FDA Analysis of Race and Ethnic Enrollment of Clinical Trials for Drug Registration in Adult Patients for Leukemia and Myelodysplastic Syndrome

Introduction: Underrepresentation of racial and ethnic minorities in cancer clinical trials is a well-documented issue and contributes to disparities in outcomes. Despite greater awareness and development of initiatives to increase diversity of cancer clinical trials, enrollment of racial and ethnic minorities remains low. Since 2012, FDA has approved 20 different drugs in adult patients with acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), chronic myeloid leukemia (CML), and myelodysplastic syndromes (MDS). Substantive scientific progress has been made with drug approvals for these malignancies, however the representativeness of racial and ethnic patient enrollment in these clinical trials is not well understood. This study analyzed the racial and ethnic enrollment in pivotal clinical trials that led to FDA-approval of therapies for ALL, AML, CML, and MDS in the context of population-based incidence data.

Methods: Demographic data was collected from registrational trials submitted to the FDA from 2012-2023 to support the approval of drugs in adults with ALL, AML, CML, and MDS. Only patients who were enrolled and treated were included in the analysis. Data was additionally divided into two-time intervals based on the date of submission: 2012-2017 and 2018-2023. The distribution of race and ethnicity for patients enrolled on trials was compared with the corresponding United States (US) population-based distribution of newly-diagnosed adult (≥20 years) ALL, AML, CML, and MDS cases. Race and ethnicity distributions of incident US cancer cases diagnosed 2012-2021 were obtained from the North American Association of Central Cancer Registries (NAACCR).

Results: Twenty-nine (29) registrational trials enrolling a total of 8,177 adult patients were included in our analysis. This included 14 trials in AML, 8 in ALL, 6 in CML, and 4 in MDS. Overall, 5571 (68%) patients identified as White, 971 (12%) identified as Asian, 256 (3.2%) identified as Black, 242 (3%) identified as an “other” race, 47 (0.6%) identified as “Latino”, 19 (0.2%) identified as Native American, and 11 (0.1%) patients identified as more than one race. Race was missing or not reported for 1060 (12.6%) patients . There were 555 (6.8%) patients that identified as Hispanic or Latino ethnicity. Between 2012-2017 there were a total 14 registrational trials enrolling 4429 patients and between 2018-2023 there were a total of 15 registrational trials enrolling 3,748 patients. During the two-time intervals, there was a relatively stable proportion of White (68% vs 70%), Black (3.4% vs 2.9%), Latino (7.4% vs. 6.1%), and Native American (0.2% vs. 0.3%) patients enrolled on registrational trials, respectively. The proportion of Asian patients (8.7% vs. 15.7%) enrolled on pivotal trials nearly doubled during the time intervals. When comparing trial enrollment numbers to population -based NAACCR data, there was an underrepresentation of Black, Latino, and Native American patients in all 4 diseases and a relative overrepresentation of Asian patients. For example, Asian patients account for 3.1% of the newly diagnosed CML cases in NAACCR but accounted for over 15% of the patients in registrational trials for CML.

Conclusions: Our analysis of registrational trials in adults with ALL, AML, CML, and MDS demonstrated that Asian patients were adequately represented, and their enrollment increased over the last decade, whereas Black, Latino, and Native American patients were underrepresented, and their rates of enrollment remained unchanged or decreased throughout the time period. It is essential thatthe demographics of patients enrolled in cancer clinical trials are representative of the US intended use population to ensure that safety and efficacy data collected from these trials can be understood for all patients. The limitations of this study are that analyses only included enrolled and treated patients and only evaluated pivotal trials that were used to support an indication for approval. Nevertheless, our analysis demonstrates that enrollment of Black, Native American, and Latino patients to registrational cancer trials remains a substantial issue. The FDA remains committed to collaboration with investigators and sponsors to find innovative solutions to increase enrollment of underrepresented minorities in cancer clinical trials.

No relevant conflicts of interest to declare.