Biomarker associations of immune checkpoint inhibitor versus chemotherapy effectiveness in first-line metastatic endometrial carcinomas: A real-world study.

Abstract

5596 Background: Limited evidence exists comparing outcomes of single-agent anti-PD1 immune checkpoint inhibitors (ICPI) vs. chemotherapy in the 1st line among metastatic endometrial carcinoma (mEC) patients. dMMR, MSI status, and Tumor Mutational Burden (TMB) have been characterized as predictive biomarkers of ICPI response in many solid tumors, including mEC. We sought to evaluate ICPI vs. chemotherapy effectiveness in 1st line mEC, stratified by TMB ≥ 10 mut/MB and MSI as assessed by next-generation sequencing. Methods: Following a prespecified analysis plan, this study used the nationwide (̃280 US cancer clinics) de-identified, EHR-derived, Flatiron Health-Foundation Medicine mEC clinico-genomic database (FH-FMI CGDB), with tissue CGP between 1/2011 - 6/2021. Cohort inclusion criteria: recurrent mEC patients with stage I-III at original diagnosis, 1st line ICPI or platinum chemotherapy treatment with TMB and MSI assessments available. Adjusted hazard ratios (aHR) from multivariable Cox proportional hazard models were utilized for time to next treatment (TTNT) and overall survival (OS) comparisons between 1L chemotherapy and 1L ICPI from start of 1L treatment stratified by TMB (≥ 10 vs. < 10) and adjusted for ECOG, BMI, and stage at diagnosis. Results: Among the 1st line cohort, patients received either Chemotherapy (n = 139, 87%) or first-line ICPI (n = 20, 13%). While the overlap between the TMB ≥ 10 (n = 46) population and MSI-H (n = 39) populations was high, with 38/46 (83%) of TMB ≥ 10 classified as MSI-H and 38/39 (97%) of MSI-H with a TMB ≥ 10, there were also 8 (17%) TMB ≥ 10 patients who were MSS. TMB ≥ 10 was associated with more favorable TTNT (aHR: 0.11 [0.03 - 0.44]) and OS (aHR: 0.1 [0.01 - 0.82]) on single-agent ICPIs vs. chemotherapy; however, TMB < 10 patients exhibited no observed differences in TTNT (aHR: 1.66 [0.79 - 3.48]) or OS (aHR: 1.79 [0.66 - 4.8]) between treatments. MSI-H pts were associated with more favorable TTNT (aHR: 0.1 [0.02 - 0.39]) and OS (aHR: 0.1 [0.01 - 0.86]) on single-agent ICPIs over chemotherapy; however, MSS pts did not observe differences in TTNT (aHR: 1.5 [0.65 - 3.5]) or OS (aHR: 2.14 [0.63 - 7.26]) between ICPI vs. chemotherapy. Conclusions: Recurrent 1st line mEC patients with TMB ≥ 10 and/or MSI-H exhibit more favorable outcomes on single-agent ICPI than standard chemotherapy in real-world settings. These findings warrant prospective randomized validation. Future studies should assess the utility of a combined biomarker approach across TMB and MSI status to identify the broadest group of patients who might benefit from ICPI.