Phase Ib study of BI 836880 (VEGF/Ang2 inhibitor) plus ezabenlimab (BI 754091; anti-PD-1 antibody) in patients (pts) with advanced hepatocellular carcinoma (HCC).

Abstract

434 Background: Combination of anti-VEGF compounds and immune checkpoint inhibitors is an approved therapy across multiple solid tumors, including advanced HCC. This phase Ib study (NCT03468426) is assessing BI 836880 (bispecific VEGF/Ang2 nanobody) + ezabenlimab (anti-PD-1 antibody) in pts with advanced solid tumors. The recommended phase 2 dose (RP2D) of BI 836880 720 mg + ezabenlimab 240 mg given IV every 3 weeks was determined in Part 1. In Part 2, RP2D was assessed in 7 expansion cohorts. We report data from cohorts in HCC after prior sorafenib/lenvatinib (cohort F) and untreated unresectable HCC (cohort G). Methods: Pts with locally advanced or metastatic HCC, Child-Pugh class A, not eligible for surgical or locoregional therapies were enrolled. Cohort F enrolled pts who had progressed on or after first-line treatment with sorafenib or lenvatinib or who had discontinued due to poor tolerability after ≥2 weeks of treatment. Cohort G enrolled pts who had received no prior systemic therapy for HCC. Treatment continued until disease progression, undue toxicity or consent withdrawal. Primary endpoint is objective response rate (ORR) by RECIST 1.1. Results: As of Aug 2021, 30/31 pts have been treated in cohorts F/G: 87/77% male; median age 65/64 yrs. Follow-up is ongoing in both cohorts. 9/19 pts in cohorts F/G remain on treatment; median (range) duration of treatment is 175 (42–532)/ 169 (42–336) days in cohorts F/G. All pts were evaluable for response in cohort F: confirmed ORR to date is 40% (1 complete response; 11 partial responses [PRs]). Of 28 evaluable pts in cohort G, confirmed ORR to date is 21% (6 PRs). 12 (40%) pts in cohort F and 18 (64%) in cohort G have stable disease. In cohort F, AEs were reported in 28 (93%) pts, most frequently hypertension and proteinuria (each 30%). In cohort G, AEs occurred in 26 (84%) pts, most frequently ascites (26%) and thrombocytopenia (19%). 24 (80%) pts in cohort F and 15 (48%) in cohort G had treatment-related AEs (TRAEs). Most frequent TRAEs were proteinuria (27%), infusion-related reactions (IRRs) and hypertension (each 20%) in cohort F, and hypertension (13%), IRRs, hypothyroidism and diarrhea (each 10%) in cohort G. There were 3 pts with G5 AEs in cohort F (COVID-19 pneumonia [n = 1]; low Glasgow coma score and dyspnea [n = 1]; hepatic cirrhosis [n = 1]) and 1 G5 AE in cohort G (hepatic failure); none were considered related to treatment. AEs leading to discontinuation occurred in 2 pts in cohort F (G3 hepatic encephalopathy and G2 duodenal ulcer) and 3 in cohort G (G5 hepatic failure [n = 1]; G2 acute kidney injury and G1 decreased appetite [n = 1]; G2 diarrhea [n = 1]). Conclusions: BI 836880 + ezabenlimab had a manageable safety profile and showed promising activity in pts with untreated and second-line post-sorafenib/lenvatinib advanced HCC. Data continue to mature, particularly in cohort G. Cohort F has been expanded by a further 30 pts. Clinical trial information: NCT03468426.