Sim2Real-PK framework for precision dosing: Integrating pharmacokinetic mechanisms and deep learning to decode clinical heterogeneity from sparse data

Sim2Real-PK framework for precision dosing: Integrating pharmacokinetic mechanisms and deep learning to decode clinical heterogeneity from sparse data | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Sim2Real-PK framework for precision dosing: Integrating pharmacokinetic mechanisms and deep learning to decode clinical heterogeneity from sparse data Lu-Yao Han, Xiang Chen, Hui-Ze Wang, Zhi-Long Zhang, Tian-Shuo Liu, and 10 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9308510/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Precision dosing is a cornerstone of individualized therapeutics, yet its implementation is frequently hindered by clinical pharmacokinetic data that are characteristically sparse, irregularly sampled, and highly heterogeneous. Traditional mechanistic models provide structural interpretability but often lack the flexibility to capture complex patient-specific deviations, whereas purely data-driven deep learning models are prone to pharmacological inconsistency and poor generalization in low-data regimes. Here we present Sim2Real-PK, a hybrid neural-mechanistic framework that reconceptualizes pharmacokinetic forecasting as a time-aware latent state-space trajectory inference task. Sim2Real-PK first pretrains a neural dynamics core on massive synthetic datasets derived from mechanistic ordinary differential equation systems to internalize universal kinetic laws, then adapts to real-world clinical heterogeneity through a constrained residual correction pathway. Validated on individual participant data from the YODA Project for two drugs representing distinct pharmacological regimes—ibrutinib and daratumumab—Sim2Real-PK significantly outperformed conventional population-level baselines. The framework improved the F50 for ibrutinib from 32.1% to 39.9% and the F20 for daratumumab from 38.8% to 51.1%, while faithfully recovering expected linear and nonlinear pharmacological behaviors. Crucially, the model successfully resolved latent pharmacokinetic heterogeneity within patient cohorts that remained obscured under traditional analytical frameworks. These results demonstrate that mechanistically-anchored neural adaptation can bridge the gap between biological rigor and predictive agility, providing a robust foundation for next-generation model-informed precision dosing. Biological sciences/Computational biology and bioinformatics Biological sciences/Drug discovery Full Text Additional Declarations No competing interests reported. Supplementary Files supplementarymaterialshly.pdf Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9308510","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":624811584,"identity":"d68e37f8-e436-4a33-b6d9-95a5bcb7b6cd","order_by":0,"name":"Lu-Yao Han","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Lu-Yao","middleName":"","lastName":"Han","suffix":""},{"id":624811585,"identity":"b1f5fb31-2bef-4178-858d-723c66f307f0","order_by":1,"name":"Xiang Chen","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Xiang","middleName":"","lastName":"Chen","suffix":""},{"id":624811586,"identity":"592ee02e-f991-4e49-9bdd-6eaf929a9aa4","order_by":2,"name":"Hui-Ze Wang","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Hui-Ze","middleName":"","lastName":"Wang","suffix":""},{"id":624811587,"identity":"0b1e27c4-7a1f-4331-aea1-e89f93cbba39","order_by":3,"name":"Zhi-Long Zhang","email":"","orcid":"","institution":"Nanjing University","correspondingAuthor":false,"prefix":"","firstName":"Zhi-Long","middleName":"","lastName":"Zhang","suffix":""},{"id":624811588,"identity":"be9ed4cd-19f7-4772-9a76-a50144a87695","order_by":4,"name":"Tian-Shuo Liu","email":"","orcid":"","institution":"Nanjing University","correspondingAuthor":false,"prefix":"","firstName":"Tian-Shuo","middleName":"","lastName":"Liu","suffix":""},{"id":624811589,"identity":"f4000196-5653-4687-bc1b-76e592a0bf09","order_by":5,"name":"Meng-Han Wu","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Meng-Han","middleName":"","lastName":"Wu","suffix":""},{"id":624811590,"identity":"6b2f33de-15f5-48ce-a241-a63e5d0eaa34","order_by":6,"name":"Zi-Qi Wang","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Zi-Qi","middleName":"","lastName":"Wang","suffix":""},{"id":624811591,"identity":"33052bdd-5fc2-48b0-8068-9777b760570f","order_by":7,"name":"Wen-Bo Pang","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Wen-Bo","middleName":"","lastName":"Pang","suffix":""},{"id":624811592,"identity":"8bedea71-e227-49e0-a932-f70206d089bc","order_by":8,"name":"Jia-Qian Li","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Jia-Qian","middleName":"","lastName":"Li","suffix":""},{"id":624811594,"identity":"a324f664-fef2-48d3-8f2b-6b52c1960030","order_by":9,"name":"Jin-Yi Wang","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Jin-Yi","middleName":"","lastName":"Wang","suffix":""},{"id":624811595,"identity":"421b3e4b-f4c6-406a-a39c-fc13c0181921","order_by":10,"name":"De-Xuan Kong","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"De-Xuan","middleName":"","lastName":"Kong","suffix":""},{"id":624811597,"identity":"7f2e3051-bb23-42af-9fd2-9844b85ef006","order_by":11,"name":"Man-Qing