Race and Gender Representation in Myelodysplastic Syndrome Clinical Trials in the US: A Cohort Study

Abstract

Introduction: Clinical trials in the US frequently under-enroll racial/ethnic minorities, and demographic reporting of minority enrollment is variable and often incomplete. Underrepresentation of minorities in clinical trials deprives these patients (pts) of the benefits of novel therapies and impacts the generalizability of clinical trial findings. Little is known about the disparities in enrollment of racial and ethnic minorities in myelodysplastic syndromes (MDS) clinical trials in the US and their trends over time. Methods: We used the aggregate analysis of ClinicalTrials.gov database to obtain records of interventional, randomized phase II and phase III studies conducted in US that reported completed results to the database using the search terms “myelodysplastic syndromes” or “MDS” with start and completion dates between January 1, 2000, to December 31, 2022. “Race/ethnicity” was adapted in accordance with US Census and Department of Health and Human Services guidelines and grouped into five categories for this analysis - White, Black/African American, American Indian or Alaska Native (AI/AN), Asian, and Other. We analyzed the reporting of race/ethnicity in clinical trials benchmarked with the passage of federal mandatory reporting requirements in September 2007 and January 2017. The proportions of enrollees by each racial group in all the MDS trials reporting on all 5 racial groups were computed for the entire study period and the distributions of these proportions were plotted in the density plots (Fig 1). Additionally, we calculated the odds ratios (OR) with 95% CI of enrollment by race and gender in clinical trials (Fig 2). All calculations were done using R version 4.2.0. Results: We identified 299 MDS clinical trials using our search criteria. After excluding phase I/II non-randomized clinical trials and trials with less than 20 pts, we included a total of 206 clinical trials (174 phases II & 32 phase III) for analysis. Of the 21,645 pts initially enrolled in the 206 trials, 20,709 pts with complete data were analyzed. With the institution of federal reporting requirements in the late 2000s, enrollment reporting by race and ethnicity has vastly improved, with 100% of MDS clinical trials reporting these demographics since 2017. In trials reporting race enrollment for all five groups, Whites had the highest enrollment proportions in all trials conducted within the study period, with a median enrollment of 85.8% (Fig 1). This level of clinical trial representation exceeded the estimates of proportion of Whites among the US population (70.3%, p < 0·001). In contrast, Blacks, AI/AN, Asian and other racial categories were significantly underrepresented relative to their US populations ( p < 0·001). Figure 2 shows a forest plot of OR of enrollment by race and gender. Blacks, Asian and AI/AN had significantly lower odds of enrollment compared to Whites. Females had significantly higher odds of enrollment compared to males. Conclusion: Our analysis highlights substantial disparities in representation of racial minorities in MDS clinical trials in the US. Unless efforts are taken to systematically address these enrollment disparities, MDS outcomes for these historically disadvantaged groups will continue to lag due to incomplete understanding of biologic and individual factors that impact treatment effectiveness in these groups.