclusive trials, enforce diversity mandates, and ensure equitable evidence generation in oncology drug development.