BACKGROUND: Nivolumab demonstrated durable responses and safety in patients with hepatocellular carcinoma (HCC) with Child-Pugh class A cirrhosis in the CheckMate 040 trial, with rates of hepatotoxicity that were similar to those of non-HCC populations. To the authors' knowledge, the safety and efficacy of nivolumab has not been established in patients with Child-Pugh class B (CPB) cirrhosis, a population with limited therapeutic options and a poor prognosis. METHODS: The authors conducted a retrospective case series of patients with advanced HCC and CPB cirrhosis who were treated with nivolumab and enrolled in the University of California at San Francisco Hepatobiliary Tissue Bank and Registry. Safety endpoints included rates of grade ≥3 adverse events (AEs) (graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events [version 4.03]) and serious AEs, immune-related AEs (irAE), steroid requirement, and discontinuation. Efficacy endpoints included time on treatment, the objective response rate according to Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1, overall survival, and progression-free survival. RESULTS: A total of 18 patients were included, with 72% of them (13 of 18 patients) previously treated with sorafenib. The majority of patients (94%; 17 of 18 patients) experienced a grade ≥3 AE, with treatment-related grade ≥3 AEs reported in 28% of patients (5 of 18 patients). irAEs were reported to occur in approximately 50% of patients (9 of 18 patients), and 28% (5 of 18 patients) required steroids. Treatment-related AEs required discontinuation in 4 patients (22%). The median time on treatment was 2.3 months (95% CI, 1.9 months to upper bound not estimable). The objective response rate was 17% (3 of 18 patients), including 2 partial responses and 1 complete response. The median overall survival from the time of nivolumab initiation was 5.9 months (95% CI, 3 months to upper bound not estimable), with a median progression-free survival of 1.6 months (95% CI, 1.4-3.5 months). CONCLUSIONS: Patients with CPB HCC experienced high rates of AEs, although the frequency of irAEs was similar to that of patients with Child-Pugh class A HCC in the CheckMate 040 trial. A subset of patients experienced prolonged tumor responses. Nivolumab warrants further study in patients with CPB HCC. Cancer 2019;125:3234-3241.

No abstract

TPS464 Background: The antiangiogenic tyrosine kinase inhibitor (TKI) SOR prolongs survival in advanced HCC, but responses occur in fewer than 5% of patients (pts) and median progression-free survival (PFS) is less than 6 months. Immune checkpoint inhibition (CPI) with the PD-1 inhibitor, NIVO, achieved an overall response rate (ORR) of approximately 20% in sorafenib-naïve pts enrolled on CheckMate 040 trial though most pts had progression. In preclinical studies, TKI can inhibit regulatory T cells and myeloid derived suppressor cells, immune cell subsets which may contribute to CPI resistance. CD8+ T cells in sorafenib-resistant tumors are characterized by PD-1 expression, providing rationale for a combined approach. The combination of TKI or the anti-angiogenic bevacizumab with CPI improves anti-tumor activity in HCC mouse models and in preliminary clinical studies. This study will examine the safety, maximum tolerated dose (MTD), and ORR of the combination of SOR plus NIVO in advanced HCC pts, along with correlative analyses of tumor and circulating immune cells. Methods: Eligible pts must have Child-Pugh A liver function and advanced HCC, without prior systemic therapy and measurable by RECIST 1.1. In Part 1 (3+3 dose escalation), SOR dose will be 400 mg QD or BID plus NIVO 240 mg IV Q2 weeks. In Part 2, Arm 1, pts will start NIVO Cycle 1, Day 1 (C1D1), with addition of SOR at MTD on C1D15; in Part 2, Arm 2, SOR is given at MTD on C1D1 with addition of NIVO on C1D15. Primary endpoints are MTD of SOR (Part 1) and ORR by RECIST 1.1 with H0 7.5% vs. H1 25% (Part 2). For expected sample size of 24 evaluable pts in Part 2, the power is 83% with 1-sided alpha 5% to determine ORR >25% by Chi-square tests. Secondary endpoints are safety, duration of response, PFS, and overall survival. Exploratory endpoints include peripheral and tumor immune cell profiling, PD-L1 expression, and alpha-fetoprotein (AFP) response. An interim safety analysis will be performed after 50% enrollment in Part 2. Optional paired biopsies in Part 2 will allow for investigation of the tumor microenvironment on SOR, NIVO, and combination. Clinical trial information: NCT03439891.

496 Background: HCC patients (pts) with Child Pugh B (CPB) cirrhosis have poor prognosis and limited treatment (Tx) options. Nivolumab demonstrated durable responses and acceptable safety in Child Pugh A (CPA) HCC in the CheckMate-040 trial with rates of hepatotoxicity similar to non-HCC populations. The safety and efficacy of nivolumab has not been established in pts with CPB cirrhosis. Methods: Design: Retrospective case series with IRB approval. Key eligibility: HCC with CPB cirrhosis; treated with nivolumab as standard Tx; enrolled in the UCSF Hepatobiliary Tissue Bank and Registry. Study endpoints: Safety during nivolumab Tx including all-cause grade(Gr) ≥ 3 adverse events (AE), serious AE (SAE), any grade immune-related (ir)AE, and systemic steroid (SS) requirement; clinical outcomes including time on Tx (TOT) with nivolumab and overall survival (OS). Results: Thirteen pts were included: male 77%; Asian 38%, white 54%; median age 66 (range: 26-86); HCV Ab+ 31%, HBsAg+ 23%; BCLC B/C 31%/69%; median Child-Pugh score 8; median prior systemic Tx 1 (range: 0-6); prior sorafenib 69%, median duration on prior sorafenib 137 days (range 10-341). The Table depicts safety outcomes on nivolumab. Median TOT on nivolumab: 44 days (95% CI: 32, 98) (range: 17-811+). Median OS from start of nivolumab: 119 days (95% CI: 40, 247) (range: 40-811+). Best response of at least stable disease occurred in 3/13 (23%) of patients, including prolonged stable disease (SD) for 6+ months and complete response (CR) for 24+ months on nivolumab (1 pt each). Conclusions: CPB HCC pts treated with nivolumab experienced high rates of all-cause Gr ≥ 3 AE and SAE and short OS, similar to prior studies in CPB HCC. Rates of irAE attributed to nivolumab were similar to rates reported in CheckMate-040 CPA population, without unexpected AE. A subset of pts experienced prolonged stable disease and CR. Nivolumab warrants further study in CPB HCC.[Table: see text]