Immunotherapy drug target identification using machine learning and patient-derived tumour explant validation

Immunotherapy drug target identification using machine learning and patient-derived tumour explant validation | 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 Immunotherapy drug target identification using machine learning and patient-derived tumour explant validation Kevin Litchfield, Marcellus Augustine, Nuno Rocha Nene, Hongchang Fu, and 13 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5499857/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Immunotherapy has revolutionised cancer treatment, yet few patients respond clinically, necessitating alternative strategies that can benefit these patients. Novel immune-oncology targets can achieve this through bypassing resistance mechanisms to standard therapies. To address this, we introduce MIDAS, a multimodal graph neural network system for immune-oncology target discovery that leverages gene interactions, multi-omic patient profiles, immune cell biology, antigen processing, disease associations, and phenotypic consequences of genetic perturbations. MIDAS generalises to time-sliced data, outcompetes existing methods, including OpenTargets, and distinguishes approved from prospective targets. Moreover, MIDAS recovers immunotherapy response-associated genes in unseen trials, thus capturing tumour-immune dynamics within human tumours. Interpretability analyses reveal a reliance on autoimmunity, regulatory networks, and relevant biological pathways. Functionally perturbing the OSM-OSMR axis, a proposed target, in TRACERx melanoma patient-derived explants yielded reduced dysfunctional CD8 + T cells, which associate with immunotherapy response. Our results present a machine learning framework for analysing multimodal data for immune-oncology discovery. Biological sciences/Computational biology and bioinformatics/Machine learning Biological sciences/Cancer/Cancer microenvironment Biological sciences/Cancer/Tumour immunology Full Text Additional Declarations Yes there is potential Competing Interest. S.T. reports personal fees from Roche, Novartis, AstraZeneca, and Ipsen outside the submitted work; and the following patents filed: indel mutations as a therapeutic target and predictive biomarker (PCTGB2018/051892 and PCTGB2018/051893) and clear-cell renal cell carcinoma biomarkers (P113326GB). N.M. has stock options in and has consulted for Achilles Therapeutics and holds a European patent in determining HLA LOH (PCT/GB2018/052004), a patent pending in determining HLA disruption (PCT/EP2023/059039), and is a co-inventor to a patent to identify responders to cancer treatment (PCT/GB2018/051912). C.S. acknowledges grant support from AstraZeneca, Boehringer-Ingelheim, Bristol Myers Squibb, Pfizer, Roche-Ventana, Invitae (previously Archer Dx Inc. - collaboration in minimal residual disease sequencing technologies), and Ono Pharmaceutical. He is an AstraZeneca Advisory Board member and Chief Investigator for the AZ MeRmaiD 1 and 2 clinical trials and is also chief investigator of the NHS Galleri trial. He has consulted for Achilles Therapeutics, Amgen, AstraZeneca, Pfizer, Novartis, GlaxoSmithKline, MSD, Bristol Myers Squibb, Illumina, Genentech, Roche-Ventana, GRAIL, Medicxi, Metabomed, Bicycle Therapeutics, Roche Innovation Centre Shanghai, and the Sarah Cannon Research Institute, C.S. had stock options in Apogen Biotechnologies and GRAIL until June 2021, and currently has stock options in Epic Bioscience, Bicycle Therapeutics, and has stock options and is co-founder of Achilles Therapeutics. C.S. holds patents relating to assay technology to detect tumour recurrence (PCT/GB2017/053289); to targeting neoantigens (PCT/EP2016/059401), identifying patent response to immune checkpoint blockade (PCT/EP2016/071471), determining HLA LOH (PCT/GB2018/052004), predicting survival rates of patients with cancer (PCT/GB2020/050221), identifying patients who respond to cancer