KEYNOTE-189: Randomized, double-blind, phase 3 study of pembrolizumab (pembro) or placebo plus pemetrexed (pem) and platinum as first-line therapy for metastatic NSCLC

Abstract

Background: The premise that chemotherapy and immunotherapy may provide greater benefit than pembro or chemotherapy alone was supported by KEYNOTE-021 cohort G, in which pembro + pem + carboplatin significantly improved ORR and PFS over pem + carboplatin in patients (pts) with advanced nonsquamous NSCLC. We present results of the KEYNOTE-189 study of pembro + pem-platinum vs placebo + pem-platinum as first-line therapy for metastatic nonsquamous NSCLC. Methods: Pts with previously untreated stage IV nonsquamous NSCLC, no EGFR or ALK alteration, and ECOG PS 0-1 were randomized 2:1 to 4 Q3W cycles of pembro 200 mg or placebo + pem 500 mg/m2 + carboplatin AUC 5 or cisplatin 75 mg/m2 followed by maintenance pembro or placebo + pem. Randomization was stratified by PD-L1 tumor proportion score (TPS; <1% vs ≥1%), platinum agent, and smoking status. Response was assessed by RECIST v1.1 per blinded, independent central review. Pts in the placebo arm with verified PD could crossover to pembro monotherapy if eligible. Primary end points were OS and PFS in the ITT population. Prespecified superiority boundaries at the first interim analysis were one-sided P = .00128 for OS and .00559 for PFS. Results: 616 pts were randomized: 410 to pembro + pem + platinum (“pembro arm”), 206 to placebo + pem + platinum (“placebo arm”). 76.5% of treated pts in the pembro arm and 66.8% in the placebo arm received ≥5 cycles of pem. 88.1% of enrolled pts were current/former smokers, carboplatin was chosen for 72.2%, and 63.0% had TPS ≥1%. As of Nov 8, 2017, median follow-up was 10.5 mo (range 0.2-20.4), and 33.8% in the pembro arm vs 17.8% in the placebo arm remained on treatment. In the placebo arm, 67 pts crossed over in-study to pembro and 18 received anti-PD-(L)1 therapy outside the study (effective crossover rate: 41.3% in the ITT population, 50.0% when pts still on treatment excluded). Pembro + pem + platinum, improved OS (HR 0.49; 95% CI 0.38-0.64; P < .00001) and PFS (HR 0.52; 95% CI 0.43-0.64; P < .00001) over placebo + pem + platinum. Median OS was not reached with pembro and was 11.3 mo with placebo. Median PFS was 8.8 mo vs 4.9 mo. OS benefit in the pembro arm was observed across subgroups, including all PD-L1 TPS categories (HR [95% CI] 0.59 [0.38-0.92] for <1%, 0.55 [0.34-0.90] for 1-49%, and 0.42 [0.26-0.68] for ≥50%). ORR was higher with pembro (47.6% vs 18.9%; P < .00001). Grade ≥3 AEs occurred in 67.2% of pts in the pembro arm vs 65.8% in the placebo arm; AEs led to discontinuation of any treatment in 27.7% vs 14.9%, all treatment in 13.8% vs 7.9%, and death in 6.7% vs 5.9%. Conclusions: Pembro + pem + platinum provided superior OS, PFS, and ORR compared with placebo + pem + platinum and had a manageable safety profile in pts with previously untreated metastatic nonsquamous NSCLC without EGFR or ALK alterations. These data suggest first-line pembro + pem and platinum may be a new standard of care for this population