Tiragolumab in combination with atezolizumab and bevacizumab in patients with unresectable, locally advanced or metastatic hepatocellular carcinoma (MORPHEUS-Liver): a randomised, open-label, phase 1b–2, study

PD-L1 and VEGF blockade with atezolizumab plus bevacizumab has been shown to improve survival in unresectable hepatocellular carcinoma. TIGIT is an immune checkpoint regulator implicated in many cancers, including unresectable hepatocellular carcinoma. Here, we evaluate the clinical activity and safety of the addition of tiragolumab, an anti-TIGIT monoclonal antibody, to atezolizumab plus bevacizumab. This randomised, open-label, phase 1b-2 umbrella study was conducted at 26 centres across China, France, Israel, New Zealand, South Korea, Taiwan, and the USA. Eligible patients were adults aged 18 years old or older with previously untreated locally advanced unresectable hepatocellular carcinoma, an Eastern Cooperative Oncology Group performance status of 0-1, Child-Pugh class A disease, and a life expectancy of at least 3 months. Eligible patients were randomly assigned (2:1) using permuted block randomisation to receive either tiragolumab 600 mg plus atezolizumab 1200 mg plus bevacizumab 15 mg/kg or atezolizumab 1200 mg plus bevacizumab 15 mg/kg, administered via intravenous infusion every 3 weeks on day 1 of each 21-day cycle. Patients received treatment until unacceptable toxic effects or loss of clinical benefit, whichever occurred first. The primary endpoint was objective response rate. Analysis of clinical activity was done in the efficacy-evaluable population (all patients who received at least one dose of each drug for their assigned treatment regimen) and safety was assessed in all patients who received any study treatment. The trial is registered with ClinicalTrials.gov, NCT04524871, and is ongoing. Between Aug 20, 2020, and Feb 10, 2022, we assessed 154 patients for eligibility and 59 eligible patients were randomly assigned to receive tiragolumab plus atezolizumab plus bevacizumab (n=41) or atezolizumab plus bevacizumab (n=18); one patient in the tiragolumab plus atezolizumab plus bevacizumab group experienced an adverse event before receiving any treatment and withdrew from the study. Median age was 65·0 years (IQR 61·0-73·0). 46 (79%) of 58 patients were male and 12 (21%) were female. Most patients were Asian (23 [40%]) or White (21 [36%]). At the time of clinical cutoff (Aug 21, 2023), median follow-up was 20·6 months (IQR 10·6-28·0) in the tiragolumab plus atezolizumab plus bevacizumab group and 14·0 months (4·2-18·5) in the atezolizumab plus bevacizumab group. The confirmed objective response rate was 43% (95% CI 27-59, n=17) in the tiragolumab plus atezolizumab plus bevacizumab group and 11% (1-35, n=2) in the atezolizumab plus bevacizumab group. All patients in both groups experienced an adverse event. The incidence of pruritis (20 [50%] of 40 patients vs three [17%] of 18 patients), arthralgia (13 [33%] vs two [11%]), and diarrhoea (12 [30%] vs one [6%]) was notably higher in the tiragolumab plus atezolizumab plus bevacizumab group than in the atezolizumab plus bevacizumab group, although these were mainly grade 1-2. The most common grade 3-4 adverse events were hypertension (six [15%] of 40 patients in the tiragolumab plus atezolizumab plus bevacizumab group vs two [11%] of 18 patients in the atezolizumab plus bevacizumab group), aspartate aminotransferase increased (three [8%] of 40 patients vs one [6%] of 18 patients), and proteinuria (two [5%] of 40 patients vs two [11%] of 18 patients). Serious adverse events occurred in 21 (53%) of 40 patients in the tiragolumab plus atezolizumab plus bevacizumab group and in ten (56%) of 18 patients in the atezolizumab plus bevacizumab group. Treatment-related deaths occurred in one patient in the tiragolumab plus atezolizumab plus bevacizumab group (due to cholestasis) and two patients in the atezolizumab plus bevacizumab group (due to oesophageal varices haemorrhage and upper gastrointestinal haemorrhage). The addition of tiragolumab to atezolizumab plus bevacizumab did not appear to result in a substantial worsening of treatment-related or immune-mediated adverse events, and no new safety signals were identified. This signal-seeking study suggests that the addition of tiragolumab to atezolizumab and bevacizumab might be more clinically active than atezolizumab plus bevacizumab alone in unresectable hepatocellular carcinoma. Based on these data, further study of combination tiragolumab plus atezolizumab plus bevacizumab is warranted. F Hoffmann-La Roche and Genentech.