Vision transformer-based model can optimize curative-intent treatment for patients with recurrent hepatocellular carcinoma
The treatment selection for recurrent hepatocellular carcinoma (rHCC) within Milan criteria after hepatectomy remains challenging. Here, we present HEROVision, a Vision Transformer-based model designed for personalized prognosis prediction and treatment optimization between thermal ablation (TA) and surgical resection (SR). HEROVision is trained on initial HCC cohorts (8492 images; 772 patients) and independently tested on rHCC cohorts (9163 images; 833 patients) from five centers. Propensity score matching (PSM) forms two groups of rHCC patients underwent TA and SR to fairly evaluate whether optimized treatment selection by HEROVision have clinical benefits. HEROVision significantly outperforms all six guideline staging systems in the external testing cohort, both in time-dependent concordance index and area under the curve (all P < 0.002). After PSM, 35.9% (23/64) and 6.6% (6/91) high-risk rHCC patients are identified, who could achieve improved prognosis by changing their treatments. HEROVision shows promise in optimizing individualized treatment between TA and SR for early-stage rHCC, complementing current clinical guidelines.
Authors
- Jinhua Pan (ORCID: https://orcid.org/0000-0002-6691-8282)
- Ping Liang (ORCID: https://orcid.org/0000-0001-8933-6392)
- Qiyu Zhao (ORCID: https://orcid.org/0000-0002-8732-8564)
- Tianan Jiang (ORCID: https://orcid.org/0000-0002-7672-8394)
- Hong Wu (ORCID: https://orcid.org/0000-0002-0885-4911)
- Liting Xie
- Zhifei Dai (ORCID: https://orcid.org/0000-0001-6231-3692)
- Jie Yu (ORCID: https://orcid.org/0000-0002-1168-8766)
- Fengwei Gao (ORCID: https://orcid.org/0000-0002-1385-0237)
- Wei Yang (ORCID: https://orcid.org/0000-0003-1190-8068)
- Qingyun Xie (ORCID: https://orcid.org/0000-0001-8664-0890)
- Jiapeng Wu (ORCID: https://orcid.org/0000-0002-0303-9686)
- Ke Zhang (ORCID: https://orcid.org/0000-0001-5606-428X)
- Haoyan Zhang (ORCID: https://orcid.org/0000-0003-3067-0710)
- Kun Wang (ORCID: https://orcid.org/0000-0003-2513-768X)
- Kai Li (ORCID: https://orcid.org/0000-0003-3982-494X)
- Jie Tian (ORCID: https://orcid.org/0000-0003-0498-0432)
- Lei Mu (ORCID: https://orcid.org/0000-0002-7922-2743)
- Qiuping Ma (ORCID: https://orcid.org/0000-0002-0000-8104)
- Jinyu Ru
- Wenbo Wang
Institutions
- Sun Yat-sen University (CN)
- Chinese Academy of Sciences (CN)
- Peking University (CN)
- Nankai University (CN)
- Sichuan University (CN)
- Chinese PLA General Hospital (CN)
- Peking University Cancer Hospital (CN)
- Shandong Institute of Automation (CN)
- Beijing Academy of Artificial Intelligence (CN)
- Third Affiliated Hospital of Sun Yat-sen University (CN)
- Ministry of Education (ET)
- University of Chinese Academy of Sciences (CN)
- First Affiliated Hospital Zhejiang University (CN)
- Beihang University (CN)
Publication Details
- Journal
- Nature Communications
- Published
- 2025-05-01
- DOI
- https://doi.org/10.1038/s41467-025-59197-0
- Citations
- 9
- Primary Topic
- Hepatocellular Carcinoma Treatment and Prognosis
- Type
- article
- Field-Weighted Citation Impact
- 11.90
Funders
- National Natural Science Foundation of China
- Chinese Academy of Sciences
- Sun Yat-sen University
- Zhejiang University
- Sichuan University
- Peking University
- General Hospital of People’s Liberation Army
- Institute of Automation, Chinese Academy of Sciences
- National Key Research and Development Program of China
- West China Hospital, Sichuan University