Updated results from a phase I trial of M7824 (MSB0011359C), a bifunctional fusion protein targeting PD-L1 and TGF-β, in patients with pretreated recurrent or refractory gastric cancer

Abstract

Background: Objective response rates (ORRs) in patients (pts) with gastric cancer (GC) treated with anti-PD-(L)1 antibodies range from 11.2% (PD-L1 unselected) to 15.5% (PD-L1+) for second line therapy. Inhibiting the transforming growth factor b (TGF-b) pathway, which plays a key role in tumor immunosuppression, may enhance the response to anti-PD-L1 treatment. We report results for M7824, an innovative first-in-class bifunctional fusion protein composed of a monoclonal antibody against PD-L1 fused with the extracellular domain of TGF-b receptor II (a TGF-b ?trap?) in pts with GC. Methods: Pts in Asia with recurrent GC or gastroesophageal junction adenocarcinoma for whom standard therapy does not exist or has failed were enrolled in this expansion cohort of the ongoing, phase 1, open-label trial NCT02699515 and received M7824 1200 mg q2w until disease progression, unacceptable toxicity, or trial withdrawal. The primary objective is to assess safety/tolerability; secondary objectives include assessment of best overall response (BOR) per RECIST v1.1. Results: As of February 15, 2018, 31 heavily pretreated pts with advanced GC (74.2% received ≥ 3 prior therapies) received M7824 for a median duration of 10.1 (range, 2- 52) wks; 4 pts remained on treatment. 6 pts (19.4%) had grade 3 treatment-related adverse events (TRAEs): anemia (2), diarrhea (1), abnormal hepatic function (1) and rash (2). No grade 4 TRAEs occurred. 1 Grade 5 AE of death occurred after 5 doses with suspected rupture of pre-existing thoracic aortic aneurysm as per investigator's assessment. 7 pts had a confirmed objective response based on investigator-assessed BOR (ORR, 22.6%; DCR, 38.7%). There were 5 partial responses (2.4, 3.6+, 4.1+, 8.3+, and 9.7+ mo) and 2 complete responses (0.3+ and 9.0 mo). The initial assessment of PDL1 tumor expression (22C3 assay) does not hint towards predictivity for ORR. Conclusions: M7824 monotherapy had a manageable safety profile in heavily pretreated Asian pts (74% ≥3 prior therapies) with GC. Early signs of clinical efficacy, with an ORR of 22.6% and long lasting responses, are encouraging. Updated data will be presented.