Artificial intelligence predicts direct-acting antivirals failure among hepatitis C virus patients: A nationwide hepatitis C virus registry program.

BACKGROUND/AIMS: Despite the high efficacy of direct-acting antivirals (DAAs), approximately 1-3% of hepatitis C virus (HCV) patients fail to achieve a sustained virological response. We conducted a nationwide study to investigate risk factors associated with DAA treatment failure. Machine-learning algorithms have been applied to discriminate subjects who may fail to respond to DAA therapy. METHODS: We analyzed the Taiwan HCV Registry Program database to explore predictors of DAA failure in HCV patients. Fifty-five host and virological features were assessed using multivariate logistic regression, decision tree, random forest, eXtreme Gradient Boosting (XGBoost), and artificial neural network. The primary outcome was undetectable HCV RNA at 12 weeks after the end of treatment. RESULTS: The training (n=23,955) and validation (n=10,346) datasets had similar baseline demographics, with an overall DAA failure rate of 1.6% (n=538). Multivariate logistic regression analysis revealed that liver cirrhosis, hepatocellular carcinoma, poor DAA adherence, and higher hemoglobin A1c were significantly associated with virological failure. XGBoost outperformed the other algorithms and logistic regression models, with an area under the receiver operating characteristic curve of 1.000 in the training dataset and 0.803 in the validation dataset. The top five predictors of treatment failure were HCV RNA, body mass index, α-fetoprotein, platelets, and FIB-4 index. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the XGBoost model (cutoff value=0.5) were 99.5%, 69.7%, 99.9%, 97.4%, and 99.5%, respectively, for the entire dataset. CONCLUSION: Machine learning algorithms effectively provide risk stratification for DAA failure and additional information on the factors associated with DAA failure.

An algorithm for simplified hepatitis C virus treatment with non-specialist care based on nation-wide data from Taiwan.

BACKGROUND: Both European Association for the Study of the Liver (EASL) and American Association for the Study of Liver Diseases and the Infectious Diseases Society of America (AASLD-IDSA) guidelines recommend simplified hepatitis C virus (HCV) treatment with pan-genotypic sofosbuvir/velpatasvir or glecaprevir/pibrentasvir for eligible patients. This observational study used real-world data to assess these regimens' safety in eligible patients and develop an algorithm to identify patients suitable for simplified treatment by non-specialists. METHODS: 7,677 HCV-infected patients from Taiwan Hepatitis C Registry (TACR) who received at least one dose of sofosbuvir/velpatasvir or glecaprevir/pibrentasvir, and fulfilled the EASL/AASLD-IDSA criteria for simplified treatment were analyzed. Multivariate analysis was conducted on patient characteristics and safety data. RESULTS: Overall, 92.8% (7,128/7,677) of patients achieved sustained virological response and only 1.9% (146/7,677) experienced Grades 2-4 laboratory abnormalities in key liver function parameters (alanine aminotransferase, aspartate aminotransferase, and total bilirubin), with only 18 patients (0.23%) experiencing Grades 3-4 abnormalities. Age > 70 years old, presence of hepatocellular carcinoma, total bilirubin > 1.2 mg/dL, estimated glomerular filtration rate < 60 mL/min/1.73 m2, and Fibrosis-4 > 3.25 were associated with higher risks of Grades 2-4 abnormalities. Patients with any of these had an odds of 4.53 times than that of those without in developing Grades 2-4 abnormalities (p < 0.01). CONCLUSIONS: Real-world data from Taiwan confirmed that simplified HCV treatment for eligible patients with pan-genotypic regimens is effective and well tolerated. The TACR algorithm, developed based on this study's results, can further identify patients who can be safely managed by non-specialist care.