Disparities in Tumor Mutational Burden, Immunotherapy Use, and Outcomes Based on Genomic Ancestry in Non–Small-Cell Lung Cancer

PURPOSE: In patients with advanced non–small-cell lung cancer (aNSCLC), tumor mutational burden (TMB) may vary by genomic ancestry; however, its impact on treatment outcomes is unclear. This retrospective, observational study describes treatment patterns of patients with aNSCLC by genomic ancestry and electronic health record (EHR)-reported race and/or ethnicity and evaluates differences in TMB, cancer immunotherapy (CIT) access, and treatment outcomes across racial and ancestral groups. METHODS: Patients diagnosed with aNSCLC after January 1, 2011, were selected from a real-world deidentified clinicogenomics database and EHR-derived database; continuously enrolled patients were evaluated. Race and/or ethnicity was recorded using variables from the EHR database; genomic ancestry was classified by single-nucleotide polymorphisms on a next-generation sequencing panel. A threshold of 16 mutations per megabase was used to categorize TMB status. RESULTS: Of 59,559 patients in the EHR-derived database and 7,548 patients in the clinicogenomics database, 35,016 (58.8%) and 4,392 (58.2%) were continuously enrolled, respectively. CIT use was similar across EHR-reported race groups, ranging from 34.4% to 37.3% for non-Hispanic Asian and non-Hispanic Black patients, respectively. TMB levels varied significantly across ancestry groups (P < .001); patients of African ancestry had the highest median TMB (8.75 mutations per megabase; interquartile range, 4.35-14.79). In patients who had received CIT, high TMB was associated with improved overall survival compared with low TMB (20.89 v 11.83 months; hazard ratio, 0.60; 95% CI, 0.51 to 0.70) across genomic ancestral groups. CONCLUSION: These results suggest that equitable access to next-generation sequencing may improve aNSCLC outcome disparities in racially and ancestrally diverse populations.