Nivolumab has been associated with longer overall survival than docetaxel among patients with previously treated non-small-cell lung cancer (NSCLC). In an open-label phase 3 trial, we compared first-line nivolumab with chemotherapy in patients with programmed death ligand 1 (PD-L1)-positive NSCLC. We randomly assigned, in a 1:1 ratio, patients with untreated stage IV or recurrent NSCLC and a PD-L1 tumor-expression level of 1% or more to receive nivolumab (administered intravenously at a dose of 3 mg per kilogram of body weight once every 2 weeks) or platinum-based chemotherapy (administered once every 3 weeks for up to six cycles). Patients receiving chemotherapy could cross over to receive nivolumab at the time of disease progression. The primary end point was progression-free survival, as assessed by means of blinded independent central review, among patients with a PD-L1 expression level of 5% or more. Among the 423 patients with a PD-L1 expression level of 5% or more, the median progression-free survival was 4.2 months with nivolumab versus 5.9 months with chemotherapy (hazard ratio for disease progression or death, 1.15; 95% confidence interval CI, 0.91 to 1.45; P=0.25), and the median overall survival was 14.4 months versus 13.2 months (hazard ratio for death, 1.02; 95% CI, 0.80 to 1.30). A total of 128 of 212 patients (60%) in the chemotherapy group received nivolumab as subsequent therapy. Treatment-related adverse events of any grade occurred in 71% of the patients who received nivolumab and in 92% of those who received chemotherapy. Treatment-related adverse events of grade 3 or 4 occurred in 18% of the patients who received nivolumab and in 51% of those who received chemotherapy. Nivolumab was not associated with significantly longer progression-free survival than chemotherapy among patients with previously untreated stage IV or recurrent NSCLC with a PD-L1 expression level of 5% or more. Overall survival was similar between groups. Nivolumab had a favorable safety profile, as compared with chemotherapy, with no new or unexpected safety signals. (Funded by Bristol-Myers Squibb and others; CheckMate 026 ClinicalTrials.gov number, NCT02041533 .).