treatment (PCT/GB2018/051912), US patent relating to detecting tumour mutations (PCT/US2017/28013), methods for lung cancer detection (US20190106751A1) and both a European and US patent related to identifying insertion/deletion mutation targets (PCT/GB2018/051892). K.L. has the following disclosures (all unrelated to the current work): patent on indel burden and CPI response pending, patent on ctDNA minimal residual disease calling methods, patent pending on a lung cancer vaccine; speaker fees from Roche tissue diagnostics and Ellipses pharma; research funding from CRUK TDL/Ono/LifeArc alliance and Genesis Therapeutics; and consulting roles with Monopteros Therapeutics, Saga diagnostics, Kynos Therapeutics and Tempus Labs, Inc. Again unrelated to this work, K.L. is currently employed by Isomorphic Labs. The remaining authors declare no competing interests. Supplementary Files SupplementaryTable1.csv Supplementary Table 1 SupplementaryMethodsRefs.docx Supplementary Methods SupplementaryTable4.xlsx Supplementary Table 4 Cite Share Download PDF Status: Under Review 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-5499857","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":385772280,"identity":"ec106deb-d22a-4558-a7f5-1bf3ab676832","order_by":0,"name":"Kevin Litchfield","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7klEQVRIie3RsWrDMBCA4TMCeTni9YxC/ApZQ2nexRg6NWvoYIqMQVnyAB36MDICdzGdA+0QL5kyeMwYqYUOBSseM+ifDsGHTgggFLrDCJjUADrhdqABYP57rn0kklqDTpUd0jcAnETAEWkHgVNIKqPKDOU3zeKq+npUa4TYHBl240T8LNaeiGNTP2xUgYBPS4aHcbJwpJHmlVOuxEZpu9gzMBxuE+JZr8TKkeTsJ+KPUKRE5Ai5WzyLpbUlXWsJ5vVq/1kgp9Oyefc8nz5qM7yUhrKdaQ6X7XqRJEV/PLfjBNj/Aw7eXwmFQqHQlK6UR1EmlTcXvAAAAABJRU5ErkJggg==","orcid":"","institution":"UCL","correspondingAuthor":true,"prefix":"","firstName":"Kevin","middleName":"","lastName":"Litchfield","suffix":""},{"id":385772281,"identity":"190f7a40-6096-4580-9407-8801802bd45a","order_by":1,"name":"Marcellus Augustine","email":"","orcid":"","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Marcellus","middleName":"","lastName":"Augustine","suffix":""},{"id":385772282,"identity":"f877b4c3-c8fa-47f9-b07a-75caa96692a2","order_by":2,"name":"Nuno Rocha Nene","email":"","orcid":"","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Nuno","middleName":"Rocha","lastName":"Nene","suffix":""},{"id":385772283,"identity":"034bb5e2-0135-4471-883f-f06e2da7356b","order_by":3,"name":"Hongchang Fu","email":"","orcid":"https://orcid.org/0009-0003-6582-8554","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Hongchang","middleName":"","lastName":"Fu","suffix":""},{"id":385772284,"identity":"6e40a5f8-7cfd-4034-bf71-c4359b3cab36","order_by":4,"name":"Christopher Pinder","email":"","orcid":"https://orcid.org/0000-0003-4149-226X","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Pinder","suffix":""},{"id":385772285,"identity":"350c826a-dcc8-48d9-8e9b-165c17fdad9c","order_by":5,"name":"Lorena Ligammari","email":"","orcid":"","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Lorena","middleName":"","lastName":"Ligammari","suffix":""},{"id":385772286,"identity":"0539c594-a9eb-4f2c-b954-252af925c59a","order_by":6,"name":"Alexander Simpson","email":"","orcid":"","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Alexander","middleName":"","lastName":"Simpson","suffix":""},{"id":385772287,"identity":"d0d2ddac-2628-446e-8234-3647533c497d","order_by":7,"name":"Irene Sanz-Fernández","email":"","orcid":"","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Irene","middleName":"","lastName":"Sanz-Fernández","suffix":""},{"id":385772288,"identity":"62b1f517-ce4c-4206-8fd1-c01b8ddc76bc","order_by":8,"name":"Krupa Thakkar","email":"","orcid":"","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Krupa","middleName":"","lastName":"Thakkar","suffix":""},{"id":385772289,"identity":"59b0c2c8-549d-4cde-b148-fd9b28a7ebde","order_by":9,"name":"Danwen