1382P Phase Ib/II Open-Label, Randomised Evaluation of Second-Line Atezolizumab (atezo) + Linagliptin (lina) Vs Ramucirumab (ram) + Paclitaxel (pac) in MORPHEUS-gastric Cancer

The MORPHEUS platform consists of multiple, global, open-label, randomised phase Ib/II trials designed to identify early efficacy signals and safety of treatment (tx) combinations across cancers. Atezo (anti–PD-L1) was tested with lina, a dipeptidyl peptidase-4 (DPP-4) inhibitor, in patients (pts) with locally advanced or metastatic gastric or gastroesophageal junction cancer (GC) within MORPHEUS-GC. Given the role of DPP-4 in the regulation of several pro-inflammatory chemokines and intratumoural T-cell recruitment, we hypothesised that adding lina to atezo could stimulate anti-tumour immune responses in GC. Eligible GC pts who progressed on first-line platinum- or fluoropyrimidine-containing regimen received atezo (1200 mg IV every 3 weeks) with lina (5 mg orally daily). Control tx was ram + pac. Primary endpoints were confirmed objective response rate (ORR, investigator-assessed RECIST 1.1) and safety. Secondary endpoints included disease control rate (DCR; response and/or stable disease [SD] ≥ 12 weeks), duration of response (DOR), progression-free survival (PFS) and overall survival (OS). 14 pts received atezo + lina and 12 ram + pac and were evaluable for efficacy and safety (data cutoff: 11 July 2019). Confirmed best overall responses were: 3 partial responses (PR; ORR 21%), 2 SD and 8 progressive disease (PD) with atezo + lina and 2 PR (ORR 17%), 8 SD and 2 PD with ram + pac. DCR was 21% with atezo + lina and 67% with ram + pac. DOR was 3.5 to 7.4+ (censored) mo and 2.9 to 3.7 mo, respectively. Median PFS was 2.0 mo (95% CI: 1.6, 4.9) with atezo + lina and 5.8 mo (95% CI: 3.7, 9.2) with ram + pac. OS data were immature. 50% of atezo + lina and 75% of ram + pac pts had a Gr 3-4 adverse event (AE), with no Gr 5 AEs in either arm. 29% and 92% of pts had dose modifications/interruptions due to AEs of any grade, respectively. 1 pt in both study arms had an AE leading to tx withdrawal. Updated data, including biomarkers and pharmacokinetics, will be presented. Limited activity was seen with atezo + lina; 3 responders had long DOR. However, 8 of 14 pts had PD as best response. AEs were consistent with each agent’s known safety profile.