OBJECTIVES: TMB represents a measurement of the number of mutations within the genome, but the prognostic effect of TMB score on OS in patients not receiving immunotherapy is not well known. This retrospective observational study aimed to describe the demographic and clinical characteristics, treatment patterns, and overall survival (OS) by TMB score for patients who are IO naïve and received non-IO in the 2L setting for advanced/metastatic solid tumors.

METHODS: This study used a nationwide (US-based) de-identified Flatiron Health-Foundation Medicine multi-tumor clinico-genomic database (CGDB). The data originated from approximately 280 US cancer clinics (~800 sites of care). The database included patients age ≥18 diagnosed between January 2011 - November 2020 with the following advanced/metastatic solid tumors: Gastric, Breast, Colorectal, Small Cell Lung Cancer, Head and Neck, Ovarian, Prostate, Pancreatic, Hepatocellular Carcinoma, Bladder, Endometrial cancers, and a known tissue TMB score. TMB-High was defined as ≥ 10 mutations / megabase using FoundationOne CDx assay. OS from start of 2L therapy was analyzed using Kaplan-Meier methodology and Cox proportional hazard models, adjusting for TMB status, age at initiation of 2L, practice type, gender, race, insurance, initial staging and ECOG Performance Score (PS) at 2L initiation.

RESULTS: 7465 patients were included: 427 TMB-High and 7038 TMB-Low, median age 63 (at index treatment), received non-IO in 2L. Majority (63%) of patients were female, and 82% of patients had ECOG PS of 0 or 1. Median OS was 19.5 [95% CI: 15.8, 23.9] months for TMB-High and 18.6 months [95% CI: 18.0, 19.4] for TMB-Low. In the adjusted Cox model, TMB-High status was not associated with a difference in the risk of death from 2L therapy initiation (HR: 0.931, 95% CI: [0.821, 1.056]).

CONCLUSIONS: Adjusted Cox regression model suggests that OS from 2L is not significantly different between TMB high and TMB low patients.