Investigating racial inequities of circulating tumor DNA (ctDNA) use in patients with non-small cell lung cancer: A real world analysis.

Abstract

9133 Background: The use of ctDNA has emerged as an invaluable adjunct to tissue-based molecular testing, though it remains unclear if racial inequities associated with ctDNA implementation exist. We sought to assess disparities in frequency of actionable alterations detected and utilization of those results for treatment decisions in a diverse urban population of NSCLC patients. Methods: We performed a retrospective study of NSCLC patients who received Guardant ctDNA testing from 11/2015-10/2022. Kaplan Meier curves with log-rank tests were used to assess time-to-treatment initiation or treatment change from date of ctDNA collection among race/ethnicity subgroups (Hispanic, Black, White). Univariate associations between race/ethnicity subgroups, NCCN actionable mutation detection, and variant allele frequency, were assessed using Chi-Square and Kruskal-Wallis tests for categorical and continuous variables, respectively. Results: We identified 258 subjects. The race/ethnicity distribution in order of prevalence was Hispanic (33.0%, n = 85), Black (32.6%, n = 84), White (24.8%, n = 64), and Asian/other/unknown (9.7%, n = 25). Median turnaround time for ctDNA result was 8 days (range 2-24 days) from collection date. Median time-to-treatment initiation or change was 25 days for the combined cohort, and 24 days for patients with treatment naïve metastatic disease. Time-to-treatment initiation or change was similar among race/ethnicity subgroups in the combined cohort, as well as those specifically with metastatic disease not yet initiated on treatment at time of ctDNA testing (p = 0.52 and p = 0.43, respectively). Similar proportions of all race/ethnicity subgroups had an NCCN actionable mutation detected (p = 0.92) and similar variant allele frequencies (p = 0.68). An NCCN actionable mutation was detected in 35.3% (n = 91) of patients, the most common being EGFR (n = 53) and KRAS G12C (n = 17). CtDNA results helped avoid the need for repeat tissue biopsy in 25.6% (n = 66) of patients. Of the 68% (n = 175) of patients with paired tissue and liquid biopsies, 24.0% (n = 42) of patients had discordant results, and 8.6% (n = 15) had additional alterations detected on liquid biopsy. Conclusions: In our diverse cohort, we observed a similarly high detection rate of actionable mutations and early time-to-treatment based on ctDNA results across all race/ethnicity subgroups. In addition, ctDNA helped avoid need for repeat tissue biopsy in 25% of patients and yielded additional alterations not detected on tissue in nearly 10% of patients with discordant results. Our findings suggest that broad based molecular testing utilizing ctDNA is not only feasible, but also that it enables faster time-to-treatment in a racially diverse urban population. Continued efforts are necessary to ensure timely molecular testing among all patients, and to identify any racial inequities that may exist.