[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-457065-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-457065-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","heterogeneous-graph-neural-network-based-prediction-of-immune-related-adverse-events","Heterogeneous graph neural network-based prediction of immune-related adverse events","","Immune-related adverse events (irAEs) are common, potentially fatal complications, yet predicting them from clinical medication regimens and basic patient information remains difficult. This study builds an irAE prediction model using heterogeneous graph neural networks with electronic health records (EHRs). Medications, diagnoses, and patient characteristics are represented as nodes, and medication-policy simulation, case studies, and interpretability analyses are used to match clinical practice. The method achieves AUC 0.902 and strong performance across evaluation metrics, with robust reliability and improved personalization monitoring.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/healthcare/","Healthcare",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/heterogeneous-graph-neural-network-based-prediction-of-immune-related-adverse-events/457065/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/heterogeneous-graph-neural-network-based-prediction-of-immune-related-adverse-events/457065.png","ImageObject",300,407,{"name":42,"@type":43},"\tCallum ","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-02","2026-09-30",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",10,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why are immune-related adverse events (irAEs) difficult to predict clinically?","Question",{"text":63,"@type":64},"irAEs are diverse and can affect multiple organs, and patients are often treated with immune checkpoint inhibitors in combination with other anticancer drugs. Predicting risk from real-world regimens and basic patient data is therefore challenging.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How does the proposed model use electronic health records for prediction?",{"text":68,"@type":64},"It constructs a heterogeneous graph where nodes represent medications, diagnoses, and patient characteristics extracted from EHRs. The model predicts whether irAEs will occur based on the learned representations.",{"name":70,"@type":61,"acceptedAnswer":71},"What evidence shows the method outperforms other baselines?",{"text":72,"@type":64},"The approach reports superior results across evaluation metrics, including AUC 0.902, precision 0.709, recall 0.799, F1 0.751, and accuracy 0.851, with additional simulation-driven improvements.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},457065,1790790412,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,110,115,120,123,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":108,"slug":109},7,40,"healthcare",{"id":111,"doc_module":4,"doc_module_name":25,"category_name":112,"show_sort_weight":113,"slug":114},8,"Research & Report",30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":55,"slug":125},"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":107,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":145,"preview_page_count":4},137451211410,"https://ap-avatar.wpscdn.com/avatar/2000bb0a9246f588df?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786362646172706240","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nHeterogeneous graph neural network-based prediction of immune-related adverse events  \nXiaojun He1,3, Qiao Ni1,3, Cui Chen1, Hongmei Li1, Linghao Ni1, Jiawei Zhou1􀀍, Lan Tang2􀀍 & Bin Peng1􀀍  \nImmune-related adverse events (irAEs) are common and potentially fatal adverse events. However, predicting irAEs based on clinical medication regimens and basic patient information remains a significant clinical challenge. This study aims to develop a prediction model using graph neural networks with electronic health records (EHRs), thereby reducing irAEs risk. Our method based on heterogeneous graph networks. It incorporates medications, diagnoses and patients characteristics from EHRs as nodes to predict irAEs occurrence. Medication-policy simulation, case studies and interpretability analyses were conducted to align the model with real-world clinical needs. Compared to other baseline methods, our method shows superior performance across all evaluation metrics: with AUC of 0.902, AUPRC of 0.85, precision of 0.709, RECALL of 0.799, F1 score of 0.751, accuracy of 0.851. About simulation study, the model demonstrated progressive improvement, reflected in a 5%– 6% increase across six evaluation metrics. Interpretability analysis revealed that distinct risk patterns emerge at different treatment stages. Our approach exhibits robust reliability and outperforms other methods for irAEs prediction. Our study further establishes a novel paradigm for personalized therapy monitoring and early intervention. This methodology holds potential for reducing irAEs risk.  \nKeywords Immune-related adverse events, Graph neural network, Precision medicine  \nThe emergence of immune checkpoint inhibitors (ICIs) has brought positive news to the treatment of cancer patients, significantly improving the clinical therapeutic effects of various cancers1,2. At present, ICIs include three types: PD-1, PD-L1, and CTLA-4 . They have been approved in many countries for the treatment of various cancers, yielding substantial benefits to cancer patients3. Similar to other cancer treatment regimens, ICI treatment also has its limitations, namely the occurrence of immune-related adverse events (irAEs) . These adverse events are diverse and vary in severity. irAEs usually triggers responses in multiple organs, such as lungs, skin, liver, heart, endocrine system, etc4–7. Mild irAEs may not affect the continuation of ICI treatment. However, the occurrence of severe irAEs may result in treatment interruption and pose health risks, and even lead to the death of patients8.  \nWith the increasing use of ICIs, current research on predicting irAEs primarily relies on collecting and integrating clinical data with conventional machine learning methods. These studies have identified factors such as age, sex, chronic disease history, and concomitant medications as potential predictors of irAEs9–12. However, in real-world clinical practice, ICIs are frequently administered in combination with other anticancer agents like chemotherapeutic and targeted drugs2, 13, presenting significant challenges for irAEs prediction that aligns with actual clinical practice. Furthermore, research on adverse events predominantly relies on spontaneously reported data, chemical structures, and drug molecular graphs. Most studies focus on the drug level—such as drug interactions and drug-induced suspected adverse event14, 15 with limited exploration based on clinical databases. We aim to integrate real-world clinical contexts as comprehensively as possible to develop potential methods for predicting irAEs.  \nAt present, the electronic health records (EHRs) system is currently experiencing rapid growth. Real-world clinical data within EHRs hold significant value for pharmacovigilance, disease prevention/control, and related domains. EHRs encompass extensive, heterogeneous, and longitudinal medical information documenting patient hea","cbCailwRSpiIXMlY","https://ap.wps.com/l/cbCailwRSpiIXMlY","pdf",3403718,13,"English","# Introduction\n## Immune checkpoint inhibitors and irAEs\n## Current prediction research and limitations\n## Value of electronic health records (EHRs)\n## Graph neural networks and heterogeneous patient graphs\n## Motivation for the proposed approach","[{\"question\":\"Why are immune-related adverse events (irAEs) difficult to predict clinically?\",\"answer\":\"irAEs are diverse and can affect multiple organs, and patients are often treated with immune checkpoint inhibitors in combination with other anticancer drugs. Predicting risk from real-world regimens and basic patient data is therefore challenging.\"},{\"question\":\"How does the proposed model use electronic health records for prediction?\",\"answer\":\"It constructs a heterogeneous graph where nodes represent medications, diagnoses, and patient characteristics extracted from EHRs. The model predicts whether irAEs will occur based on the learned representations.\"},{\"question\":\"What evidence shows the method outperforms other baselines?\",\"answer\":\"The approach reports superior results across evaluation metrics, including AUC 0.902, precision 0.709, recall 0.799, F1 0.751, and accuracy 0.851, with additional simulation-driven improvements.\"}]","Heterogeneous graph neural network-based prediction of immune-related adverse events | PDF",1790748220,33]