Durvalumab with tremelimumab produced responses with manageable toxicity in sicker liver cancer patients

Design

Phase IIIb, single-arm, open-label study (SIERRA)

Population

111 adults with unresectable hepatocellular carcinoma: Child-Pugh B7/B8 (44), ECOG performance status 2 (47), main trunk portal vein thrombosis (20)

Primary outcome

Grade 3 or 4 treatment-related adverse events within 6 months, and objective response rate

Effect

Grade 3/4 events 19.8% (95% CI 12.9 to 28.5); objective response 14.4% (95% CI 8.5 to 22.4)

SIERRA was a phase IIIb, single-arm study of the STRIDE regimen (a single dose of tremelimumab 300 mg with durvalumab 1500 mg, then durvalumab 1500 mg every four weeks) in 111 adults with unresectable hepatocellular carcinoma who would be excluded from many trials. They fell into three cohorts: Child-Pugh B7 or B8 liver function (44), ECOG performance status 2 (47) and main trunk portal vein thrombosis (20). The co-primary end points were grade 3 or 4 treatment-related adverse events within six months and objective response rate.

Grade 3 or 4 treatment-related events occurred in 19.8% (95% CI 12.9 to 28.5): 18.2% in Child-Pugh B, 19.1% in performance status 2 and 25.0% in portal vein thrombosis. The objective response rate was 14.4% (95% CI 8.5 to 22.4): 6.8%, 21.3% and 15.0% respectively. The authors describe a favourable risk-benefit profile.

There is no comparison group and survival results are not in the abstract. Response was lowest in the Child-Pugh B group. It supports considering immunotherapy for patients who were previously not studied, in specialist centres, and it does not show a survival gain. Access and cost in India are not addressed.

• Grade 3 or 4 treatment-related events within six months occurred in 19.8% of 111 patients.
• Objective response was 14.4% overall: 6.8% in Child-Pugh B, 21.3% in performance status 2 and 15.0% with portal vein thrombosis.
• Safety looked similar across the three poor-prognosis groups.
• Consider the regimen in selected patients with Child-Pugh B7 or B8, performance status 2 or portal vein thrombosis, at specialist centres.
• There is no control arm and survival data are not given in the abstract.

Why it matters

Up to two thirds of real-world patients with unresectable liver cancer have poor-prognosis features that trials exclude.

Don't overread it

A single-arm phase IIIb study shows tolerability and response, not that survival improves.

The statistics, in plain English

In a single-arm study, response rate and toxicity can be described but there is nothing to compare them with, so benefit cannot be proved. The confidence interval for response in the Child-Pugh B group (1.4% to 18.7%) is wide, because only 44 patients were in it.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.