Race-Associated Genomic Correlates of Therapeutic Response in African American Patients With Non–Small-Cell Lung Cancer

African American individuals are disproportionately affected by lung cancer in terms of incidence and mortality. In oncogene-driven non-small-cell lung cancer (NSCLC), emerging evidence indicates that underlying molecular heterogeneity, which can be affected by ancestry, contributes to variable drug sensitivity and therapeutic responses. The purpose of this study was to evaluate race-associated differences in reported treatment decisions, therapeutic outcomes, and molecular features in KRAS- and EGFR-mutant NSCLC.This is a retrospective study using real-world clinical-genomic data from health systems in the United States to evaluate race-associated outcomes in advanced-stage KRAS- or EGFR-driven NSCLC. Our overall objectives were to evaluate race-associated therapeutic outcomes and to describe molecular features in non-Hispanic Black (NHB) and non-Hispanic White (NHW) patients with NSCLC.A total of 723 NSCLC patients with KRAS and 315 patients with EGFR oncogenic mutations were evaluated. In KRAS-mutant patients, variable outcomes were observed in NHB and NHW patients on the basis of receiving chemotherapy alone or in combination with immune checkpoint inhibitors. NHB patients received treatment at significantly lower rates compared with NHW patients. In the EGFR-mutant cohort, NHB and NHW patients received EGFR-targeted agents at similar rates, and overall survival was not significantly different. Race-associated differences in molecular features included a higher frequency of TP53 comutation in KRAS-mutant NHB patients and higher prevalence of EGFR G719S subtype in NHB patients.In a real-world cohort of patients with NSCLC, we identified race-associated differences in therapeutic outcomes and described molecular characteristics in NHB and NHW patients with NSCLC. To proactively identify patients most likely to respond to systemic therapies, a more comprehensive approach is needed to help guide therapy selection in individualized patient populations.