Jia","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Man-Qing","middleName":"","lastName":"Jia","suffix":""},{"id":624811599,"identity":"07897ef0-89f1-4bfb-bd53-3e0d2dfc53c6","order_by":12,"name":"Haiping Hao","email":"","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Haiping","middleName":"","lastName":"Hao","suffix":""},{"id":624811600,"identity":"34b6623b-0db7-4191-bc7c-91821897d068","order_by":13,"name":"Yang Yu","email":"","orcid":"","institution":"Nanjing University","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Yu","suffix":""},{"id":624811601,"identity":"451258f2-ff55-475c-9da1-f49f68a61e87","order_by":14,"name":"Guo Yu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAs0lEQVRIiWNgGAWjYNCCCgglQYKWMyRrYWwjRYt8RPIxad55dnkGB5gP3uZhsMsjqMXwRlqaNO+25GKDA2zJ1jwMycWEtczIMQNqOZC44QCPmTQPw4HEBuK0zAFp4f9GnBZ5CZCWBrAtbMRpMeB5lmw551hyseRhNmPLOQbJRNjSnnzwxpsauzy+480Pb7ypsCPClgMMLKDoSGBgBnMJqQfZ0sDA/AGsZRSMglEwCkYBLgAAkyc3+JmrAdkAAAAASUVORK5CYII=","orcid":"","institution":"China Pharmaceutical University","correspondingAuthor":true,"prefix":"","firstName":"Guo","middleName":"","lastName":"Yu","suffix":""}],"badges":[],"createdAt":"2026-04-03 03:54:54","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9308510/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9308510/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108806133,"identity":"8107204d-937c-416e-af51-b1550ea76ff6","added_by":"auto","created_at":"2026-05-08 15:27:46","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":932541,"visible":true,"origin":"","legend":"","description":"","filename":"maintext0320final.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9308510/v1_covered_7897f1f5-fae0-488b-9b6f-1a4fb1106984.pdf"},{"id":107693323,"identity":"f8dab0bf-1bdf-4f22-a277-decff0d9e908","added_by":"auto","created_at":"2026-04-24 06:33:25","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1709899,"visible":true,"origin":"","legend":"","description":"","filename":"supplementarymaterialshly.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9308510/v1/59eee8527ba734361fd4833d.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Sim2Real-PK framework for precision dosing: Integrating pharmacokinetic mechanisms and deep learning to decode clinical heterogeneity from sparse data","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9308510/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9308510/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003ePrecision dosing is a cornerstone of individualized therapeutics, yet its implementation is frequently hindered by clinical pharmacokinetic data that are characteristically sparse, irregularly sampled, and highly heterogeneous. Traditional mechanistic models provide structural interpretability but often lack the flexibility to capture complex patient-specific deviations, whereas purely data-driven deep learning models are prone to pharmacological inconsistency and poor generalization in low-data regimes. Here we present Sim2Real-PK, a hybrid neural-mechanistic framework that reconceptualizes pharmacokinetic forecasting as a time-aware latent state-space trajectory inference task. Sim2Real-PK first pretrains a neural dynamics core on massive synthetic datasets derived from mechanistic ordinary differential equation systems to internalize universal kinetic laws, then adapts to real-world clinical heterogeneity through a constrained residual correction pathway. Validated on individual participant data from the YODA Project for two drugs representing distinct pharmacological regimes\u0026mdash;ibrutinib and daratumumab\u0026mdash;Sim2Real-PK significantly outperformed conventional population-level baselines. The framework improved the F50 for ibrutinib from 32.1% to 39.9% and the F20 for daratumumab from 38.8% to 51.1%, while faithfully recovering expected linear and nonlinear pharmacological behaviors. Crucially, the model successfully resolved latent pharmacokinetic heterogeneity within patient cohorts that remained obscured under traditional analytical frameworks. These results demonstrate that mechanistically-anchored neural adaptation can bridge the gap between biological rigor and predictive agility, providing a robust foundation for next-generation model-informed precision dosing.\u003c/p\u003e","manuscriptTitle":"Sim2Real-PK framework for precision dosing: Integrating pharmacokinetic mechanisms and deep learning to decode clinical heterogeneity from sparse data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-24 06:33:18","doi":"10.21203/rs.3.rs-9308510/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"340016ed-71c0-43a5-b251-6dd3eec03a6c","owner":[],"postedDate":"April 24th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Rejected","date":"2026-05-07T20:15:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-07T14:52:21+00:00","index":17,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T10:17:12+00:00","index":16,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-04T23:57:48+00:00","index":15,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":66505418,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":66505419,"name":"Biological sciences/Drug discovery"}],"tags":[],"updatedAt":"2026-05-07T20:24:04+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-24 06:33:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9308510","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9308510","identity":"rs-9308510","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}