Tumor mutational burden as a predictive biomarker for immune checkpoint inhibitor versus chemotherapy benefit in first-line metastatic urothelial carcinoma: A real-world study.

Abstract

547 Background: There is an unmet need to identify metastatic urothelial carcinoma (mUC) patients who might be spared chemotherapy in 1st line. Anti-PD-(L)1 immune checkpoint inhibitors (ICPI) alone without chemotherapy did not show superiority to platinum-based chemotherapy in ITT populations of DANUBE, KEYNOTE-361, and IMvigor130. However, DANUBE and IMvigor130 reported secondary subgroup analyses, both suggesting enhanced benefit for ICPI vs. chemotherapy in patients with tumor mutational burden (TMB) ≥ 10 (mutations/megabase), using same cutoff and assay as pan-tumor CDx for pembrolizumab approved in later lines of therapy. We sought to determine if TMB ≥ 10 identified a group of enhanced relative ICPI benefit (single-agent anti-PD[L]1 w/o chemo) in real-world settings where patients are less eligible for chemotherapy. Methods: Association of genomic data with clinical variables and outcomes in cohort of patients with mUC treated January 2011- April 2021. Longitudinal de-identified clinical data from approximately 280 U.S. academic or community-based cancer clinics were derived from electronic health records, curated via technology-enabled abstraction by Flatiron Health and linked to genomic testing by Foundation Medicine. 849 1st line mUC patients received either ICPI (n = 307) or chemotherapy (n = 542) at physician’s discretion in standard of care settings. All patients underwent genomic testing using Foundation Medicine comprehensive genomic profiling assays (FoundationOne© or FoundationOne©CDx). PFS and OS were assessed unadjusted and adjusted for imbalances using propensity scores. Results: 273 of 849 (32.2%) patients had TMB ≥ 10. Pre-therapy characteristics: patients assigned ICPI vs. chemotherapy had comparable TMB, primary disease site, histology, smoking status, and PD-L1 staining, but were generally older (median years: 72 vs. 67, p < 0.001), higher ECOG scores (p < 0.001), lower CrCl (median ml/min: 49.8 vs. 59.7, p < 0.001), and lower hemoglobin (median: 11.5 vs. 12.1, p < 0.001). Unadjusted, TMB ≥ 10 group showed more favorable PFS (HR: 0.72, 95%CI: 0.52 – 0.99, p = 0.041) and OS (HR: 0.70, 95%CI: 0.49 – 0.1, p = 0.048) for ICPI vs. chemotherapy despite imbalances favoring outcomes on chemotherapy. ICPI vs. chemotherapy outcomes adjusted for imbalances: TMB ≥ 10 group showed more favorable PFS (HR: 0.65, 95%CI: 0.45 – 0.95, p = 0.026) and OS (HR: 0.61, 95%CI: 0.39 – 0.93, p = 0.022), while TMB < 10 had comparable or worse PFS (HR: 1.30, 95%CI: 0.98 – 1.72, p = 0.06) and OS (HR: 1.03; 95%CI: 0.78– 1.34, p = 0.85). Conclusions: In real-world settings, 1st line mUC patients with TMB ≥ 10 have more favorable PFS and OS on single agent ICPI than chemotherapy, adding clinical validity to TMB as a predictive biomarker in patient populations less eligible for chemotherapy than reported trials.