Background T-cell–inflamed gene expression profile (Tcell_infGEP) and tumor mutational burden (TMB) are clinically validated biomarkers that independently predict pembrolizumab response. This study investigated prospective Tcell_infGEP and TMB assessment in evaluating first-line pembrolizumab-based combination therapies; the different treatment combinations evaluated may provide insight into the unique biology of each biomarker subgroup.

Methods KEYNOTE-495/KeyImPaCT is a group-sequential, adaptively randomized, multisite, open-label, phase 2 study investigating first-line pembrolizumab plus the VEGF/FGFR inhibitor lenvatinib, CTLA-4 inhibitor quavonlimab (MK-1308), or LAG-3 inhibitor favezelimab (MK-4280) in patients with advanced NSCLC. DNA and RNA were extracted from tumor tissue to determine Tcell_infGEP and TMB; patients were assigned to one of four biomarker-defined subgroups (Tcell_infGEP^lowTMB^low, Tcell_infGEP^lowTMB^high, Tcell_infGEP^highTMB^low, Tcell_infGEP^highTMB^high) and randomly assigned 1:1:1 to receive pembrolizumab (200mg IV Q3W)+lenvatinib (20mg oral QD), pembrolizumab+quavonlimab (75mg IV Q6W), or pembrolizumab+favezelimab (200mg [n=30] or 800mg [n=34] Q3W; the initial prespecified dose was 200mg but changed to 800mg based on emerging data). The primary end point was investigator-assessed ORR per RECIST v1.1. Multiple interim analyses will be performed until the prespecified clinical signal is observed. The first interim analysis for each combination therapy occurred after ≥10 patients had ≥12 weeks of follow-up.

Results At the data cutoff (January 11, 2021), 208 patients were treated (pembrolizumab+lenvatinib, n=72; pembrolizumab+quavonlimab, n=72; pembrolizumab+favezelimab 200mg, n=30; pembrolizumab+favezelimab 800mg, n=34). The overall assay success rate for testing and determining Tcell_infGEP and TMB was 94%. In patients treated with pembrolizumab+lenvatinib, pembrolizumab+quavonlimab, or pembrolizumab+favezelimab, ORRs were generally highest in the Tcell_infGEP^highTMB^high subgroup (table 1); response rates were similar across combinations within this subgroup. ORR was low across combinations within the Tcell_infGEP^lowTMB^low subgroup. Treatment-related adverse events (TRAEs) occurred in 88%, 65%, 57%, and 59% of patients in the pembrolizumab+lenvatinib, pembrolizumab+quavonlimab, pembrolizumab+favezelimab 200mg and pembrolizumab+favezelimab 800mg arms, respectively. Consistent with the known TRAEs of these agents, most TRAEs were grade 1 or 2 in severity except in the pembrolizumab+lenvatinib arm (grade 3–5, 63%). Three deaths from TRAEs occurred (pembrolizumab+lenvatinib [n=2], brain hemorrhage and myocardial infarction; pembrolizumab+favezelimab 800 mg [n=1], pneumonitis).

Conclusions These data demonstrate the feasibility and clinical usefulness of prospective Tcell_infGEP and TMB assessment to study the clinical activity of three first-line pembrolizumab-based combination therapies in patients with advanced NSCLC. Although sample sizes were small, the Tcell_infGEP^highTMB^high subgroup demonstrated the best response among the biomarker subgroups for all three combination therapies; further validation is needed to determine additional signals and may be addressed as more mature data become available.

Acknowledgements Jeanne Fahey, PhD, of Merck & Co., Inc., Kenilworth, New Jersey, USA, provided critical review of the abstract. Elisha Dettman PhD, Mark Ayers MS, and Andrey Loboda PhD of Merck & Co., Inc., Kenilworth, New Jersey, USA, provided critical review of study translational data. Medical writing and/or editorial assistance was provided by Shane Walton, PhD, and Lei Bai, PhD, of ApotheCom (Yardley, PA, USA). This assistance was funded by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA.

Trial Registration ClinicalTrials.

gov, NCT03516981

Ethics Approval The study protocol and all amendments were approved by the relevant institutional review board or ethics committee at each study site. All patients provided written informed consent to participate in the clinical trial.