Machine learning for individualized prediction of hepatocellular carcinoma development after the eradication of hepatitis C virus with antivirals

Tatsuya Minami; Masaya Sato; Hidenori Toyoda; Satoshi Yasuda; Tomoharu Yamada; Takuma Nakatsuka; Kenichiro Enooku; Hayato Nakagawa; Hidetaka Fujinaga; Masashi Izumiya; Yasuo Tanaka; Motoyuki Otsuka; Takamasa Ohki; Masahiro Arai; Yoshinari Asaoka; Atsushi Tanaka; Kiyomi Yasuda; Hideaki Miura; Itsuro Ogata; Toshiro Kamoshida; Kazuaki Inoue; Ryo Nakagomi; Masatoshi Akamatsu; Hiroshi Mitsui; Hajime Fujie; Keiji Ogura; Koji Uchino; Hideo Yoshida; Kazuyuki Hanajiri; Tomonori Wada; Kiyohiko Kurai; Hisato Maekawa; Yuji Kondo; Shuntaro Obi; Takuma Teratani; Naohiko Masaki; Kayo Nagashima; Takashi Ishikawa; Naoya Kato; Hiroshi Yotsuyanagi; Kyoji Moriya; Takashi Kumada; Mitsuhiro Fujishiro; Kazuhiko Koike; Ryosuke Tateishi

doi:10.1016/j.jhep.2023.05.042

Machine learning for individualized prediction of hepatocellular carcinoma development after the eradication of hepatitis C virus with antivirals

J Hepatol. 2023 Jun 24:S0168-8278(23)00424-5. doi: 10.1016/j.jhep.2023.05.042. Online ahead of print.

Authors

Tatsuya Minami¹, Masaya Sato¹, Hidenori Toyoda², Satoshi Yasuda², Tomoharu Yamada¹, Takuma Nakatsuka¹, Kenichiro Enooku¹, Hayato Nakagawa¹, Hidetaka Fujinaga¹, Masashi Izumiya¹, Yasuo Tanaka¹, Motoyuki Otsuka¹, Takamasa Ohki³, Masahiro Arai⁴, Yoshinari Asaoka⁵, Atsushi Tanaka⁵, Kiyomi Yasuda⁶, Hideaki Miura⁷, Itsuro Ogata⁸, Toshiro Kamoshida⁹, Kazuaki Inoue¹⁰, Ryo Nakagomi¹¹, Masatoshi Akamatsu¹², Hiroshi Mitsui¹³, Hajime Fujie¹⁴, Keiji Ogura¹⁵, Koji Uchino¹⁶, Hideo Yoshida¹⁶, Kazuyuki Hanajiri¹⁷, Tomonori Wada¹⁷, Kiyohiko Kurai¹⁸, Hisato Maekawa¹⁹, Yuji Kondo²⁰, Shuntaro Obi²⁰, Takuma Teratani²¹, Naohiko Masaki²², Kayo Nagashima²³, Takashi Ishikawa²⁴, Naoya Kato¹, Hiroshi Yotsuyanagi²⁵, Kyoji Moriya¹, Takashi Kumada², Mitsuhiro Fujishiro¹, Kazuhiko Koike¹, Ryosuke Tateishi²⁶

Affiliations

¹ Department of Gastroenterology, Graduate School of Medicine, The University of Tokyo.
² Department of Gastroenterology and Hepatology, Ogaki Municipal Hospital.
³ Department of Gastroenterology, Mitsui Memorial Hospital.
⁴ Department of Gastroenterology, Toshiba General Hospital.
⁵ Department of Medicine, Teikyo University School of Medicine.
⁶ Department of Gastroenterology, Kiyokawa Hospital.
⁷ Department of Gastroenterology, Tokyo Yamate Medical Center.
⁸ Department of Gastroenterology, Kawakita General Hospital.
⁹ Department of Gastroenterology, Hitachi General Hospital.
¹⁰ Department of Gastroenterology, Showa University Fujigaoka Hospital.
¹¹ Department of Gastroenterology, Kanto Central Hospital of the Mutual Aid Association of Public School Teacher.
¹² Department of Gastroenterology, JR Tokyo General Hospital.
¹³ Department of Gastroenterology, Tokyo Teishin Hospital.
¹⁴ Department of Gastroenterology, Tokyo Shinjuku Medical Center.
¹⁵ Department of Gastroenterology, Tokyo Metropolitan Police Hospital.
¹⁶ Department of Gastroenterology, Japanese Red Cross Medical Center.
¹⁷ Department of Gastroenterology, Sanraku Hospital.
¹⁸ Kurai Kiyohiko Medical Clinic.
¹⁹ Department of Gastroenterology and Hepatology, Tokyo Takanawa Hospital.
²⁰ Department of Gastroenterology and Hepatology, Kyoundo Hospital.
²¹ Department of Hepato-Bililary-Pancreatic Medicine, NTT Medical Center Tokyo.
²² Clinical Laboratory Department, Center Hospital of the National Center for Global Health and Medicine.
²³ Department of Gastroenterology, National Disaster Medical Center.
²⁴ Marunouchi Clinic.
²⁵ Division of Infectious Disease and Applied Immunology, The University of Tokyo the Institute of Medical Science Research Hospital.
²⁶ Department of Gastroenterology, Graduate School of Medicine, The University of Tokyo. Electronic address: tateishi-tky@umin.ac.jp.

PMID: 37716372
DOI: 10.1016/j.jhep.2023.05.042

Abstract

Background and aims: Accurate risk stratification for hepatocellular carcinoma (HCC) after achieving a sustained viral response (SVR) is necessary for optimal surveillance. We aimed to develop and validate a machine learning (ML) model to predict the risk of HCC after achieving an SVR in individual patients.

Methods: In this multicenter cohort study, 1742 patients with chronic hepatitis C who achieved an SVR were enrolled. Five ML models were developed including DeepSurv, gradient boosting survival analysis, random survival forest (RSF), survival support vector machine, and a conventional Cox proportional hazard model. Model performance was evaluated using Harrel' c-index and was externally validated in an independent cohort (977 patients).

Results: During the mean observation period of 5.4 years, 122 patients developed HCC (83 in the derivation cohort and 39 in the external validation cohort). The RSF model showed the best discrimination ability using seven parameters at the achievement of an SVR with a c-index of 0.839 in the external validation cohort and a high discriminative ability when the patients were categorized into three risk groups (P <0.001). Furthermore, this RSF model enabled the generation of an individualized predictive curve for HCC occurrence for each patient with an app available online.

Conclusions: We developed and externally validated an RSF model with good predictive performance for the risk of HCC after an SVR. The application of this novel model is available on the website. This model could provide the data to consider an effective surveillance method. Further studies are needed to make recommendations for surveillance policies tailored to the medical situation in each country.

Impact and implications: A novel prediction model for HCC occurrence in patients after hepatitis C virus eradication was developed using machine learning algorithms. This model, using seven commonly measured parameters, has been shown to have a good predictive ability for HCC development and could provide a personalized surveillance system.

Keywords: Hepatitis C; Hepatocellular carcinoma; Machine learning; SVR; Surveillance.