Impact of atezolizumab (atezo) treatment beyond disease progression (TBP) in advanced NSCLC: Results from the randomized phase III OAK study.

Abstract

9001 Background: Cancer immunotherapy (CIT) can have a positive impact on OS that exceeds response rate or PFS effects, termed post progression prolongation of survival (PPPS). This effect can also result from unconventional CIT response due to tumor immune infiltration or delayed response, reducing reliability of RECIST v1.1 (RECIST) PD as an indicator of treatment failure. In the primary analysis (N = 850) of OAK, a study of atezo vs docetaxel (doc) in 2L/3L NSCLC, OS favored atezo (HR 0.73; 95% CI: 0.62, 0.87), despite similar PFS between arms (HR 0.95; 95% CI: 0.82, 1.10). Here we evaluate clinical benefit from TBP, defined by post PD tumor regression, OS and safety. Methods: Patients (pts) received atezo 1200 mg IV q3w until PD or loss of clinical benefit per investigator or doc 75 mg/m 2 IV q3w until PD per RECIST. No crossover was allowed. Primary outcome measure: OS. Atezo TBP pts were evaluated for post PD tumor change and for safety pre and post PD. OS from time of PD per RECIST was evaluated in both arms (data cutoff, July 7, 2016; minimum follow-up, 19 mo). Results: Among 332 atezo pts with PD, 51% (n = 168) continued atezo TBP; 7% (12/168) achieved subsequent response in target lesions (≥ 30% reduction from new baseline at PD); 49% (83/168) had stable target lesions (best change between +20% and −30%). mOS was 12.7 mo (95% CI: 9.3, 14.9) post PD for pts on atezo TBP (Table). Atezo TBP was not associated with increased safety risk. Conclusions: This is the first report from a Phase III study of CIT in NSCLC to evaluate post PD OS in pts continuing treatment beyond RECIST PD. Atezo TBP was associated with high frequency of stable or decreased target lesions, mOS > 1 year and a tolerable safety profile, all supporting prolonged treatment benefit consistent with PPPS. NCT02008227 Clinical trial information: NCT02008227. [Table: see text]