Racial disparities in oncology clinical trials by absolute neutrophil count eligibility criteria: A single center retrospective analysis.

Abstract

1583 Background: Black individuals have been historically underrepresented in clinical trials. The Duffy-null phenotype leads to clinically insignificant lower absolute neutrophil counts (ANC) and is found in 66% of non-Hispanic Blacks (NHB) and < 1% non-Hispanic Whites (NHW) in the United States. Although the ANC lower limit of normal for Duffy-null individuals is estimated to be 1,210/µL, little is known about the impact of ANC criteria on racial underrepresentation in clinical trial eligibility and enrollment. Methods: We conducted a single-center retrospective study of patients newly diagnosed with five cancers (lung, breast, prostate, colorectal, non-Hodgkin lymphoma) between 1988-2024. Baseline organ function data (Hgb, ANC, platelets, AST, ALT, creatinine, and bilirubin) were compared between NHB and NHW. We also identified ANC thresholds above which eligibility proportions differ between NHB and NHW. Results: We identified 23,854 patients. 14,585 had available baseline lab data within one year prior to starting anti-cancer therapy. The table lists the percentage of NHW and NHB who meet ANC eligibility criteria at thresholds of 1,000, 1,500, and 2,000/µL. Among patients with breast cancer, the median ANC was 3,500 for NHB and 3,890 for NHW (p = 0.63) and fewer NHB were eligible at all ANC thresholds of 1,500/µL or greater (94.5% NHB vs 97.2% NHW, p = 0.03). For those with prostate cancer, the median ANC was 3,780 for NHB and 4,450 for NHW (p = 0.03) and significantly fewer NHB were eligible at all ANC thresholds of 1,400/µL or greater (97.6% NHB vs 99.6% NHW, p = 0.04). For those with non-Hodgkin lymphoma (NHL), the median ANC was 4,260 for NHB and 4,020 for NHW (p = 0.34) and significantly fewer NHB were eligible at all ANC thresholds of 1,600/µL or greater (90.4% NHB vs 97.1% NHW, p = 0.008). Patients with lung cancer (ANC NHB 4,820 vs NHW 4,920; p = 0.85) and colorectal cancer (ANC NHB 3,930 vs NHW 4,150; p = 0.27) did not have significant ANC eligibility differences between NHB and NHW. No significant differences were found for other baseline labs. Conclusions: Our findings provide further evidence that ANC criteria may contribute to differences in clinical trial eligibility between NHB and NHW for breast cancer, prostate cancer, and NHL. Further work is needed to identify optimal cutoffs for each disease group. Limitations include lack of Duffy status for most patients and race is an imperfect approximation. Future clinical trials should proactively address these differences by using ANC eligibility criteria based on Duffy phenotyping. Cancer(n=NHB; NHW) NHB ANC >2000 NHW ANC >2000 p value* NHB ANC >1500 NHW ANC >1500 p value* NHB ANC >1000 NHW ANC >1000 p value* Lung(n=73; 726) 90.4% 94.4% 0.27 93.2% 97.4% 0.10 95.9% 99% 0.08 Breast(n=237; 2,905) 86.1% 92.7% <0.001 94.5% 97.2% 0.03 98.7% 99% 0.99 Prostate(n=124; 961) 88.6% 97.8% <0.001 96.8% 99.4% 0.02 100% 99.9% - Colorectal(n=80; 800) 87.5% 93.5% 0.08 100% 98.6% - 100% 99.9% - NHL(n=103; 1,414) 76.7% 85.6% 0.02 87.4% 90.2% 0.39 94.2% 93.6% 1.0 *Calculated by Chi-squared tests.