CoChat

Nivolumab plus ipilimumab versus lenvatinib or sorafenib as first-line treatment for unresectable hepatocellular carcinoma (CheckMate 9DW): an open-label, randomised, phase 3 trial

Open access
Yau, Thomas, Galle, Peter R, Decaens, Thomas, Sangro, Bruno, Qin, Shukui, da Fonseca, Leonardo G, Karachiwala, Hatim, Blanc, Jean-Frederic, Park, Joong-Won, Gane, Edward, Pinter, Matthias, Peña, Ana Matilla, Ikeda, Masafumi, Tai, David, Santoro, Armando, Pizarro, Gonzalo, Chiu, Chang-Fang, Schenker, Michael, He, Aiwu, Chon, Hong Jae, Wojcik-Tomaszewska, Joanna, Verset, Gontran, Wang, Qi Qi, Stromko, Caitlyn, Neely, Jaclyn, Singh, Prianka, Jimenez Exposito, Maria Jesus, Kudo, Masatoshi · 2025 · The Lancet

The DOI matches its Crossref record; 27 of the 27 references it cites were found in a registry or an index; 21 of the 30 sources its claims cite were checked for support.

Abstract

BACKGROUND: Patients with unresectable hepatocellular carcinoma have a poor prognosis, and treatments with long-term benefits are needed. We report results from the preplanned interim analysis of the CheckMate 9DW trial assessing nivolumab plus ipilimumab versus lenvatinib or sorafenib for unresectable hepatocellular carcinoma in the first-line setting. METHODS: This open-label, randomised, phase 3 trial enrolled patients aged 18 years or older with unresectable hepatocellular carcinoma without previous systemic therapy at 163 hospitals and cancer centres across 25 countries in Asia, Australia, Europe, North America, and South America. Patients had at least one measurable untreated lesion per Response Evaluation Criteria in Solid Tumours (RECIST) version 1.1, a Child-Pugh score of 5 or 6, and an Eastern Cooperative Oncology Group performance status of 0 or 1. Patients were randomly assigned (1:1) via an interactive response technology system to receive nivolumab (1 mg/kg) plus ipilimumab (3 mg/kg) intravenously every 3 weeks for up to four doses, followed by nivolumab 480 mg every 4 weeks or investigator's choice of either oral lenvatinib (8 mg or 12 mg mg daily depending on bodyweight) or oral sorafenib (400 mg twice daily). Randomisation was stratified by aetiology; the presence of macrovascular invasion, extrahepatic spread, or both; and baseline alpha-fetoprotein concentration. The primary endpoint was overall survival, which was assessed in all randomly assigned patients; safety was an exploratory endpoint and was assessed in all randomly assigned patients who received at least one dose of study medication. This trial is registered with ClinicalTrials.gov, NCT04039607 (ongoing). FINDINGS: Between Jan 6, 2020, and Nov 8, 2021, 668 patients were randomly assigned to nivolumab plus ipilimumab (n=335) or lenvatinib or sorafenib (n=333). Early crossing of the overall survival Kaplan-Meier curves reflected a higher number of deaths during the first 6 months after randomisation with nivolumab plus ipilimumab (hazard ratio 1·65 [95% CI 1·12-2·43]) but was followed by a sustained separation of the curves thereafter in favour of nivolumab plus ipilimumab (0·61 [0·48-0·77]). After a median follow-up of 35·2 months (IQR 31·1-39·9), overall survival was significantly improved with nivolumab plus ipilimumab versus lenvatinib or sorafenib (median 23·7 months [95% CI 18·8-29·4] vs 20·6 months [17·5-22·5]; hazard ratio 0·79 [0·65-0·96]; two-sided stratified log-rank p=0·018); respective overall survival rates were 49% (95% CI 44-55) versus 39% (34-45) at 24 months and 38% (32-43) versus 24% (19-30) at 36 months. Overall, 137 (41%) of 332 patients receiving nivolumab plus ipilimumab and 138 (42%) of 325 patients receiving lenvatinib or sorafenib had grade 3-4 treatment-related adverse events. 12 deaths were attributed to treatment with nivolumab plus ipilimumab and three were attributed to treatment with lenvatinib or sorafenib. INTERPRETATION: Nivolumab plus ipilimumab showed a significant overall survival benefit versus lenvatinib or sorafenib and manageable safety in patients with previously untreated unresectable hepatocellular carcinoma. These results support nivolumab plus ipilimumab as a first-line treatment in this setting. FUNDING: Bristol Myers Squibb.

