Pembrolizumab in patients with advanced non-small-cell lung cancer (KEYNOTE-001): 3-year results from an open-label, phase 1 study

Abstract

Background: The anti-programmed death 1 monoclonal antibody pembrolizumab has shown antitumour activity and is a first-line and second-line treatment option for patients with programmed death ligand 1 (PD-L1)-expressing advanced non-small-cell lung cancer. We report updated 3-year safety and efficacy outcomes from the phase 1 study, KEYNOTE-001. Methods: KEYNOTE-001 is a multicohort, open-label, phase 1 study of pembrolizumab (2 mg/kg every 3 weeks or 10 mg/kg every 2 or 3 weeks) in treatment naive or previously treated patients with locally advanced or metastatic non-small-cell lung cancer with measurable disease at baseline. Two cohorts were randomly assigned to a pembrolizumab dose by use of a computer-generated randomisation schedule at cohort-dependent ratios, and a further four cohorts were assigned to a pembrolizumab dose without randomisation. We present 3-year outcomes for the full analysis set of patients who received at least one dose of study treatment, pooled for all pembrolizumab doses. The primary efficacy endpoint was proportion of patients with objective response, analysed here as investigator-assessed response according to immune-related response criteria. Secondary efficacy endpoints included overall survival, duration of response, and progression-free survival. Safety endpoints included incidence of adverse events. This study is registered at ClinicalTrials.gov, number NCT01295827, and is ongoing. Findings: Between May 8, 2012 and July 13, 2014, 550 patients (101 treatment naive and 449 previously treated) were enrolled. Median follow-up was 34·5 months at data cutoff (Sept 1, 2016). At 36 months, investigator-assessed objective response according to immune-related response criteria was achieved for 41 of 101 treatment naive patients (41% [95% CI 30·9–50·8]; median duration of response was 16·7 months [95% CI 12·6–not reached]) and 102 of 449 previously treated patients (23% [18·9–26·9]; 33·3 ([22·5–not reached]). The Kaplan-Meier estimate of overall survival at 36 months was 26·4% (95% CI 14·3–40·1) for treatment naive patients and 19·0% (15·0–23·4) for previously treated patients, with median overall survival of 22·3 months (95% CI 17·1–31·5) and 10·5 months (8·6–13·2). PD-L1 tumour proportion score ≥50% was associated with longer median overall survival (95% CI) versus tumour proportion score 1–49% (treatment naive: 34·9 [20·3–not reached] vs 19·5 [10·7–26·3] months; previously treated: 15·4 [10·5–18·5] vs 8·5 [6·0–12·7] months). Grade 3–5 treatment-related adverse events occurred in 66 patients (12%), and 30 (6%) discontinued owing to a treatment-related adverse event. The most frequent grade 3–4 treatment-related adverse events were pneumonitis (10 [2%] of 550) and fatigue (5 [1%] of 550). Overall, 227 patients (41%) of 550 had serious adverse events, of which 50 (9%) were treatment related. Interpretation: Pembrolizumab provides durable response and long-term effects on overall survival, with tolerable safety, for treatment naive and previously treated patients with advanced non-small-cell lung cancer expressing PD-L1. Funding: Merck Sharp & Dohme Corp.