ABSTRACT Aim: E has proven efficacy as first-line therapy for pts with EGFR mut+ NSCLC. The phase 2, open-label, single-arm ASPIRATION study assessed the efficacy of first-line E until RECIST PD, efficacy beyond PD if E was continued by the investigator, and safety in pts with EGFR mut+ NSCLC in Asia. Methods: Pts ≥18 yrs with stage IV, EGFR mut+ NSCLC received E 150mg/day orally. Primary endpoint: PFS1 (time to RECIST PD/death). Secondary endpoints: PFS2 (time to off-E PD if E was extended beyond RECIST PD) in all pts and in pts with exon 19 deletion/L858R mutations, objective response rate (ORR), disease control rate (DCR), best objective response (BOR), PFS1 in the exon 19 deletion/L858R subset, OS and safety. Results: ITT population: 207 pts; 150 had a RECIST PD event at data cut-off, 46 had no PFS1, 11 withdrew, 81 continued post-PD E. Median follow-up: 213 days. PFS1: median 10.8 mo (95% CI 9.2–11.1). For pts with a RECIST PD event, data from pts receiving post-PD E vs pts not receiving post-PD E are shown in the Table. Median PFS2 (n = 81; 67 PD events): 13.0 mo (95% CI 11.5–14.8). In pts receiving post-PD E the difference between PFS1 and PFS2 was 3.7 mo. In pts with centrally confirmed exon 19 deletion/L858R mutations, median PFS1 (n = 144): 11.0 mo; median PFS2 (n = 54): 13.1 mo. ORR: 65.2%, DCR: 82.6%. OS data are immature. In the safety population (n = 207), 94 pts (45.4%) had grade ≥3 AEs, 7 (3.4%) had grade 5 AEs and 49 (23.7%) had a serious AE. The most common treatment-related AEs (grade ≥3) were skin disorders (15.9%) and GI disorders (2.9%); these were more common (all grades) in post-PD E pts vs pts not receiving post-PD E. More non-post-PD E pts (11%) had new lung lesions at PFS1 than post-PD E pts (5%). Post-PD E N = 81 No post-PD E N = 69 PFS1*, mos 9.3 (95% CI 9.1–11.0) 7.2 (95% CI 5.4–9.2) Depth of response*, % -49.6† -38.9‡ Baseline to BOR*, days 55 59 BOR to PFS1*, days 140 105 ECOG 0/1§, % 95.1 78.3§§ Grade 4 AEs§, % 2.5 7.2 *Median, †n = 79, ‡n = 61, §at PFS1, §§n = 62. Conclusions: The ASPIRATION data show that continuing E beyond RECIST PD is feasible, with a difference between PFS1 and PFS2 of 3.7 mo in post-PD E pts. However, validation of the optimal pt subset to benefit from post-PD E needs further research. Disclosure: K. Park: I have served on advisory boards for A-Z, BI, Clovis, EliLilly, KHK, Novartis and Roche. I have also participated in corporate sponsored research with A-Z; M. Lin: I have served on advisory boards for A-Z, Roche and Boehringer Ingelheim; J. Kang: I have served on advisory boards for Amgen and Pfizer; P. Perez-Moreno: I am an employee and stock owner of F. Hoffmann-La Roche Ltd.; P. Button: I am an employee of Roche Products Pty Ltd.; D. Gregory: I am an employee of Roche Products Pty Ltd.; T.S.K. Mok: Advisory boards: A-Z, Roche, Eli Lilly, Merck Serono, Eisai, BMS, AVEO, Pfizer, Taiho, BI, Novartis, GSK Biologicals, Clovis Oncology, Amgen Inc, Janssen and BioMarin Pharmaceutical Inc. Board of Directors: IASLC Corporate sponsored research: A-Z. All other authors have declared no conflicts of interest.