Abstract 1262P
Background
In KEYNOTE-010, pembro improved OS vs doce as 2L+ therapy for advanced NSCLC with PD-L1 TPS ≥1% and ≥50%. KEYNOTE-010 did not enroll any pts from mainland China, which has high NSCLC mortality. KEYNOTE-033 (NCT02864394) evaluates pembro vs doce in pts with previously treated advanced NSCLC with PD-L1 TPS ≥1%, with most pts enrolled in mainland China.
Methods
Eligible pts (≥18 y) were randomized to pembro 2 mg/kg Q3W (35 cycles) or doce 75 mg/m2 Q3W (per local standard of care), stratified by TPS (≥50% vs 1–49%). Response was assessed Q9W per RECIST v1.1 by BICR. PD-L1 expression was assessed centrally (PD-L1 IHC 22C3 pharmDx assay). OS and PFS (primary objectives) were evaluated sequentially using stratified log-rank tests, first in pts with TPS ≥50% and then in pts with TPS ≥1% (1-sided α=0.025).
Results
425 pts were enrolled. At data cutoff (Sep 9, 2019), median follow-up was 18.8 (range, 0.2-38.8) mo, and 291 (68%) pts had died. Pembro numerically improved OS in all groups analyzed, but did not achieve predefined statistical significance in pts with PD-L1 TPS ≥50% (Table); thus, sequential testing of OS and PFS ceased. HR for OS in TPS ≥1% pts from mainland China (n=311) was 0.68 (95% CI, 0.51–0.89). In all treated pts, incidence of treatment-related AEs was lower with pembro vs doce (any grade, 70% vs 88%; grade 3-5, 11% vs 47%). Table: 1262P
| Pembro | Doce | ||
| PD-L1 TPS ≥50% | N=114 | N=113 | |
| OS | Median (95% CI), mo | 12.3 (10.0-16.3) | 10.9 (8.3-13.1) |
| HR (95% CI) | 0.83 (0.61-1.14) | ||
| P | 0.1276 | ||
| PFS | Median (95% CI), mo | 4.0 (2.1-8.0) | 2.5 (2.1-4.2) |
| HR (95% CI) | 0.76 (0.54-1.07) | ||
| ORR | % (95% CI) | 28.1 (20.1-37.3) | 7.1 (3.1-13.5) |
| PD-L1 TPS ≥1% | N=213 | N=212 | |
| OS | Median (95% CI), mo | 12.9 (10.3-16.5) | 10.6 (8.7-12.5) |
| HR (95% CI) | 0.75 (0.60-0.95) | ||
| PFS | Median (95% CI), mo | 3.3 (2.1-4.1) | 3.0 (2.3-4.0) |
| HR (95% CI) | 0.84 (0.66-1.08) | ||
| ORR | % (95% CI) | 20.7 (15.4-26.7) | 5.7 (3.0-9.7) |
Conclusions
While pembro did not meet statistical significance for OS in pts with PD-L1 TPS ≥50%, HRs for OS and PFS numerically favored pembro and ORR was higher with pembro in both the PD-L1 TPS ≥50% and ≥1% groups. Toxicity was consistent with the established pembro safety profile. These data support the value of pembro for previously treated advanced NSCLC in China.
Clinical trial identification
NCT02864394.
Editorial acknowledgement
Writing support was provided by Michael S. McNamara, MS, of ICON plc (North Wales, PA, USA), funded by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA.
Legal entity responsible for the study
Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA.
Funding
Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA.
Disclosure
C. Zhou: Honoraria (self): Eli Lily; Honoraria (self): Sanofi; Honoraria (self): Merck Sharp & Dohme Corp.; Honoraria (self): BI; Honoraria (self): Hengrui. B. Li, J. Ge: Full/Part-time employment: MSD China. T. Dang: Full/Part-time employment: Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. All other authors have declared no conflicts of interest.