Biologically-informed deep neural networks provide quantitative assessment of intratumoral heterogeneity in post-treatment glioblastoma

Biologically-informed deep neural networks provide quantitative assessment of intratumoral heterogeneity in post-treatment glioblastoma | 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 Biologically-informed deep neural networks provide quantitative assessment of intratumoral heterogeneity in post-treatment glioblastoma Jing Li, Hairong Wang, Michael Argenziano, Hyunsoo Yoon, Deborah Boyett, and 12 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3891425/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 19 Oct, 2024 Read the published version in npj Digital Medicine → Version 1 posted You are reading this latest preprint version Abstract Intratumoral heterogeneity poses a significant challenge to the diagnosis and treatment of glioblastoma (GBM). This heterogeneity is further exacerbated during GBM recurrence, as treatment-induced reactive changes produce additional intratumoral heterogeneity that is ambiguous to differentiate on clinical imaging. There is an urgent need to develop non-invasive approaches to map the heterogeneous landscape of histopathological alterations throughout the entire lesion for each patient. We propose to predictively fuse Magnetic Resonance Imaging (MRI) with the underlying intratumoral heterogeneity in recurrent GBM using machine learning (ML) by leveraging image-localized biopsies with their associated locoregional MRI features. To this end, we develop BioNet, a biologically-informed neural network model, to predict regional distributions of three tissue-specific gene modules: proliferating tumor, reactive/inflammatory cells, and infiltrated brain tissue. BioNet offers valuable insights into the integration of multiple implicit and qualitative biological domain knowledge, which are challenging to describe in mathematical formulations. BioNet performs significantly better than a range of existing methods on cross-validation and blind test datasets. Voxel-level prediction maps of the gene modules by BioNet help reveal intratumoral heterogeneity, which can improve surgical targeting of confirmatory biopsies and evaluation of neuro-oncological treatment effectiveness. The non-invasive nature of the approach can potentially facilitate regular monitoring of the gene modules over time, and making timely therapeutic adjustment. These results also highlight the emerging role of ML in precision medicine. Health sciences/Diseases/Cancer Biological sciences/Cancer Physical sciences/Engineering/Biomedical engineering Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryTable1.xlsx Supplementary Table 1 MRIparameters.xlsx Supplementary Table 2 NMISIclean.pdf Cite Share Download PDF Status: Published Journal Publication published 19 Oct, 2024 Read the published version in npj Digital Medicine → 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-3891425","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":274229313,"identity":"11dc6513-de9e-4fe9-bb06-a640383f028f","order_by":0,"name":"Jing Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2klEQVRIiWNgGAWjYBACxnbG5geJ//7JQXgNQOIAIS3NjG0GD9gOGDMwMBOpBaRQEqglsYFoLczNzA0GCTx30ue7nz/44OcOBjm+GwkEHdbwIEHiWe7GM8nMhr1nGIwlidFikGDAnLuxIZlNmrGNIXEDMVokEhKY0w37H7P/BmqpJ1LLgcMJ8hLJbMxALQkGRGhpM0hsSDPcIPHYWLK3TcJw5pkH+LUYtrc/fvizwUZevj/x4YefbTbyfMcJ2GLYAGUYHABTEviVg4A8nNGAR9UoGAWjYBSMbAAAzi1KSnn5pP4AAAAASUVORK5CYII=","orcid":"","institution":"Georgia Institute of Technology","correspondingAuthor":true,"prefix":"","firstName":"Jing","middleName":"","lastName":"Li","suffix":""},{"id":274229314,"identity":"25153425-b519-4212-a8dd-967f39909117","order_by":1,"name":"Hairong Wang","email":"","orcid":"","institution":"Georgia Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Hairong","middleName":"","lastName":"Wang","suffix":""},{"id":274229315,"identity":"487fa4c3-c68c-4119-bba5-1e2ef759e910","order_by":2,"name":"Michael Argenziano","email":"","orcid":"","institution":"Columbia University Irving Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Argenziano","suffix":""},{"id":274229316,"identity":"f02fe01a-ded1-4a17-8f35-603f2a0b6103","order_by":3,"name":"Hyunsoo Yoon","email":"","orcid":"","institution":"Arizona State University","correspondingAuthor":false,"prefix":"","firstName":"Hyunsoo","middleName":"","lastName":"Yoon","suffix":""},{"id":274229317,"identity":"39a36039-7966-4754-92c3-6ca0f8d25d4d","order_by":4,"name":"Deborah