Phase 2 study of cabozantinib (cabo) and pembrolizumab (pembro) in metastatic gastric/ gastroesophageal adenocarcinoma (mGEA) refractory to immune checkpoint inhibitors (ICIs).

345 Background: Most mGEA patients (pt) progress on therapy including ICI, with about 5% progression-free at 6mo (PFS-6) with single agent ICI in later lines of therapy. Alternative therapeutic approaches are needed. Cabo is a multi-tyrosine kinase inhibitor which might modulate immunosuppression in the tumor microenvironment. This study reports on safety and efficacy of the combination of Cabo+Pembro in mGEA refractory to ICI. Methods: Investigator-initiated, single‐arm, two-stage, single institution, phase 2 study in pts with mGEA after ≥1L of fluoropyrimidine and platinum-containing regimens. Additionally, pts with a PD-L1 CPS score ≥10% had also progressed on ICI. Cabo dose was 40mg p.o. daily on days 1 through 21 of a 21‐day cycle, with Pembro 200 mg i.v. on day 1. The primary endpoint was PFS-6, with H0=5%, Ha≥20%. Stage one results (n=19) exceeded the futility threshold. The complete design requires 27 evaluable pts and n≥4 pts with PFS-6 to reject H0 in favor of Ha. Results: 25 pts are currently enrolled. Because stage one did not signal futility, enrollment continues. Median age 58 years (24-75), female (n=13), ECOG 0/1= 13/12, GC/GEJ= 14/11, Caucasian/Hispanic/Asian/other = 7/7/8/3. After a median follow-up of 5.1mo (1.2-24.6), the median PFS and OS are 2.9mo (95% CI 2.1-4.6) and 9.5mo (95% CI 3.1-14.6), respectively. 11/25 had disease progression on or before the first imaging timepoint at 8 weeks, i.e. the clinical benefit rate is 56% (14/25). 13/20 pts had measurable disease by RECISTv1.1: PR 1(7.7%), SD 7(41.2%), PD 5(38.5%). The most common any grade (G) treatment related adverse events (TRAE) were diarrhea (28%), fatigue (20%), and hypertension (16%). G3-4 TRAE were seen in n=3 pts (Hypertension, Thromboembolic event, esophageal perforation; each n=1). No G5 were reported. Conclusions: In this interim analysis of this ongoing study, the combination of Cabo+Pembro shows promising clinical benefit with an estimated OS of 9.5mo in mGEA pts resistant to prior ICI. Follow-up of the primary endpoint is ongoing. The regimen appears tolerable with no new safety signals. More mature data including the primary endpoint and correlative studies will be presented at the meeting. Clinical trial information: NCT04164979 .