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. Conclusions Machine learning algorithms effectively provide risk stratification for DAA failure and additional information on the factors associated with DAA failure.

Objective To compare the differences in pathogenicity and gene expression profiles between adult Schistosoma japonicum isolated from hilly and marshland and lake regions of Anhui Province, so as to provide the scientific evidence for formulating the precise schistosomiasis control strategy in different endemic foci. Methods C57BL/6 mice were infected with cercariae of S. japonicum isolates from Shitai County (hilly regions) and Susong County (marshland and lake regions) of Anhui Province in 2021, and all mice were sacrificed 44 days post-infection and dissected. The worm burdens, number of S. japonicum eggs deposited in the liver, and the area of egg granulomas in the liver were measured to compare the difference in the pathogenicity between the two isolates. In addition, female and male adult S. japonicum worms were collected and subjected to transcriptome sequencing, and the gene expression profiles were compared between Shitai and Susong isolates of S. japonicum. The differentially expressed genes (DEGs) were subjected to Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. Results The total worm burdens [(14.50 ± 3.96) worms/mouse vs. (16.10 ± 3.78) worms/mouse; t = 0.877, P = 0.392], number of female and male paired worms [(4.50 ± 0.67) worms/mouse vs. (5.10 ± 1.45) worms/mouse; t = 1.129, P = 0.280], number of unpaired male worms [(5.50 ± 4.01) worms/mouse vs. (5.60 ± 1.69) worms/mouse; t = 0.069, P = 0.946], number of eggs deposited in per gram liver [(12 116.70 ± 6 508.83) eggs vs. (16 696.70 ± 4 571.56) eggs; t = 1.821, P = 0.085], and area of a single egg granuloma in the liver [(74 359.40 ± 11 766.34) µm² vs. (74 836.90 ± 13 086.12) µm²; t = 0.081, P = 0.936] were comparable between Shitai and Susong isolates of S. japonicum. Transcriptome sequencing identified 584 DEGs between adult female worms and 1 598 DEGs between adult male worms of Shitai and Susong isolates of S. japonicum. GO enrichment analysis showed that the DEGs between female adults were predominantly enriched in biological processes of stimulus response, cytotoxicity, multiple cell biological processes, metabolic processes, cellular processes and signaling pathways, cellular components of cell, organelles and cell membranes and molecular functions of binding and catalytic ability, and KEGG enrichment analysis showed that these DEGs were significantly enriched in pathways of vascular endothelial growth factor signaling, glutathione metabolism, arginine and proline metabolism. In addition, the DEGs between male adults were predominantly enriched in biological processes of signaling transduction, multiple cell biological processes, regulation of biological processes, metabolic processes, development processes and stimulus responses, cellular components of extracellular matrix and cell junction and molecular functions of binding and catalytic ability, and these DEGs were significantly enriched in pathways of Wnt signaling, Ras signaling, natural killer cells-mediated cytotoxicity, extracellular matrix-receptor interactions and arginine biosynthesis. Conclusions There is no significant difference in the pathogenicity between S. japonicum isolates from hilly and marshland and lake regions of Anhui Province; however, the gene expression profiles vary significantly between S. japonicum isolates.