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

Background & Aims: Accurate risk stratification for hepatocellular carcinoma (HCC) following the achievement of a sustained virologic 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 achievement of an SVR in individual patients. Methods: In this multicenter cohort study, 1,742 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's 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). An RSF model, based on seven parameters at the achievement of an SVR, showed the best discriminative ability, with a c-index of 0.839 in the external validation cohort and a high discriminative ability when patients were categorized into three risk groups (p <0.001). Furthermore, using an app that has been made available online, this RSF model (termed the SMART model) enabled the generation of an individualized predictive curve for HCC occurrence for each patient. Conclusions: We developed and externally validated an RSF model with good predictive performance for the risk of HCC after an SVR. This model can be used for risk stratification and, subject to validation and cost-effectiveness analysis, could be applied to personalized surveilance approaches in each country.(c) 2023 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.

Automated summary

A machine learning model has been developed to predict the risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis C after achieving a sustained virologic response (SVR). The model showed good discriminative ability in external validation and can generate individualized predictive curves for each patient using an online application.