DOI: 10.1001/jamaoncol.2026.3901 ISSN: 2374-2437

Immune Checkpoint Inhibitor–Based Downstaging Therapy for Hepatocellular Carcinoma

Meiching Ong, Kang He, Feng Wang, Ziqiang Li, Yipeng Pan, Wenlei Li, Jun Fang, Aibo Mu, Zhaoxian Li, Siyi Zhong, Zebin Zhu, Songming Li, Tielong Wang, Xuanyi Zhu, Shugeng Zhang, Zhiyong Guo, Jinzhen Cai, Jian Xu, Qiang Xia, Shusen Zheng, Guangming Li, Qi Ling

Importance

In patients with hepatocellular carcinoma (HCC) that is initially beyond the Milan criteria, immune checkpoint inhibitor (ICI)–based regimens have been explored before liver transplant (LT) to improve tumor control as part of downstaging strategies that most often involve locoregional therapy, but associations with posttransplant oncologic and survival benefit remain uncertain.

Objective

To evaluate the associations of an ICI-based multimodal pretransplant strategy with posttransplant oncologic outcomes and acute rejection among patients with HCC that was initially beyond the Milan criteria who underwent LT.

Design, Setting, and Participants

This multicenter, retrospective, matched cohort study included adults (≥18 years) with HCC that was initially beyond the Milan criteria who underwent LT at 8 high-volume LT centers (>55 LTs annually) in China between January 2019 and December 2024. Propensity score matching balanced measured covariates between ICI and non-ICI groups. The data were analyzed in February 2026.

Exposures

ICI-based multimodal pretransplant strategy (typically combined with locoregional and/or targeted therapies) vs therapy without ICIs.

Main Outcomes and Measures

Primary outcomes were overall survival (OS) and recurrence-free survival (RFS) from LT. Secondary outcomes were posttransplant recurrence and acute rejection within 90 days.

Results

Among 416 matched recipients, 208 (50.5%) received ICI-based multimodal therapy and 208 (50.0%) did not. The median age was 53 years (IQR, 46-59 years), 39 recipients (9.4%) were female, and 377 (90.6%) were male. Median OS was 4.57 years (95% CI, 4.34 to not estimable) in the ICI group and 2.53 years (95% CI, 2.17-3.10; P  < .001) in the non-ICI group. Median RFS was 1.47 years (95% CI, 1.18-4.34) in the ICI group and 0.89 years (95% CI, 0.75-1.39; P  = .02) in the non-ICI group. In multivariable Cox models, the ICI-based multimodal strategy was associated with better OS (hazard ratio [HR], 0.63; 95% CI, 0.43-0.92) and RFS (HR, 0.59; 95% CI, 0.44-0.80). In a Fine-Gray model, the ICI-based multimodal strategy was associated with lower posttransplant recurrence risk (subdistribution HR, 0.54; 95% CI, 0.40-0.72). Acute rejection within 90 days after transplant was more frequent in the ICI group (34 [16.3%] vs 17 [8.2%]; P  = .01).

Conclusions and Relevance

The results of this cohort study suggest that an ICI-based multimodal pretransplant strategy was associated with better posttransplant oncologic outcomes and survival in selected patients with HCC that was initially beyond the Milan criteria, despite more frequent acute rejection.