{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T18:15:07Z","timestamp":1781720107990,"version":"3.54.5"},"reference-count":17,"publisher":"Wiley","issue":"6","license":[{"start":{"date-parts":[[2026,2,6]],"date-time":"2026-02-06T00:00:00Z","timestamp":1770336000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,2,6]],"date-time":"2026-02-06T00:00:00Z","timestamp":1770336000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["ascpt.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Clin Pharma and Therapeutics"],"published-print":{"date-parts":[[2026,6]]},"abstract":"<jats:p>\n                    Drug\u2010induced QT\u2010interval prolongation, a non\u2010specific biomarker of increased risk for Torsades de Pointes (TdP), is a major safety concern in drug development. While\n                    <jats:italic>in vitro<\/jats:italic>\n                    hERG inhibition assays are required for early\u2010phase screening, pharmacovigilance data from sources like the FDA Adverse Event Reporting System (FAERS) provide complementary insights. Integrating these heterogeneous data with molecular structure offers a promising, yet underutilized approach to predict proarrhythmic risk. We developed an interpretable graph neural network (GNN) framework integrating\n                    <jats:italic>in vitro<\/jats:italic>\n                    hERG inhibition data (PubChem AID 588834), FAERS\u2010derived pharmacovigilance signals, and molecular structure information. Canonical SMILES were converted into molecular graphs using RDKit, and atom\u2010 and bond\u2010level features were encoded. Four GNN architectures (GINE, GCN, GraphSAGE, and GATv2) were compared via stratified five\u2010fold cross\u2010validation. The best\u2010performing model, GATv2, was further interpreted using Integrated Gradients to identify structural features contributing to QT liability. The final data set comprised 4,808 small molecules with binary QT\u2010risk labels. GATv2 achieved a cross\u2010validated ROC\u2010AUC of 0.838, PR\u2010AUC of 0.830, and F1\u2010score of 0.756. Retraining on the full data set yielded ROC\u2010AUC 0.918, PR\u2010AUC 0.908, and F1\u2010score 0.847. External validation on an independent hERG assay (AID 1671200,\n                    <jats:italic>n<\/jats:italic>\n                    \u2009=\u20092,405) confirmed strong performance (ROC\u2010AUC 0.859, sensitivity 0.80, specificity 0.82). Atomic degree and hydrogen count were dominant predictors, consistent with known SARs. This GNN\u2010based framework integrates structural and pharmacological data to predict QT risk, providing a transparent, structure\u2010based decision\u2010support tool aligned with ICH S7B\/E14 and CiPA guidelines.\n                  <\/jats:p>","DOI":"10.1002\/cpt.70220","type":"journal-article","created":{"date-parts":[[2026,2,7]],"date-time":"2026-02-07T06:01:22Z","timestamp":1770444082000},"page":"1662-1670","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Structure\u2010Based Prediction of\n                    <scp>QT<\/scp>\n                    Prolongation Risk Using Graph Neural Networks: An Integrative Approach Combining\n                    <i>In Vitro<\/i>\n                    <scp>hERG<\/scp>\n                    Assays and Pharmacovigilance Data"],"prefix":"10.1002","volume":"119","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0977-1181","authenticated-orcid":false,"given":"Tomoyuki","family":"Enokiya","sequence":"first","affiliation":[{"name":"Laboratory of Pharmacoinformatics, Department of Pharmaceutical Sciences, Faculty of Pharmaceutical Sciences Suzuka University of Medical Science  Suzuka Japan"},{"name":"Laboratory of Pharmacoinformatics, Division of Clinical Pharmacy and Pharmaceutical Sciences, Graduate School of Pharmaceutical Sciences Suzuka University of Medical Science  Suzuka Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ryosuke","family":"Kunitomo","sequence":"additional","affiliation":[{"name":"Laboratory of Pharmacoinformatics, Department of Pharmaceutical Sciences, Faculty of Pharmaceutical Sciences Suzuka University of Medical Science  Suzuka Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Takamasa","family":"Yamaguchi","sequence":"additional","affiliation":[{"name":"Laboratory of Pharmacoinformatics, Division of Clinical Pharmacy and Pharmaceutical Sciences, Graduate School of Pharmaceutical Sciences Suzuka University of Medical Science  Suzuka Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,2,6]]},"reference":[{"key":"e_1_2_11_1_1","doi-asserted-by":"publisher","DOI":"10.1113\/JP270526"},{"key":"e_1_2_11_2_1","doi-asserted-by":"publisher","DOI":"10.1093\/europace\/eut214"},{"key":"e_1_2_11_3_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1476-5381.2009.00487.x"},{"key":"e_1_2_11_4_1","doi-asserted-by":"publisher","DOI":"10.3389\/fdsfr.2022.866898"},{"key":"e_1_2_11_5_1","doi-asserted-by":"publisher","DOI":"10.3389\/fphar.2023.1259611"},{"key":"e_1_2_11_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.clinthera.2018.07.012"},{"key":"e_1_2_11_7_1","doi-asserted-by":"publisher","DOI":"10.1177\/20420986221125006"},{"key":"e_1_2_11_8_1","doi-asserted-by":"publisher","DOI":"10.3389\/fphar.2023.1117391"},{"key":"e_1_2_11_9_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btaa075"},{"key":"e_1_2_11_10_1","first-page":"1561","article-title":"Equivariant graph neural networks for toxicity prediction","volume":"36","author":"Cremer J.","year":"2023","journal-title":"Chem. Res. Toxicol."},{"key":"e_1_2_11_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2022.106464"},{"key":"e_1_2_11_12_1","first-page":"143","article-title":"AttenhERG: a reliable and interpretable graph neural network framework for predicting hERG channel blockers","volume":"16","author":"Yang T.","year":"2024","journal-title":"J. Chem."},{"key":"e_1_2_11_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.crtox.2023.100121"},{"key":"e_1_2_11_14_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.est.3c00653"},{"key":"e_1_2_11_15_1","doi-asserted-by":"publisher","DOI":"10.1021\/jm0208875"},{"key":"e_1_2_11_16_1","first-page":"1561","article-title":"Equivariant graph neural networks for toxicity prediction","volume":"36","author":"Cremer J.","year":"2023","journal-title":"Chem Res Toxicol."},{"key":"e_1_2_11_17_1","first-page":"1067","article-title":"Mold2, molecular descriptors from 2D structures for chemoinformatics and toxicoinformatics","volume":"55","author":"Hong H.","year":"2015","journal-title":"J. Chem. Inf. Model."}],"container-title":["Clinical Pharmacology &amp; Therapeutics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/ascpt.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/cpt.70220","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/ascpt.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/cpt.70220","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/ascpt.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/cpt.70220","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,9]],"date-time":"2026-05-09T03:29:25Z","timestamp":1778297365000},"score":1,"resource":{"primary":{"URL":"https:\/\/ascpt.onlinelibrary.wiley.com\/doi\/10.1002\/cpt.70220"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,6]]},"references-count":17,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2026,6]]}},"alternative-id":["10.1002\/cpt.70220"],"URL":"https:\/\/doi.org\/10.1002\/cpt.70220","archive":["Portico"],"relation":{},"ISSN":["0009-9236","1532-6535"],"issn-type":[{"value":"0009-9236","type":"print"},{"value":"1532-6535","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,6]]},"assertion":[{"value":"2025-07-07","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-01-16","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-02-06","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}