TPS57 Background: Hepatocellular carcinoma (HCC) is a frequent cause of cancer-related death worldwide and incidence is rising. Sorafenib and lenvatinib are approved for first-line use in unresectable HCC; however, survival benefit remains low. Nivolumab, a human monoclonal antibody to programmed death-1 (PD1) receptor, has been approved as second-line therapy in patients with unresectable HCC refractory to sorafenib. BMS-986205 selectively inhibits human indoleamine-2,3- dioxygenase 1 (IDO1), an intracellular enzyme expressed in various malignancies that promotes immune tolerance. Thus, the combination of PD-1 antibody and IDO1 inhibitor is a promising strategy to augment immunotherapy against HCC. We aim to evaluate the combination of BMS-986205 with nivolumab 3 mg/kg every 2 weeks in unresectable and metastatic HCC. Methods: Primary objectives were to identify tolerable dosing of BMS-986205, and to evaluate safety, tolerability, and efficacy of combination therapy, the latter defined by objective response rate. Secondary objectives were to evaluate disease control rate, duration of response, progression free survival, and overall survival. Phase I (dose escalation phase) will monitor two dose levels of BMS-986205: 50mg daily with conditional escalation to 100mg daily depending on dose-limiting toxicity. Phase II (expansion phase) will begin after maximum tolerated dose is reached. The phase II cohort will be run as a Simon two-stage design to allow for early trial stoppage for futility. Estimated study duration is 36 months with a 24-month accrual period for a total of 17 patients. Patients will be followed for 100 days after the last dose of treatment or until all treatment-related clinical toxicities resolve to baseline or grade ≤ 1. Assessments prior to therapy include imaging, tissue biopsy, serum correlatives, and stool samples. Tissue biopsy will be repeated at 4 weeks. Imaging will be repeated at 8 and 16 weeks, then every 12 weeks until end of treatment. Serum correlatives will be repeated intermittently throughout. At progression, imaging, tissue biopsy, and serum correlatives will be repeated. Data will be reported using 95% confidence intervals, Kaplan-Meier estimates, and frequency tables. Clinical trial information: NCT03695250.