Background: Capmatinib is a selective MET inhibitor approved for patients (pts) with metastatic NSCLC harboring MET exon 14 skipping mutation. Pembrolizumab (pembro) is a programmed death protein-1 (PD-1) inhibitor approved as monotherapy for pts with advanced NSCLC expressing PD-ligand 1 (PD-L1). Preclinical studies have shown that capmatinib enhances T cell mediated antitumor response in mice treated with PD-1 inhibitors. Combining capmatinib with pembro may be beneficial in pts with advanced NSCLC expressing high tumor PD-L1. Methods: Herein we evaluated the efficacy and safety of capmatinib plus pembro (combo) versus pembro alone in tx-naïve pts with advanced MET-unselected NSCLC expressing PD-L1 with tumor proportion score (TPS) ≥50%, and no ALK or EGFR tumor aberrations. Pts received pembro 200 mg IV q3w in the pembro alone arm or with capmatinib 400 mg orally BID in the combo arm. The primary endpoint was investigator-assessed progression-free survival (PFS) using RECIST v1.1. Secondary endpoints included overall response rate (ORR), disease control rate (DCR), pharmacokinetics and safety. Results: As of the data cut off (DCO) of 28 Feb 2021, 76 pts were randomized 2:1 to the combo arm (n = 51) or pembro alone arm (n = 25). Baseline demographics and disease status were mostly comparable across study arms. At this interim analysis, PFS data were not mature. Median PFS (95% CI) was 6.3 (3.2, not evaluable [NE]) months in the combo arm and 4.3 (2.3, NE) months in the pembro alone arm. The ORR (95% CI) was 15.7% (7.0%, 28.6%) and 28.0% (12.1%, 49.4%) in the combo and pembro alone arms, respectively. The DCR (95% CI) was comparable across study arms; combo: 56.9% (42.2%, 70.7%) and pembro alone: 56.0% (34.9%, 75.6%). In the combo arm, capmatinib exposure (C_max: 3580 ng/mL [n = 7] and AUC_tau: 19700 hr*ng/mL [n = 2]) was consistent with data from previous studies. Tx-related grade (GR) ≥3 adverse events (AEs) were more common in the combo (37.3%) vs pembro alone arm (16%). Tx-related AEs occurring in ≥10% of pts are shown in the Table. Tx discontinuation and dose adjustment/interruptions were more common in the combo (27.5% and 52.9%) vs pembro alone arm (16% and 16%). Capmatinib was stopped prematurely in the combo arm, and at DCO, 32 (62.7%) pts in the combo arm and 18 (72%) pts in the pembro alone arm were receiving pembro monotherapy. Conclusions: Combination tx with capmatinib and pembro was not well tolerated and did not improve antitumor activity in tx-naïve pts with advanced NSCLC with PD-L1 TPS ≥50%. Clinical trial information: NCT04139317.

All data are n (%).