LAUSR

2553 Background: Spartalizumab and LAG525 are monoclonal antibodies targeting PD-1 and LAG-3, respectively. Dual blockade of PD-1 and LAG-3 has shown synergistic antitumor activity in preclinical models. Here we describe preliminary efficacy of spartalizumab + LAG525 across seven tumor types. Methods: Phase II, open-label, parallel-cohort study was conducted in pts with solid or hematologic malignances relapsed and/or refractory to standard-of-care therapies. Prior immunotherapy was prohibited. LAG525 (400 mg) and spartalizumab (300 mg) were dosed intravenously every 3 weeks. Primary endpoint was clinical benefit rate at 24 weeks (CBR24), assessed using a Bayesian hierarchical model-based futility analysis. Posterior probability that clinical benefit exceeds historical control (Pr) was estimated to determine futility at interim against prespecified thresholds. Results: As of January 7 2019, 76 pts received spartalizumab + LAG525; 72 pts were eligible for analysis (Table). The Pr cut-off for all arms was > 0.70. Hence, neuroendocrine tumors (NET), small cell lung cancer (SCLC) and diffuse large B-cell lymphoma (DLBCL) cohorts all met the expansion criteria with Pr of 0.971, 0.975 and 0.804 respectively. CBR24 were as follows; NET: 0.86, SCLC: 0.27, DLBCL: 0.43. Prostate, sarcoma and ovarian cohorts did not meet the expansion criterion (Pr > 0.70) but were not declared futile; enrollment was paused pending results of next analysis. Gastroesophageal (GE) cancer cohort was stopped for futility due to Pr (0.071) below the futility threshold. Conclusions: Spartalizumab and LAG525 showed promising activity in NET, SCLC and DLBCL that met expansion criteria. The GE cohort was declared futile. Remaining cohorts are paused pending further analysis. Clinical trial information: NCT03365791. [Table: see text]