Comparison of sociodemographic characteristics of a phase 1 clinical trial population at an NCI-designated comprehensive cancer center in the Southeast to catchment area.

Abstract

e18591 Background: Racial and ethnic minority populations are consistently under-represented in oncology clinical trials despite comprising a disproportionate share of cancer burden. Due in part to difficulties associated with participation, phase 1 trials pose a unique challenge and opportunity for minority inclusion. Here we examine the sociodemographics of phase 1 clinical trial patients at an NCI-designated Comprehensive Cancer Center compared to all patients treated at the center, patients with new cancer diagnoses in metropolitan Atlanta, and the state of Georgia (GA). Methods: Patients enrolled on phase 1 trials at the Winship Cancer Institute (WCI) from 2015-2020, identified from a data warehouse, were compared to new patient registrations at WCI from 2015-2019. Patients with cancer in metro Atlanta and GA were identified from the Surveillance, Epidemiology, and End Results (SEER) Program: Nov 2018 Submission. Covariates for the phase 1 and institutional cohort included sex, race, ECOG PS, insurance, and age at consent. Covariates for SEER patients included sex, race, and age at diagnosis. Summary statistics are reported for categorical variables using frequencies and percentages, and for continuous variables using mean, median, standard deviation, and range. One-sample proportion tests were utilized to compare observed demographic proportions on phase 1 trials with proportions calculated from new patient registrations at WCI and from SEER case data. Results: From 2015-2020, 2325 patients (43.4% F, 56.6% M) signed consent for phase 1 trials. Grouped race distribution was 70.3% White, 26.2% Black, 3.5% Other. Insurance distribution was 42.9% private, 48% government, 9.1% uninsured/other. Mean age at consent was 61.7 years (MD 63, ran 19-92). Of new patient registrations at WCI in 2015 ( N= 12358) (49.7% F, 50.3% M), grouped race distribution was 64.0% W, 28.2% B, 8% O. Atlanta 2015 SEER patients ( N= 31101) (50.3%F, 49.7%M) showed grouped race distribution 58.4% W, 37.2% B, 4.3% O. Mean age at diagnosis was 62.9. GA 2015 SEER patients ( N= 99487) (48.6%F, 51.4%M) showed grouped race distribution 71.2% W, 26.7% B, 2.1% O. Mean age at diagnosis was 64.2. The race and sex distribution of phase 1 patients was significantly different than proportions calculated from new patient registrations at WCI and SEER data from patients in Atlanta ( p< 0.001). Sex distribution was significantly different than GA SEER patients. Conclusions: Phase 1 patients were significantly more likely to be white and male, compared with patients treated at the cancer center, as well as patients with cancer in Atlanta. Race distribution was not significantly different from patients with cancer in GA. Our intent is to characterize existing disparities in order to increase representation of patients from racial and ethnic minority backgrounds in phase 1 clinical trials.