The impact of food and drug administration-based eligibility criteria on demographic representation in current phase III clinical trials for newly diagnosed breast cancer.

Abstract

e23021 Background: Restrictive eligibility criteria may limit demographic representation in clinical trials. Beginning in 2020, the US Food and Drug Administration (FDA) published draft guidance on broadening eligibility criteria by aligning them with investigational agents’ safety data. The extent to which this guidance has been adopted in current curative-intent phase III breast cancer trials, as well as its potential to impact demographic representation, is unclear. Methods: Eligibility criteria from US trials registered on ClinicalTrials.gov from 10/2023-10/2024 were catalogued. “FDA” guidance-based eligibility criteria were developed from available safety data using FDA drug labels and study protocols. A cohort of newly diagnosed patients treated with curative-intent therapy from 2006-2024 was identified using a real-world electronic health records database (TriNetX; Cambridge, MA) comprised of 87 US healthcare organizations. The primary endpoint was the median difference in the proportion of patients eligible using FDA versus actual criteria. Secondary endpoints were changes in the number of criteria and in eligibility by race/ethnicity. Endpoints were summarized descriptively and differences tested by univariate analysis. Results: There were 14 trials identified and 2990 patients in the analysis. Median age was 54 years, 10.7% were Hispanic, 2.5% Non-Hispanic (NH) Asian (NHA), 16.9% NH Black (NHB), 59.7% NH White (NHW), and 10.0% Unknown/Other (U/O). The median number of criteria per trial decreased from 13 (actual) to 4 (FDA). Cohort eligibility increased 12.8% (95% CI: 9.9, 18.2; p = 0.002) when moving from actual (86.7%) to FDA (98.2%) criteria. Changes in eligibility proportions when moving from actual to FDA criteria by trial ranged from -0.7-17.6%, with significant increases for 12 trials. Eligibility increased most through broadening the following criteria: creatinine clearance (6.6%), hemoglobin (1.6%), ALT (1.3%). Using actual criteria, median eligibility differences by race and ethnicity, compared to NHW (89.5% eligible), were: Hispanic (-3.1%), NHA (3.4%), NHB (-9.6%), O/U (-6.5%; all pairwise p < 0.002). Median differences using FDA criteria, compared to NHW (98.9% eligible) were: Hispanic (0.1%; p = 0.8), NHA (-2.1%; p = 0.002), NHB (-1.8%; p = 0.003), O/U (-3.4%; p = 0.002). Changes in median eligibility proportions between actual and FDA criteria for race/ethnic groups relative to NHW are shown in the Table. Conclusions: FDA guidance has not been widely adopted in current phase III trials. Utilizing FDA guidance will likely improve trial representation, especially for minoritized racial and ethnic groups. Race/Ethnicity NHB NHA Hispanic O/U Median % change compared to NHW % change* 8.1 -4.9 3.1 3.2 p-value 0.007 0.001 0.001 0.007 *Positive change indicates relatively more eligibility.