Qian","email":"","orcid":"","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Danwen","middleName":"","lastName":"Qian","suffix":""},{"id":385772290,"identity":"59cdc6d6-d81c-49a9-ba62-dac77dabfe14","order_by":10,"name":"Evelyn Fitzsimons","email":"","orcid":"https://orcid.org/0000-0001-9280-9971","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Evelyn","middleName":"","lastName":"Fitzsimons","suffix":""},{"id":385772291,"identity":"2bef914e-aab1-4148-9b6a-b6dd772a793b","order_by":11,"name":"Benjamin Simpson","email":"","orcid":"","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Benjamin","middleName":"","lastName":"Simpson","suffix":""},{"id":385772292,"identity":"06cd5b00-6f14-4011-8fa7-077a4cf64b0e","order_by":12,"name":"Heather Niederer","email":"","orcid":"","institution":"Cancer Research Horizons","correspondingAuthor":false,"prefix":"","firstName":"Heather","middleName":"","lastName":"Niederer","suffix":""},{"id":385772293,"identity":"fd078fbf-224d-40a4-9859-f0135f0401f1","order_by":13,"name":"Samra Turajlic","email":"","orcid":"","institution":"The Francis Crick Institute","correspondingAuthor":false,"prefix":"","firstName":"Samra","middleName":"","lastName":"Turajlic","suffix":""},{"id":385772294,"identity":"748f9124-43da-47db-b81b-1c3fe42a00f2","order_by":14,"name":"Nicholas McGranahan","email":"","orcid":"","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Nicholas","middleName":"","lastName":"McGranahan","suffix":""},{"id":385772295,"identity":"8d677383-b7d2-45ec-bbfd-be5639326717","order_by":15,"name":"Chris Watkins","email":"","orcid":"","institution":"Royal Holloway University of London","correspondingAuthor":false,"prefix":"","firstName":"Chris","middleName":"","lastName":"Watkins","suffix":""},{"id":385772296,"identity":"f1f5f27d-70cb-4b0d-93cb-fc991c084bd0","order_by":16,"name":"Charles Swanton","email":"","orcid":"","institution":"The Francis Crick Institute, London, United Kingdom","correspondingAuthor":false,"prefix":"","firstName":"Charles","middleName":"","lastName":"Swanton","suffix":""}],"badges":[],"createdAt":"2024-11-21 18:25:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5499857/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5499857/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":70636562,"identity":"e4e561d5-06bc-4e6f-b905-2b100f57ac40","added_by":"auto","created_at":"2024-12-05 06:35:45","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2212660,"visible":true,"origin":"","legend":"","description":"","filename":"MIDASManuscriptsubmission.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5499857/v1_covered_30468d67-f724-4740-9dad-55d36346c410.pdf"},{"id":70634655,"identity":"ad791695-9c4d-445a-b334-b90aced3fbd5","added_by":"auto","created_at":"2024-12-05 06:11:37","extension":"csv","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":46931,"visible":true,"origin":"","legend":"Supplementary Table 1","description":"","filename":"SupplementaryTable1.csv","url":"https://assets-eu.researchsquare.com/files/rs-5499857/v1/fa9ef18a150cfa05fb316829.csv"},{"id":70634658,"identity":"2f0ba746-59f8-463c-b037-01ebacd34dda","added_by":"auto","created_at":"2024-12-05 06:11:37","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":69978,"visible":true,"origin":"","legend":"Supplementary Methods","description":"","filename":"SupplementaryMethodsRefs.docx","url":"https://assets-eu.researchsquare.com/files/rs-5499857/v1/7c0303e60e04e24efb5fa240.docx"},{"id":70635953,"identity":"71743f3f-e2af-4be5-8f82-964dee0900d0","added_by":"auto","created_at":"2024-12-05 06:27:37","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":9822,"visible":true,"origin":"","legend":"Supplementary Table 4","description":"","filename":"SupplementaryTable4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5499857/v1/19c5d23aa4c6751f33c33efc.xlsx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nS.T. reports personal fees from Roche, Novartis, AstraZeneca, and Ipsen outside the submitted work; and the following patents filed: indel mutations as a therapeutic target and predictive