Abstract
INTRODUCTION: The efficacy and safety of atezolizumab in previously treated patients with NSCLC have been established in the registrational phase 3 OAK trial. In this study, we evaluated the effectiveness and safety of atezolizumab monotherapy in a large real-world cohort to confirm the reproducibility of the results of the registrational trial. METHODS: This was a multicenter, prospective, single-arm observational study. Consecutive patients with previously treated NSCLC scheduled to receive atezolizumab monotherapy were enrolled. The primary end point was the 18-month overall survival (OS) rate. The incidence of adverse events (AEs) and immune-related AEs was evaluated. RESULTS: Overall, 1002 patients were included in the safety analysis set and 1000 in the full analysis set. Median follow-up was 11.5 months. Of the full analysis set, 62% were ineligible for the OAK trial (OAK-unlike subpopulation). The 18-month OS rate was 41.1%, with a median OS of 13.0 months (95% confidence interval: 12.2-15.1). The 18-month OS rate was 49.4% and 36.1% in OAK-like and OAK-unlike subpopulations, respectively; that in patients with Eastern Cooperative Oncology Group performance status greater than or equal to 2 was 14.3%. The incidence of AEs overall, in the OAK-like, and OAK-unlike subpopulations was 43.9%, 46.2%, and 42.5%; that of immune-related AEs was 19.0%, 20.1%, and 18.3%, respectively. CONCLUSIONS: The findings suggest that atezolizumab may be effective and safe for previously treated patients with NSCLC in real-world settings; however, atezolizumab administration should be considered carefully regarding the benefit-risk balance for the OAK-unlike subpopulation, especially in patients with Eastern Cooperative Oncology Group performance status greater than or equal to 2.