Identifying eligibility criteria that perpetuate race/ethnic disparities in acute myeloid leukemia (AML) clinical trial participation.

1589 Background: The extent to which eligibility criteria perpetuate race/ethnic disparities in adult AML clinical trial participation is unknown. Methods: This multicenter cohort study counterfactually assessed race/ethnic disparities in trial eligibility between Non-Hispanic Whites (NHW) and Minoritized Individuals (MI: Hispanic, Black, Asian, and/or Other) diagnosed with AML between 2010-2022 at six hospitals in Chicago, IL and one in Boston, MA. All newly diagnosed patients were catalogued; those with complete data were analyzed. Eligibility criteria were obtained from 190 front-line phase II/III therapeutic studies active during the same period (1). Common sets of criteria for intensive and non-intensive trials were generated based on prevalent criteria and medians of any associated limits (e.g., QTc ≤480 ms). Proportions of MI and NHW that would have been eligible based on these common criteria sets were assessed using X2 tests with odds ratios (OR). Mann-Whitney U tests assessed differences in the median number of trials for which each MI and NHW patient would have been eligible. Analyses with MI groups disaggregated were also performed. Results: There were 1283 patients (405 MI, 878 NHW) with a median age of 60 years (57 MI, 61 NHW; p<0.001); 44% were female. According to the common set of 23 intensive trial criteria that was generated, similar proportions of MI and NHW were eligible (24.7% and 24.5%, OR 1.01, 95% CI [0.77,1.33]). With the age limit criterion removed, fewer MI than NHW were eligible (42.7% and 36.0%, OR 0.76, 95% CI [0.59,0.96]). In this set of criteria, prior malignancy (excluding 22.5% and 17.5%, p=0.04), prolonged QTc (17.0% and 8.7%, p<0.001), and coronary artery disease (CAD; 8.6% and 4.1%, p=0.002) generated the largest disparity individually. According to the non-intensive set (14 criteria), fewer MI were eligible than NHW (42.2% and 54.4%, OR 0.61, 95% CI [0.48,0.78]); there was no age criterion in this set. Prior malignancy (22.5% and 17.5%, p=0.04), prolonged QTc (20.5% and 12.8%, p<0.001), and hepatitis B infection (4.0% and 1.0%, p=0.001) generated the largest disparities in this set. Overall, individuals were eligible for a mean of 51.6% of the 190 trials (median: 98 trials, IQR [59,115]). MI were eligible for 6.8% fewer trials than NHW (median difference: 13 trials [88 and 101], p<0.001). Asian, Black, and Other were eligible for 7.7%, 4.8%, and 16.9% less trials than NHW, respectively (all p<0.05). Criteria with the largest individual disparities in median eligibility were lower age limits, hepatitis B, and CAD (all p<0.01). Conclusions: Eligibility criteria disparately exclude minoritized individuals from AML clinical trial participation. Based on ASCO recommendations ( 2), several of the criteria driving disparate eligibility (hepatitis, prior malignancy) are modifiable or removable. 1. Hantel, Haematologica, 2023. 2. ASCO, Clin Cancer Res, 2021.