Overview

CoChat Summary

TL;DR

In CheckMate 9DW, first-line nivolumab plus ipilimumab significantly improved overall survival versus lenvatinib or sorafenib in unresectable hepatocellular carcinoma, with median overall survival 23.7 vs 20.6 months. The regimen also produced higher objective response rates and durable responses, with a manageable safety profile despite an early survival crossing.

Key findings

  • The primary endpoint was met: median overall survival was 23.7 months (95% CI 18.8–29.4) with nivolumab plus ipilimumab vs 20.6 months (17.5–22.5) with lenvatinib or sorafenib; HR 0.79 (95% CI 0.65–0.96), two-sided stratified log-rank p=0.018. Overall survival rates were 49% vs 39% at 24 months and 38% vs 24% at 36 months.
  • Early crossing of the overall survival curves reflected more deaths in the first 6 months with nivolumab plus ipilimumab (HR 1.65 [95% CI 1.12–2.43]), followed by sustained separation favouring nivolumab plus ipilimumab thereafter (HR 0.61 [0.48–0.77]). A competing risk analysis suggested this was substantially driven by an imbalance in deaths attributed to adverse events unrelated to study treatment.
  • Objective response rate by blinded independent central review was 36% (121/335; 95% CI 31–42) vs 13% (44/333; 10–17), p<0.0001; complete responses occurred in 7% vs 2%. Median duration of response was 30.4 months (21.2–not estimable) vs 12.9 months (10.2–31.2), and 47% of responders in the nivolumab plus ipilimumab group had an ongoing response at 36 months.
  • Grade 3–4 treatment-related adverse events occurred in 41% vs 42%; treatment-related deaths were reported in 12 patients vs 3 patients.

Methods

  • Open-label, randomised, phase 3 trial at 163 hospitals and cancer centres across 25 countries. Patients were adults aged ≥18 years with unresectable hepatocellular carcinoma without previous systemic therapy, at least one measurable untreated lesion per RECIST v1.1, Child-Pugh score 5 or 6, and ECOG performance status 0 or 1.
  • Patients were randomly assigned 1:1 via interactive response technology to nivolumab 1 mg/kg plus ipilimumab 3 mg/kg intravenously every 3 weeks for up to four doses followed by nivolumab 480 mg every 4 weeks, or investigator’s choice of oral lenvatinib (8 mg or 12 mg daily by bodyweight) or oral sorafenib 400 mg twice daily. Randomisation was stratified by aetiology, macrovascular invasion/extrahepatic spread, and baseline alpha-fetoprotein.
  • The primary endpoint was overall survival, assessed in all randomly assigned patients. Analysis used stratified log-rank tests and stratified Cox proportional hazards models; because the proportional hazards assumption was violated, supporting analyses included restricted mean survival time, piecewise HRs, and the MaxCombo test. Median follow-up was 35.2 months (IQR 31.1–39.9).

Limitations

  • The paper states that limitations include exclusion of patients with Vp4 portal vein thrombosis and the open-label design; adverse event causality and health-related quality of life might have been influenced by awareness of treatment received, although BICR mitigated bias for objective response rate and progression-free survival.
  • The text makes evident that the proportional hazards assumption was violated due to early crossing of the overall survival curves, so the treatment effect required interpretation through supporting analyses such as restricted mean survival time, piecewise HRs, and MaxCombo.
  • The comparator was investigator’s choice, and 85% of treated comparator patients received lenvatinib while 15% received sorafenib; separate comparisons against lenvatinib or sorafenib were post hoc.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1016/s0140-6736(25)00403-9

Cites

27 of 27 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Trial profile
Figure 1 Trial profileCC-BY-NC-ND · From the PDF
Image 2
CC-BY-NC-ND · From the PDF
Image 3
CC-BY-NC-ND · From the PDF

Verification

Claims the paper makes

8 Valid · 4 Not valid · 3 Not checked · 76 Own results and interpretations

29 more claims

Create a free account to see all 91 claims and to verify the 3 references still open.

