Trifluridine/tipiracil and pembrolizumab combination as later-line treatment in patients with advanced gastric cancer: A multi-cohort phase II study with lead-in safety cohort (LonKey trial).

389 Background: Despite recent advances in therapeutics, including immune checkpoint inhibitors (ICIs), approved third-line treatments for advanced gastric cancer (AGC) remain limited to trastuzumab deruxtecan for HER2-positive disease and trifluridine/tipiracil (FTD/TPI; Lonsurf). However, in the TAGS phase III trial, third- or later-line FTD/TPI monotherapy showed limited efficacy, with an objective response rate (ORR) of 4%, median progression-free survival (PFS) of 2.0 months, and overall survival (OS) of 5.7 months. Methods: This open-label, multi-institutional, multi-cohort phase II trial evaluated the efficacy and safety of FTD/TPI plus pembrolizumab as a third- or later-line palliative treatment for AGC. A safety lead-in (SLI) part using a 3+3 design with two dose levels determined the recommended phase II dose (RP2D). Phase II included two cohorts: cohort 1 (n=45) comprised ICI-naïve patients, while cohort 2 (n=30) included ICI-pretreated patients. The primary endpoints were RP2D (SLI) and ORR (phase II); secondary endpoints included disease control rate (DCR), PFS, OS, and safety. Results: Between December 2022 and November 2024, 81 patients were enrolled (SLI, n=6; phase II, n=75). In the SLI part, one of six patients experienced a dose-limiting toxicity (grade 4 neutropenia), and dose level 0 (pembrolizumab 400 mg IV every 6 weeks plus FTD/TPI 35 mg/m² orally BID on days 1–5 and 8–12 every 4 weeks) was selected as the RP2D. Among all enrolled patients, 40% (n=30) received this regimen as fourth-line or beyond. At the time of data cutoff (March 31, 2025), the median follow-up duration was 19.1 months. In cohort 1, ORR and DCR were 2.2% and 37.8%, with median PFS and OS of 2.1 and 5.2 months, respectively. In cohort 2, ORR and DCR were 3.4% and 55.2%, with median PFS and OS of 3.1 and 7.1 months, respectively. Patients who received the regimen as third-line therapy had longer OS compared to those treated at fourth-line or beyond: 6.0 vs. 3.5 months (P=0.233) in cohort 1 and 7.8 vs. 3.8 months (P=0.047) in cohort 2. The regimen was generally tolerable; grade ≥3 adverse events were primarily hematologic, including neutropenia (32%) and anemia (9.3%). Conclusions: FTD/TPI combined with pembrolizumab demonstrated modest antitumor activity in unselected AGC patients receiving third- or later-line treatment. A potential signal of efficacy in the ICI-pretreated patients (cohort 2) supports further exploration of ICI rechallenge combined with cytotoxic therapy. Clinical trial information: NCT05508737 .