Boyett","email":"","orcid":"","institution":"Columbia University Irving Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Deborah","middleName":"","lastName":"Boyett","suffix":""},{"id":274229318,"identity":"7ec8f6d3-7e8d-4dbc-bd42-0544ca293169","order_by":5,"name":"Akshay Save","email":"","orcid":"","institution":"Columbia University Irving Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Akshay","middleName":"","lastName":"Save","suffix":""},{"id":274229319,"identity":"350fa7fc-5441-4abd-8df3-f985f299a9b5","order_by":6,"name":"Petros Petridis","email":"","orcid":"https://orcid.org/0000-0001-7608-5723","institution":"NYU Grossman School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Petros","middleName":"","lastName":"Petridis","suffix":""},{"id":274229320,"identity":"8e53c9fb-09a7-4900-a32d-76b5ae20238b","order_by":7,"name":"William Savage","email":"","orcid":"","institution":"Columbia University Irving Medical Center","correspondingAuthor":false,"prefix":"","firstName":"William","middleName":"","lastName":"Savage","suffix":""},{"id":274229321,"identity":"fb26c0f4-f7de-47f2-bf33-7cd9183b0955","order_by":8,"name":"Pamela Jackson","email":"","orcid":"","institution":"Mayo Clinic","correspondingAuthor":false,"prefix":"","firstName":"Pamela","middleName":"","lastName":"Jackson","suffix":""},{"id":274229322,"identity":"7843b45f-1491-4499-8aec-c9755b36a5c2","order_by":9,"name":"Andrea Hawkins-Daarud","email":"","orcid":"","institution":"Mayo Clinic","correspondingAuthor":false,"prefix":"","firstName":"Andrea","middleName":"","lastName":"Hawkins-Daarud","suffix":""},{"id":274229323,"identity":"b383cdf5-06f9-43b7-b757-85f94d07b8de","order_by":10,"name":"Nhan Tran","email":"","orcid":"","institution":"Mayo Clinic","correspondingAuthor":false,"prefix":"","firstName":"Nhan","middleName":"","lastName":"Tran","suffix":""},{"id":274229324,"identity":"3541f47b-06fe-4a22-b471-ec32b1db6e81","order_by":11,"name":"Leland Hu","email":"","orcid":"","institution":"Mayo Clinic","correspondingAuthor":false,"prefix":"","firstName":"Leland","middleName":"","lastName":"Hu","suffix":""},{"id":274229325,"identity":"0c130b1d-ebd9-4dcc-ba59-bd89886e14f0","order_by":12,"name":"Osama Al-Dalahmah","email":"","orcid":"","institution":"Columbia University Irving Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Osama","middleName":"","lastName":"Al-Dalahmah","suffix":""},{"id":274229326,"identity":"da3aa755-93ac-496a-93b1-7fdeb10a9b6d","order_by":13,"name":"Jeffrey Bruce","email":"","orcid":"","institution":"Columbia University Irving Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Jeffrey","middleName":"","lastName":"Bruce","suffix":""},{"id":274229327,"identity":"32185dc8-fd5a-4459-9524-4743633e3b69","order_by":14,"name":"Jack Grinband","email":"","orcid":"https://orcid.org/0000-0001-7658-6755","institution":"Columbia University Irving Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Jack","middleName":"","lastName":"Grinband","suffix":""},{"id":274229328,"identity":"2e2965b6-56f5-41b4-aa01-c8c32863bc79","order_by":15,"name":"Kristin Swanson","email":"","orcid":"","institution":"Mayo Clinic","correspondingAuthor":false,"prefix":"","firstName":"Kristin","middleName":"","lastName":"Swanson","suffix":""},{"id":274229329,"identity":"90906784-2043-44c0-937b-9ba54bc7c8ea","order_by":16,"name":"Peter Canoll","email":"","orcid":"","institution":"Columbia University Irving Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"","lastName":"Canoll","suffix":""}],"badges":[],"createdAt":"2024-01-23 15:11:52","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3891425/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3891425/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41746-024-01277-4","type":"published","date":"2024-10-19T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":67039529,"identity":"da7e81e9-efba-483a-b4ad-249a76b86ec9","added_by":"auto","created_at":"2024-10-20 07:06:00","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7973655,"visible":true,"origin":"","legend":"","description":"","filename":"NMIpaperclean.