Atezolizumab Plus Chemotherapy for First-Line Treatment of Nonsquamous NSCLC: Results From the Randomized Phase 3 IMpower132 Trial.

Document Type


Publication Date


Publication Title

J Thorac Oncol


oregon; portland; chiles; Antibodies, Monoclonal, Humanized; Antineoplastic Combined Chemotherapy Protocols; Carboplatin; Carcinoma, Non-Small-Cell Lung; Cisplatin; Humans; Lung Neoplasms; Pemetrexed


INTRODUCTION: We report the final results of the phase 3 IMpower132 study evaluating atezolizumab plus carboplatin or cisplatin plus pemetrexed (APP) in patients with nonsquamous NSCLC.

METHODS: Chemotherapy-naive patients with stage IV nonsquamous NSCLC without sensitizing EGFR or ALK genetic alterations were randomized in a one-to-one ratio to receive four or six cycles of carboplatin or cisplatin plus pemetrexed (PP) or APP every 3 weeks, followed by maintenance therapy with atezolizumab plus pemetrexed or pemetrexed alone. Co-primary end points were overall survival (OS) and investigator-assessed progression-free survival (PFS).

RESULTS: The intention-to-treat population included 578 patients (APP, n = 292; PP, n = 286). At the primary PFS analysis (May 22, 2018; median follow-up, 14.8 mo), APP exhibited significant PFS improvement versus PP (median = 7.6 versus 5.2 mo, stratified hazard ratio [HR] = 0.60, 95% confidence interval [CI]: 0.49-0.72, p < 0.0001). OS for the APP group was numerically better but not statistically significant at the interim (May 22, 2018; median = 18.1 versus 13.6 mo, stratified HR = 0.81, 95% CI: 0.64-1.03, p = 0.0797) and final analyses (July 18, 2019; median = 17.5 versus 13.6 mo; stratified HR = 0.86, 95% CI: 0.71-1.06, p = 0.1546). The OS and PFS results favored APP versus PP across subgroups. Grade 3 or 4 treatment-related adverse events occurred in 54.6% (APP) and 40.1% (PP) of patients; grade 5 treatment-related events occurred in 3.8% and 2.9%, respectively.

CONCLUSIONS: IMpower132 met its co-primary PFS end point but not its co-primary OS end point, with numerical improvement for OS in the APP arm. APP had a manageable safety profile, with no new or unexpected safety signals identified.

Clinical Institute



Earle A. Chiles Research Institute