Skip to content
Bending the Curve Trial by Trial
?

EMERALD-1

Durvalumab with or without bevacizumab with transarterial chemoembolisation in hepatocellular carcinoma (EMERALD-1): a multiregional, randomised, double-blind, placebo-controlled, phase 3 study

Phase 3N = 616NCT03778957

Population
Adults with unresectable HCC amenable to embolisation, ECOG performance status 0 or 1, and at least one measurable intrahepatic lesion
Design
Multiregional, randomised (1:1:1), double-blind, placebo-controlled phase 3 study at 157 sites in 18 countries, stratified by TACE method, region and portal vein invasion: TACE plus durvalumab plus bevacizumab, TACE plus durvalumab plus placebo, or TACE plus placebo alone. The primary end point was progression-free survival by blinded independent central review (BICR) per RECIST 1.1 for durvalumab plus bevacizumab versus placebo alone in the ITT population; overall survival is to be reported later.
Primary endpoint
PFS
Arms
  • TACE plus durvalumab plus bevacizumab (experimental, n = 204): TACE plus durvalumab 1500 mg every 4 weeks then 1120 mg with bevacizumab 15 mg/kg every 3 weeks
  • TACE plus durvalumab (experimental, n = 207): TACE plus durvalumab (same regimen) plus placebo in place of bevacizumab
  • TACE plus placebo (control, n = 205): TACE plus placebo in place of durvalumab and bevacizumab
People
Sangro B
PFSHR 0.77 (95% CI 0.61–0.98)Δ median PFS+6.8 monthsPFSHR 0.94 (95% CI 0.75–1.19)Δ median PFS+1.8 months
Show the numbers
Results reported for EMERALD-1
EndpointAnalysisTACE plus durvalumab plus bevacizumab / TACE plus durvalumabTACE plus placeboEffect
PFSProgression-free survival by BICR per RECIST 1.1 (primary end point), durvalumab plus bevacizumab versus placebo alonefinalmedian follow-up 27.9 monthsSangro 2025median 15 months (11.1–18.9)median 8.2 months (6.9–11.1)HR 0.77 (95% CI 0.61–0.98)P=0.032Two-sided p. Median follow-up for progression-free survival was 27.9 months (95% CI 27.4-30.4). The paper describes this as the final progression-free survival analysis (final-analysis significance threshold 0.044); an earlier interim progression-free survival analysis (13 June 2022) had not met significance.
PFSProgression-free survival by BICR per RECIST 1.1 (key secondary end point), durvalumab plus placebo versus placebo alonefinalmedian follow-up 27.9 monthsSangro 2025median 10 months (9.0–12.7)median 8.2 months (6.9–11.1)HR 0.94 (95% CI 0.75–1.19)P=0.64Two-sided p. Key secondary endpoint, tested after the primary objective. Reported at the same final progression-free survival analysis.
Results reported for EMERALD-1
EndpointAnalysisTACE plus durvalumab plus bevacizumab / TACE plus durvalumabTACE plus placeboEffect
PFSProgression-free survival by BICR per RECIST 1.1 (primary end point), durvalumab plus bevacizumab versus placebo alonefinalmedian follow-up 27.9 monthsSangro 2025median 15 months (11.1–18.9)median 8.2 months (6.9–11.1)HR 0.77 (95% CI 0.61–0.98)P=0.032Two-sided p. Median follow-up for progression-free survival was 27.9 months (95% CI 27.4-30.4). The paper describes this as the final progression-free survival analysis (final-analysis significance threshold 0.044); an earlier interim progression-free survival analysis (13 June 2022) had not met significance.
PFSProgression-free survival by BICR per RECIST 1.1 (key secondary end point), durvalumab plus placebo versus placebo alonefinalmedian follow-up 27.9 monthsSangro 2025median 10 months (9.0–12.7)median 8.2 months (6.9–11.1)HR 0.94 (95% CI 0.75–1.19)P=0.64Two-sided p. Key secondary endpoint, tested after the primary objective. Reported at the same final progression-free survival analysis.
Hazard ratio (95% CI) by endpoint. Values below 1 favour the experimental arm.
PFSPFS0.512◀ experimentalcontrol ▶
Data table
EndpointHR95% CI
PFS0.770.61–0.98
PFS0.940.75–1.19
Illustrative — modelled from reported values, not trial data. PFS, final.
0%25%50%75%100%0102030Months (modelled)
  • TACE plus durvalumab plus bevacizumab (experimental)
  • TACE plus placebo (control)

Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.

Data table (modelled values)
MonthsTACE plus durvalumab plus bevacizumabTACE plus placebo
0100.0%100.0%
966.0%46.7%
1941.6%20.1%
2827.4%9.4%
3817.3%4.0%
Illustrative — modelled from reported values, not trial data. PFS, final.
0%25%50%75%100%0510152025Months (modelled)
  • TACE plus durvalumab (experimental)
  • TACE plus placebo (control)

Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.

Data table (modelled values)
MonthsTACE plus durvalumabTACE plus placebo
0100.0%100.0%
666.0%60.2%
1340.6%33.3%
1926.8%20.1%
2517.7%12.1%
Primary publication: Sangro B, Kudo M, Erinjeri JP, Qin S, Ren Z, Chan SL, et al. Durvalumab with or without bevacizumab with transarterial chemoembolisation in hepatocellular carcinoma (EMERALD-1): a multiregional, randomised, double-blind, placebo-controlled, phase 3 study. Lancet. 2025;405(10474):216-232. PubMedDOI

Appears in

Sources

  1. Sangro B, Kudo M, Erinjeri JP, Qin S, Ren Z, Chan SL, et al. Durvalumab with or without bevacizumab with transarterial chemoembolisation in hepatocellular carcinoma (EMERALD-1): a multiregional, randomised, double-blind, placebo-controlled, phase 3 study. Lancet. 2025;405(10474):216-232.

    PubMedDOI