See all claims — free account
  • It may state two separate claims as one. 42
  • Probably not a substantial claim. 23

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 27 references

Cited by

See all 390

Metadata

Authors
Yau, T., Galle, P. R., Decaens, T., Sangro, B., Qin, S., da Fonseca, L. G., Karachiwala, H., Blanc, J.-F., Park, J.-W., Gane, E., Pinter, M., Peña, A. M., Ikeda, M., Tai, D., Santoro, A., Pizarro, G., Chiu, C.-F., Schenker, M., He, A., Chon, H. J., Wojcik-Tomaszewska, J., Verset, G., Wang, Q. Q., Stromko, C., Neely, J., Singh, P., Jimenez Exposito, M. J., & Kudo, M.
Affiliations
  • University of Hong Kong (HK)
  • University Medical Center of the Johannes Gutenberg University Mainz (DE)
  • Centre National de la Recherche Scientifique (FR)
  • Clinica Universidad de Navarra (ES)
  • China Pharmaceutical University (CN)
  • Cross Cancer Institute (CA)
  • National Cancer Center (KR)
  • University of Auckland (NZ)
  • Medical University of Vienna (AT)
  • National Cancer Center Hospital East (JP)
  • National Cancer Centre Singapore (SG)
  • Humanitas University (IT)
  • China Medical University Hospital (TW)
  • MedStar Georgetown University Hospital (US)
  • CHA University Bundang Medical Center (KR)
  • Université Libre de Bruxelles (BE)
  • Bristol-Myers Squibb (United States) (US)
  • Kindai University (JP)
Corresponding author
Galle, Peter R
Venue
The Lancet
Publisher
Elsevier BV
Published
2025-05
Volume
405
Issue
10492
Pages
1851-1864
ISSN
  • 0099-5355 (ISSN-L)
  • 0140-6736
OpenAlex
W4410206247
Cited by
390
References
25
Citations by year
2026: 277 · 2025: 113
FWCI
216.23
Type
journal article
Language
en
Open access
Licence
Locations
  • The Lancet (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • Gutenberg Open Science (submittedVersion, cc-by-nc-nd)
  • The HKU Scholars Hub (University of Hong Kong) (submittedVersion)
  • Gutenberg Open Science
Funders
Bristol-Myers Squibb Company
Topics
  • Health Sciences › Medicine › Hepatology › Hepatocellular Carcinoma Treatment and Prognosis
  • Health Sciences › Medicine › Oncology › Cancer Immunotherapy and Biomarkers
  • Health Sciences › Medicine › Oncology › Colorectal Cancer Treatments and Studies
Keywords
hepatocellular carcinoma, unresectable liver cancer, nivolumab, ipilimumab, lenvatinib, sorafenib, immune checkpoint inhibitors, first-line therapy, overall survival, immune-related adverse events, alpha-fetoprotein, macrovascular invasion, extrahepatic spread, treatment safety
MeSH terms
Ipilimumab, Sorafenib, Nivolumab, Adult, Aged, Antineoplastic Combined Chemotherapy Protocols, Female, Carcinoma, Hepatocellular, Humans, Liver Neoplasms, Male, Middle Aged, Phenylurea Compounds, Quinolines, Treatment Outcome

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Nivolumab plus ipilimumab versus lenvatinib or sorafenib as first-line treatment for unresectable hepatocellular carcinoma (CheckMate 9DW): an open-label, randomised, phase 3 trial. https://app.cochat.ai/papers/10.1016/s0140-6736(25)00403-9
The paper
Yau, T., Galle, P. R., Decaens, T., Sangro, B., Qin, S., da Fonseca, L. G., Karachiwala, H., Blanc, J.-F., Park, J.-W., Gane, E., Pinter, M., Peña, A. M., Ikeda, M., Tai, D., Santoro, A., Pizarro, G., Chiu, C.-F., Schenker, M., He, A., . . . Kudo, M. (2025). Nivolumab plus ipilimumab versus lenvatinib or sorafenib as first-line treatment for unresectable hepatocellular carcinoma (CheckMate 9DW): an open-label, randomised, phase 3 trial. The Lancet. https://doi.org/10.1016/s0140-6736(25)00403-9

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.