De-risking 1st line chemotherapy-free immune checkpoint inhibitor (ICI) use for patients with advanced gastroesophageal cancer (aGEJ) with tumor mutational burden (TMB).

Abstract

407 Background: ICI + chemotherapy (chemo-ICI) is becoming an increasingly common 1st line regimen for patients with aGEJ. However, direct outcome comparison between biomarker-enriched groups receiving chemo-ICI vs. ICI monotherapy (mono-ICI) is limited in phase III trials. We sought to evaluate if TMB might identify patients for whom mono-ICI vs. chemo-ICI might be de-risked. We prospectively hypothesized that (1) patients with TMB <10 mut/mB would have more favorable outcomes receiving ICI-chemo vs. mono-ICI and (2) patients with TMB ≥10 mut/mB would have diminished additional benefit from chemo-ICI vs. mono-ICI. Methods: This study included patients with aGEJ who received 1st line chemo-ICI or mono-ICI and tumor tissue genomic profiling by Foundation Medicine. Patient data was obtained by the nationwide (US-based) de-identified Flatiron Health-Foundation Medicine gastric real-world clinico-genomic database (FH-FMI CGDB), originating from approximately 280 US cancer clinics (January/2011–March/2023). Real-world progression-free survival (rwPFS) and overall survival (rwOS) were compared between patients receiving chemo-ICI vs. mono-ICI by Cox models, adjusted by propensity scores accounting for disease stage at diagnosis, ECOG PS, PD-L1 score and microsatellite instability (MSI) status. Comparisons of the explanatory power of TMB (<10 vs. ≥10 mut/mB) vs. MSI status (MSI-H vs MSS) were performed by likelihood ratio test (LRT). Results: A total of 263 patients (176 receiving chemo-ICI and 87 mono-ICI) were included in the study. Patients with TMB <10 mut/mB had more favorable outcomes receiving chemo-ICI vs. mono-ICI for rwPFS (median 5.8 vs. 1.9 months, hazard ratio [HR]: 0.44, 95% confidence interval [CI] 0.26-0.76), p=0.0029) and rwOS (median 11.1 vs. 7.9 months, HR: 0.68, 95% CI 0.40-1.15, p=0.1489). Patients with TMB ≥10 mut/mB did not have reduced benefit from mono-ICI; curiously less favorable outcomes were observed with chemo-ICI vs. mono-ICI for rwPFS (HR: 3.27, 95% CI 1.19-8.97, p=0.0212) and rwOS (HR: 3.56, 95% CI 1.14-11.08, p=0.0286). Evaluating only the mono-ICI group, the LRT indicated that when TMB was added to a model evaluating only MSI status, there was an improvement in ICI outcome prediction (p=0.006 for rwPFS and p=0.003 for rwOS), but not when MSI was added to TMB (p=0.91 for rwPFS and p=0.40 for rwOS). Conclusions: aGEJ patients with TMB <10 mut/mB had more favorable outcomes receiving chemo-ICI vs. mono-ICI. In the TMB ≥10 mut/mB we did not observe more favorable outcomes among patients receiving chemo-ICI vs. mono-ICI. Curiously, we observed more favorable outcomes among those receiving mono-ICI in the TMB ≥10 mut/mB group. Among patients receiving mono-ICI, TMB better predicted mono-ICI outcomes compared to MSI status alone.