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3891425/v1_covered_676de2d0-b28f-4891-905a-7e35c1781e83.pdf"},{"id":53528582,"identity":"1b991156-d190-4be5-aa89-33f500fbbacc","added_by":"auto","created_at":"2024-03-27 05:40:50","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1028746,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table 1\u003c/p\u003e","description":"","filename":"SupplementaryTable1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3891425/v1/cdeaf0380ef8969d98e258de.xlsx"},{"id":53528581,"identity":"3a0a7177-2a2a-4f5b-a97c-009953316b07","added_by":"auto","created_at":"2024-03-27 05:40:50","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":52761,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table 2\u003c/p\u003e","description":"","filename":"MRIparameters.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3891425/v1/2223057c0ea19e479959d7ab.xlsx"},{"id":53528584,"identity":"ab07e17e-25c0-4619-8994-4b6741f03795","added_by":"auto","created_at":"2024-03-27 05:40:51","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":288278,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"NMISIclean.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3891425/v1/7bed2d82ba307a34a07d9c7b.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Biologically-informed deep neural networks provide quantitative assessment of intratumoral heterogeneity in post-treatment glioblastoma","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-3891425/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3891425/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Intratumoral heterogeneity poses a significant challenge to the diagnosis and treatment of glioblastoma (GBM). This heterogeneity is further exacerbated during GBM recurrence, as treatment-induced reactive changes produce additional intratumoral heterogeneity that is ambiguous to differentiate on clinical imaging. There is an urgent need to develop non-invasive approaches to map the heterogeneous landscape of histopathological alterations throughout the entire lesion for each patient. We propose to predictively fuse Magnetic Resonance Imaging (MRI) with the underlying intratumoral heterogeneity in recurrent GBM using machine learning (ML) by leveraging image-localized biopsies with their associated locoregional MRI features. To this end, we develop BioNet, a biologically-informed neural network model, to predict regional distributions of three tissue-specific gene modules: proliferating tumor, reactive/inflammatory cells, and infiltrated brain tissue. BioNet offers valuable insights into the integration of multiple implicit and qualitative biological domain knowledge, which are challenging to describe in mathematical formulations. BioNet performs significantly better than a range of existing methods on cross-validation and blind test datasets. Voxel-level prediction maps of the gene modules by BioNet help reveal intratumoral heterogeneity, which can improve surgical targeting of confirmatory biopsies and evaluation of neuro-oncological treatment effectiveness. The non-invasive nature of the approach can potentially facilitate regular monitoring of the gene modules over time, and making timely therapeutic adjustment. These results also highlight the emerging role of ML in precision medicine.","manuscriptTitle":"Biologically-informed deep neural networks provide quantitative assessment of intratumoral heterogeneity in post-treatment glioblastoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-03-27 05:40:44","doi":"10.21203/rs.3.rs-3891425/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4db6a14f-7eb2-4900-a3a2-ed47aebe0457","owner":[],"postedDate":"March 27th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":28904943,"name":"Health sciences/Diseases/Cancer"},{"id":28904944,"name":"Biological sciences/Cancer"},{"id":28904945,"name":"Physical sciences/Engineering/Biomedical engineering"}],"tags":[],"updatedAt":"2024-10-20T07:05:42+00:00","versionOfRecord":{"articleIdentity":"rs-3891425","link":"https://doi.org/10.1038/s41746-024-01277-4","journal":{"identity":"npj-digital-medicine","isVorOnly":false,"title":"npj Digital Medicine"},"publishedOn":"2024-10-19 04:00:00","publishedOnDateReadable":"October 19th, 2024"},"versionCreatedAt":"2024-03-27 05:40:44","video":"","vorDoi":"10.1038/s41746-024-01277-4","vorDoiUrl":"https://doi.org/10.1038/s41746-024-01277-4","workflowStages":[]},"version":"v1","identity":"rs-3891425","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3891425","identity":"rs-3891425","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}