Randomized phase III trial of concurrent chemoradiation followed by nivolumab or placebo for locally advanced non-small cell lung cancer (NSCLC) (RTOG 3505).

Document Type

Conference Proceeding

Publication Date


Publication Title

J Clin Oncol


Background: Despite aggressive therapy with concurrent chemoradiation, fewer than 25% of patients with stage 3 NSCLC achieve 5-year survival and are presumably cured. To date, treatment modifications-including consolidation chemotherapy, maintenance therapy with molecularly targeted agents, concomitant administration of monoclonal antibodies, and escalation of radiation therapy (RT) dose-have not improved these outcomes. Immune checkpoint inhibitors represent an effective treatment for advanced NSCLC and may enhance RT-associated anti-tumor immunity. RTOG 3505 will test whether the addition of the anti-programmed death 1 (PD1) antibody nivolumab after chemoradiation improves overall survival (OS) and progression-free survival (PFS) in this population. Methods: Key eligibility criteria include surgically unresectable stage 3 NSCLC, ECOG 0-1, adequate organ function, available archival tissue, and absence of active autoimmune disease. Patients will receive thoracic RT to 60 Gy with concurrent cisplatin 50 mg/m2IV on Days 1, 8, 29, and 36, and etoposide 50 mg/m2IV on Days 1-5 and 29-33. This regimen was selected to (1) minimize risk of pulmonary toxicity and steroid requirements, and (2) optimize timing of immunotherapy. Between 4 and 12 weeks after completion of chemoradiation, eligible patients will be randomized to nivolumab 240 mg IV or placebo every 2 weeks for 1 year. Stratification factors include performance status, histology, and tumor PD-L1 status. Co-primary endpoints are OS and PFS, as determined by central radiology review. Secondary objectives include toxicity assessment, patient-reported outcomes and quality of life, and OS and PFS according to PD-L1 expression. Exploratory objectives include biomarkers to predict treatment efficacy and toxicity. A total of 660 patients will be enrolled to provide ≥90% power to detect (1) a hazard ratio (HR) of 0.7 for OS with two-sided type I error of 0.04, and (2) HR of 0.667 for PFS twosided type I error of 0.01, allowing a 16.7% drop-out rate before randomization.





This document is currently not available here.