Comparing Representativeness of Race and Ethnicity in Real-World Evidence, Randomized Clinical Trial, and Surveillance, Epidemiology, and End Results (SEER) Cancer Patient Data

Speaker(s)

Balanean A1, Gentile D1, Brown M1, Asgarisabet P2, Chopra D1, Swain R2, Schuler T3, Klink A1, Savill K4, John W5, Feinberg B6
1Cardinal Health Specialty Solutions, Dublin, OH, USA, 2Cardinal Health Real-World Evidence and Insights, Dublin, OH, USA, 3Cardinal Health Specialty Solutions, Issaquah, WA, USA, 4Cardinal Health Specialty Solutions, EL DORADO HILLS, CA, CA, USA, 5Cardinal Health Specialty Solutions, Durham, NC, USA, 6Cardinal Health Specialty Solutions, ATLANTA, GA, USA

Presentation Documents

OBJECTIVES: In the United States (US), randomized clinical trial (RCT) participants tend to be younger, healthier, less racially/ethnically diverse, and predominantly male compared with their real-world patient counterparts. Using contemporaneous time periods spanning 2017–2022, we assessed representation of race and ethnicity of patients with advanced melanoma, breast, lung, liver, or kidney cancer in real-world evidence (RWE) chart reviews, corresponding RCTs, and the Surveillance, Epidemiology, and End Results (SEER) database, which represents cancer diagnoses in the US irrespective of treatment.

METHODS: From RWE studies assessing first-line (1L) cancer treatment, we selected tumor types on which recent RCTs had also been conducted in the 1L setting and built SEER patient cohorts around their eligibility criteria. We then compared demographics of 83,298 patients with advanced breast, lung, liver, kidney, or melanoma skin cancer (RWE: n=2,980, 3.6%; RCT: n=6,168, 7.4%; SEER: n=74,150, 89.0%). Race and ethnicity were compared across data sources using 2-sided chi-square or Monte-Carlo exact tests.

RESULTS: Total representation of Black patients was 20.0% in RWE, 1.9% in RCT, and 12.9% in SEER studies. Across tumor types, Black patients represented 25.3% (RWE), 2.9% (RCT), 13.4% (SEER) in breast; 24.0% (RWE), 1.3% (RCT), 12.8% (SEER) in lung; 17.2% (RWE), 1.6% (RCT), 4.8% (SEER) in kidney; and 7.3% (RWE), 0.0% (RCT), 1.9% (SEER) in melanoma. Hispanic representation was 11.7% in RWE, 3.4% in RCT (2 trials), and 15.2% in SEER data (P<.001 for all comparisons).

CONCLUSIONS: Significant differences were observed among the data sources. Despite known limitations (sparse Hispanic reporting in RCTs, differential SEER participation by US region, and possible patient overlap), underrepresentation of Black race and Hispanic ethnicity in RCTs is noteworthy. Hispanic ethnicity was collected in only 25% of RCTs assessed. As clinical decision-making and payer reimbursement are largely driven by RCTs, research should leverage RWE to improve representativeness and generalizability.

Code

RWD95

Topic

Health Policy & Regulatory, Real World Data & Information Systems, Study Approaches

Topic Subcategory

Distributed Data & Research Networks, Health Disparities & Equity, Meta-Analysis & Indirect Comparisons, Registries

Disease

No Additional Disease & Conditions/Specialized Treatment Areas