Start Your Search
Poster Display session (Friday) (ID 65)
- Event: ELCC 2018
- Type: Poster Display session
- Presentations: 1
- Coordinates: 4/13/2018, 12:30 - 13:00, Hall 1
188P - Treatment switching–adjusted overall survival (OS) in KEYNOTE-024: First-line pembrolizumab versus chemotherapy in patients with advanced non-small cell lung cancer (NSCLC) (ID 622)
12:30 - 13:00 | Author(s): A.G. Robinson
In KEYNOTE-024, pembrolizumab was superior to chemotherapy for advanced untreated NSCLC with PD-L1 tumor proportion score (TPS) ≥50% and no sensitizing EGFR mutations/ALK translocations. We present OS adjusted for the potential bias introduced by switching treatment from chemotherapy to pembrolizumab after a median of 25.2 months of follow-up.
Patients were randomized to pembrolizumab 200 mg every 3 weeks or investigator-choice platinum-doublet chemotherapy with optional pemetrexed maintenance for patients with nonsquamous histology. Patients in the chemotherapy arm could switch to pembrolizumab (prior to the second interim analysis that revealed superiority, only patients with progressive disease confirmed by blinded, independent central radiology review were eligible). Three statistical methods were applied to adjust for treatment switching: the simplified 2-stage method, the rank preserving structural failure time (RPSFT) method, and the inverse probability of censoring weighting (IPCW) method.
305 patients were randomized; all but 1 patient in the chemotherapy arm received protocol-specified treatment. 82 (54%) patients had switched from chemotherapy to pembrolizumab per protocol. OS results per the intent-to-treat (ITT) analysis and the 3 adjustment methods are listed in the Table. The 3 methods adjusting for treatment switching in the chemotherapy arm provided larger treatment estimates (lower HRs) than the ITT estimate. As there was some evidence towards a deviation from the common treatment effect assumed in the RPSFT method, and the IPCW method is more prone to bias in the presence of relatively small sample sizes, the simplified 2-stage method was the favored method in this study. Results obtained with the RPSFT and IPCW methods were consistent with those from the 2-stage analyses.Table: (Abstract: 188P)Treatment switching–adjusted OS in KEYNOTE-024 after median follow-up of 25.2 months
Median OS, months (95% CI) Pembrolizumab(n = 154) Chemotherapy(n = 151) HR (95% CI) ITT analysis 30.0 (18.3–not reached) 14.2 (9.8–19.0) 0.63 (0.47–0.86) 2-sided log-rank P = 0.003 2-stage method N/A 8.7 (7.3–11.5) 0.49 (0.34–0.69) RPSFT method N/A 11.8 (8.7–not reached) 0.52 (0.33–0.75) IPCW method N/A 11.8 (8.7–21.3) 0.52 (0.33–0.80)
In this study, pembrolizumab demonstrated an OS advantage compared to chemotherapy as first-line therapy, which becomes more pronounced when utilizing treatment switching analyses that adjust for bias introduced when switching from chemotherapy to pembrolizumab. Pembrolizumab remains a standard-of-care first-line therapy for patients with advanced NSCLC expressing PD-L1 TPS ≥50%.
Clinical trial identification:
Legal entity responsible for the study:
Merck & Co., Inc., Kenilworth, NJ, USA
This study and medical writing assistance were funded by Merck & Co., Inc., Kenilworth, NJ, USA.
M. Reck: Advisory board member for Roche, Lilly, AstraZeneca, BMS, MSD, Merck, Boehringer-Ingelheim, Celgene, Pfizer, and Novartis; Speakers’ bureaus for Roche, Lilly, AstraZeneca, BMS, MSD, Merck, Boehringer-Ingelheim, Celgene, Pfizer, and Novartis; Research funding from Boehringer-Ingelheim. A.G. Robinson: Honoraria from Merck & Co., Inc., Kenilworth, NJ; his institution has received research funding from AstraZeneca, Merck & Co., Inc. (Kenilworth, NJ USA), Bristol-Myers Squibb, and Roche Canada. R. Hui: Honoraria from MSD, BMS, AstraZeneca, Boehringer-Ingelheim, and Pfizer; Consulting or Advisory role for MSD, AstraZeneca, Pfizer, and Novartis. K. Vandormael: Employee of MSD Europe. W. Malbecq: Employee of and owns stock in MSD Europe. M.C. Pietanza: Employee of Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ USA. J. Brahmer: Advisory board member for BMS (uncompensated), Celgene, Eli Lilly, Merck; Research funding from BMS, MedImmune/AstraZeneca, Merck. All other authors have declared no conflicts of interest.