Abstract CT034: Results from phase 3 KEYLYNK-008: Pembrolizumab (pembro) with and without maintenance olaparib (ola) after first-line (1L) pembro plus chemotherapy (chemo) for metastatic squamous non-small-cell lung cancer (sqNSCLC)

Abstract Introduction: 1L pembro + chemo significantly improved OS vs chemo for metastatic sqNSCLC in the KEYNOTE-407 study. Poly(ADP-ribose) polymerase inhibitors (PARPi) can upregulate PD-L1 expression and promote immune-mediated response, which may increase the efficacy of anti‒PD-(L)1‒based therapies. The randomized, double-blind (and in-house blinding), phase 3 KEYLYNK-008 study (NCT03976362) evaluated addition of ola (PARPi) as maintenance therapy for metastatic sqNSCLC with SD or CR/PR after 1L pembro + chemo. Methods: Eligible patients (pts) aged ≥18 y had previously untreated stage IV NSCLC, confirmed squamous histology, measurable disease per RECIST v1.1, and ECOG PS 0-1. Pts with CR, PR, or SD after 4 cycles of Q3W induction therapy with pembro 200 mg, carboplatin AUC 6, and paclitaxel 200 mg/m2 or nab-paclitaxel 100 mg/m2 were randomized 1:1 to ola 300 mg BID or placebo (pbo), both given with ≤31 cycles of pembro 200 mg Q3W. Randomization was stratified by ECOG PS (0 vs 1), PD-L1 TPS (<50% vs ≥50%), and response to induction treatment (CR/PR vs SD). Primary endpoints were PFS (RECIST v1.1 by BICR) and OS. Final PFS testing occurred at interim analysis (IA) 2 (data cutoff, Feb 23, 2023); other endpoints were assessed at IA3 (data cutoff, Sep 21, 2023). Efficacy boundaries based on actual observed events were 1-sided P = 0.003202 for PFS (IA2) and P = 0.014647 for OS (IA3). Results: Of 851 pts who received induction treatment, 591 were randomized to pembro + ola (n = 296) or pembro + pbo (n = 295). Median PFS (95% CI) was 8.3 (6.7-9.7) mo for pembro + ola and 5.4 (4.1-5.6) mo for pembro + pbo (HR 0.77; 95% CI 0.63-0.93; P = 0.0040). 24-mo PFS rates (95% CI) were 27.5% (21.9%-33.4%) and 21.1% (16.1%-26.5%), respectively. Median follow-up to IA3 was 33.4 (19.0-46.8) mo. Median (95% CI) OS was 19.1 (15.9-22.2) mo for pembro + ola and 18.6 (16.0-21.6) mo for pembro + pbo (HR 1.01; 95% CI 0.83-1.24; P = 0.5481). 24-mo OS rates (95% CI) were 40.9% (35.1%-46.5%) vs 41.2% (35.4%-46.8%). ORR (95% CI) was 25.0% (20.2%-30.3%) for pembro + ola vs 15.3% (11.3%-19.9%) for pembro + pbo. Median DOR (range) was 21.0 (2.1 to 42.8+) mo vs 26.0 (1.2+ to 39.6+) mo. 225/294 pts (76.5%; grade 3-5, 29.6%) had treatment-related AEs in the pembro + ola arm and 190/292 (65.1%; grade 3-5, 12.7%) in pembro + pbo arm. 2 pts (pneumonitis, multiple organ dysfunction syndrome) in the pembro + ola arm and 1 (thrombocytopenia) in the pembro + pbo arm died due to treatment-related AEs. Conclusions: In KEYLYNK-008, the addition of ola to pembro as maintenance therapy after 1L pembro + chemo for metastatic sqNSCLC numerically prolonged PFS vs pembro + pbo, but neither PFS nor OS met the prespecified statistical significance boundary. No new safety signals were identified. The study has been discontinued per the recommendation of an independent data monitoring committee. Citation Format: Maximilian Hochmair, Michael Schenker, Manuel Cobo Dols, Tae Min Kim, Sang-We Kim, Ozgur Ozyilkan, Maria Smagina, Leonova Viktoriya, Terufumi Kato, Alexander Fedenko, Flavia De Angelis, Achim Rittmeyer, Jhanelle E. Gray, Alastair Greystoke, Qinlei Huang, Bin Zhao, Humberto Lara-Guerra, Ernest Nadal. Results from phase 3 KEYLYNK-008: Pembrolizumab (pembro) with and without maintenance olaparib (ola) after first-line (1L) pembro plus chemotherapy (chemo) for metastatic squamous non-small-cell lung cancer (sqNSCLC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr CT034.