Abstract

4012 Background: Anti-PD-1 monoclonal antibody (mAbs) plus chemotherapy has become standard first-line therapy for patients (pts) with advanced/metastatic G/GEJ cancer. In this phase Ib/II study (NCT04982276), we assessed the safety and efficacy of cadonilimab, a PD-1/CTLA-4 bispecific antibody, in combination with pulocimab, an anti-VEGFR-2 mAb, and paclitaxel in G/GEJ cancer pts who failed immunochemotherapy. Methods: This study consisted of two parts: safety run-in (arms A, B and C, n = 6 each) and expansion (a randomized, double-blind phase II study). In safety part, pts received intravenous pulocimab 10mg/kg, every 2 weeks (Q2W) + paclitaxel 80 mg/m 2 , days 1, 8, 15 (arm A), or cadonilimab 10mg/kg + pulocimab 10mg/kg, Q2W (arm B), or cadonilimab 10mg/kg + pulocimab 10mg/kg + paclitaxel 150 mg/m 2 , Q3W (arm C). In expansion part, pts were randomly assigned at 1:1 ratio to receive cadonilimab 10mg/kg + pulocimab 10mg/kg + paclitaxel 150 mg/m 2 , Q3W (arm 1) or placebo + pulocimab + paclitaxel (arm 2). The primary endpoint was safety and objective response rate (ORR), secondary endpoints included progression-free survival (PFS) and overall survival (OS). Results: As of 30 Oct. 2023, a total of 77 pts were enrolled. In safety part (n = 18), dose-limiting toxicity (DLT) was observed in 1/6 pts in arm B, whereas no DLT was observed in either arm A or C. In expansion part, 59 pts were randomly assigned to arm 1 (n = 29) and arm 2 (n = 30). 6 pts were excluded due to treatment interruptions caused by Covid-19 epidemic between September and December, 2022. Baseline characteristics were generally balanced between treatment arms. With median follow-up of 7.3 months (mo, range: 1.4-16.1), 53 pts had at least one post-baseline tumor evaluation. ORR was 48.0% (12/25) for arm 1 vs 35.7% (10/28) for arm 2, disease control rate (DCR) was 96.0% (24/25) vs 92.9% (26/28), mPFS was 6.8 mo (95% CI 4.1, 11.2) vs 4.9 mo (95% CI 3.2, 7.1) , median duration of response (mDoR) was not reached vs 4.0 mo (95%CI 1.58, NE), the median OS was not reached for the two arms, 9-mo OS rate was 65.5% (95% CI 33.2, 85.0) vs 34.0% (95% CI 9.5, 60.9). Grade 3-4 treatment-related adverse events (TRAEs) occurred in ≥10% pts were neutrophil count decreased (27.6% vs 33.3%), white blood cell count decreased (10.3% vs 26.7%) and blood pressure increased (13.8% vs 10.0%). No new safety signals were identified. Conclusions: The combination of cadonilimab with pulocimab and paclitaxel demonstrated manageable safety profile and encouraging efficacy as second-line therapy in pts with G/GEJ cancer progressed after immunochemotherapy. Incorporation of dual immunotherapy to VEGFR-2 targeting therapy could be a potential strategy to overcome immunotherapy resistance. Additional phase III study is warranted to confirm it. Clinical trial information: NCT04982276 .