In patients with advanced NSCLC and PD-L1 expression on at least 50% of tumor cells, pembrolizumab was associated with significantly longer progression-free and overall survival and with fewer adverse events than was platinum-based chemotherapy. (Funded by Merck; KEYNOTE-024 ClinicalTrials.gov number, NCT02142738 .).

BACKGROUNDStandard first-line therapy for metastatic, squamous non-small-cell lung cancer (NSCLC) is platinum-based chemotherapy or pembrolizumab (for patients with programmed death ligand 1 [PD-L1] expression on ≥50% of tumor cells). More recently, pembrolizumab plus chemotherapy was shown to significantly prolong overall survival among patients with nonsquamous NSCLC. METHODSIn this double-blind, phase 3 trial, we randomly assigned, in a 1:1 ratio, 559 patients with untreated metastatic, squamous NSCLC to receive 200 mg of pembrolizumab or saline placebo for up to 35 cycles; all the patients also received carboplatin and either paclitaxel or nanoparticle albumin-bound [nab]-paclitaxel for the first 4 cycles. Primary end points were overall survival and progression-free survival. RESULTSAfter a median follow-up of 7.8 months, the median overall survival was 15.9 months (95% confidence interval [CI], 13.2 to not reached) in the pembrolizumab-combination group and 11.3 months (95% CI, 9.5 to 14.8) in the placebo-combination group (hazard ratio for death, 0.64; 95% CI, 0.49 to 0.85; P<0.001). The overall survival benefit was consistent regardless of the level of PD-L1 expression. The median progression-free survival was 6.4 months (95% CI, 6.2 to 8.3) in the pembrolizumabcombination group and 4.8 months (95% CI, 4.3 to 5.7) in the placebo-combination group (hazard ratio for disease progression or death, 0.56; 95% CI, 0.45 to 0.70; P<0.001). Adverse events of grade 3 or higher occurred in 69.8% of the patients in the pembrolizumab-combination group and in 68.2% of the patients in the placebocombination group. Discontinuation of treatment because of adverse events was more frequent in the pembrolizumab-combination group than in the placebo-combination group (13.3% vs. 6.4%). CONCLUSIONSIn patients with previously untreated metastatic, squamous NSCLC, the addition of pembrolizumab to chemotherapy with carboplatin plus paclitaxel or nab-paclitaxel resulted in significantly longer overall survival and progression-free survival than chemotherapy alone. (Funded by Merck Sharp & Dohme; KEYNOTE-407 ClinicalTrials.gov number, NCT02775435.

The implementation of clinical-decision support algorithms for medical imaging faces challenges with reliability and interpretability. Here, we establish a diagnostic tool based on a deep-learning framework for the screening of patients with common treatable blinding retinal diseases. Our framework utilizes transfer learning, which trains a neural network with a fraction of the data of conventional approaches. Applying this approach to a dataset of optical coherence tomography images, we demonstrate performance comparable to that of human experts in classifying age-related macular degeneration and diabetic macular edema. We also provide a more transparent and interpretable diagnosis by highlighting the regions recognized by the neural network. We further demonstrate the general applicability of our AI system for diagnosis of pediatric pneumonia using chest X-ray images. This tool may ultimately aid in expediting the diagnosis and referral of these treatable conditions, thereby facilitating earlier treatment, resulting in improved clinical outcomes. VIDEO ABSTRACT.

Purpose In the randomized, open-label, phase III KEYNOTE-024 study, pembrolizumab significantly improved progression-free survival and overall survival (OS) compared with platinum-based chemotherapy in patients with previously untreated advanced non–small-cell lung cancer (NSCLC) with a programmed death ligand 1 tumor proportion score of 50% or greater and without EGFR/ALK aberrations. We report an updated OS and tolerability analysis, including analyses adjusting for potential bias introduced by crossover from chemotherapy to pembrolizumab. Patients and Methods Patients were randomly assigned to pembrolizumab 200 mg every 3 weeks (for up to 2 years) or investigator’s choice of platinum-based chemotherapy (four to six cycles). Patients assigned to chemotherapy could cross over to pembrolizumab upon meeting eligibility criteria. The primary end point was progression-free survival; OS was an important key secondary end point. Crossover adjustment analysis was done using the following three methods: simplified two-stage method, rank-preserving structural failure time, and inverse probability of censoring weighting. Results Three hundred five patients were randomly assigned (pembrolizumab, n = 154; chemotherapy, n = 151). At data cutoff (July 10, 2017; median follow-up, 25.2 months), 73 patients in the pembrolizumab arm and 96 in the chemotherapy arm had died. Median OS was 30.0 months (95% CI, 18.3 months to not reached) with pembrolizumab and 14.2 months (95% CI, 9.8 to 19.0 months) with chemotherapy (hazard ratio, 0.63; 95% CI, 0.47 to 0.86). Eighty-two patients assigned to chemotherapy crossed over on study to receive pembrolizumab. When adjusted for crossover using the two-stage method, the hazard ratio for OS for pembrolizumab versus chemotherapy was 0.49 (95% CI, 0.34 to 0.69); results using rank-preserving structural failure time and inverse probability of censoring weighting were similar. Treatment-related grade 3 to 5 adverse events were less frequent with pembrolizumab compared with chemotherapy (31.2% v 53.3%, respectively). Conclusion With prolonged follow-up, first-line pembrolizumab monotherapy continues to demonstrate an OS benefit over chemotherapy in patients with previously untreated, advanced NSCLC without EGFR/ALK aberrations, despite crossover from the control arm to pembrolizumab as subsequent therapy.