biomarker (PCTGB2018/051892 and PCTGB2018/051893) and clear-cell renal cell carcinoma biomarkers (P113326GB). N.M. has stock options in and has consulted for Achilles Therapeutics and holds a European patent in determining HLA LOH (PCT/GB2018/052004), a patent pending in determining HLA disruption (PCT/EP2023/059039), and is a co-inventor to a patent to identify responders to cancer treatment (PCT/GB2018/051912). C.S. acknowledges grant support from AstraZeneca, Boehringer-Ingelheim, Bristol Myers Squibb, Pfizer, Roche-Ventana, Invitae (previously Archer Dx Inc. - collaboration in minimal residual disease sequencing technologies), and Ono Pharmaceutical. He is an AstraZeneca Advisory Board member and Chief Investigator for the AZ MeRmaiD 1 and 2 clinical trials and is also chief investigator of the NHS Galleri trial. He has consulted for Achilles Therapeutics, Amgen, AstraZeneca, Pfizer, Novartis, GlaxoSmithKline, MSD, Bristol Myers Squibb, Illumina, Genentech, Roche-Ventana, GRAIL, Medicxi, Metabomed, Bicycle Therapeutics, Roche Innovation Centre Shanghai, and the Sarah Cannon Research Institute, C.S. had stock options in Apogen Biotechnologies and GRAIL until June 2021, and currently has stock options in Epic Bioscience, Bicycle Therapeutics, and has stock options and is co-founder of Achilles Therapeutics. C.S. holds patents relating to assay technology to detect tumour recurrence (PCT/GB2017/053289); to targeting neoantigens (PCT/EP2016/059401), identifying patent response to immune checkpoint blockade (PCT/EP2016/071471), determining HLA LOH (PCT/GB2018/052004), predicting survival rates of patients with cancer (PCT/GB2020/050221), identifying patients who respond to cancer treatment (PCT/GB2018/051912), US patent relating to detecting tumour mutations (PCT/US2017/28013), methods for lung cancer detection (US20190106751A1) and both a European and US patent related to identifying insertion/deletion mutation targets (PCT/GB2018/051892). K.L. has the following disclosures (all unrelated to the current work): patent on indel burden and CPI response pending, patent on ctDNA minimal residual disease calling methods, patent pending on a lung cancer vaccine; speaker fees from Roche tissue diagnostics and Ellipses pharma; research funding from CRUK TDL/Ono/LifeArc alliance and Genesis Therapeutics; and consulting roles with Monopteros Therapeutics, Saga diagnostics, Kynos Therapeutics and Tempus Labs, Inc. Again unrelated to this work, K.L. is currently employed by Isomorphic Labs. The remaining authors declare no competing interests.","formattedTitle":"Immunotherapy drug target identification using machine learning and patient-derived tumour explant validation","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5499857/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5499857/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eImmunotherapy has revolutionised cancer treatment, yet few patients respond clinically, necessitating alternative strategies that can benefit these patients. Novel immune-oncology targets can achieve this through bypassing resistance mechanisms to standard therapies. To address this, we introduce MIDAS, a multimodal graph neural network system for immune-oncology target discovery that leverages gene interactions, multi-omic patient profiles, immune cell biology, antigen processing, disease associations, and phenotypic consequences of genetic perturbations. MIDAS generalises to time-sliced data, outcompetes existing methods, including OpenTargets, and distinguishes approved from prospective targets. Moreover, MIDAS recovers immunotherapy response-associated genes in unseen trials, thus capturing tumour-immune dynamics within human tumours. Interpretability analyses reveal a reliance on autoimmunity, regulatory networks, and relevant biological pathways. Functionally perturbing the OSM-OSMR axis, a proposed target, in TRACERx melanoma patient-derived explants yielded reduced dysfunctional CD8\u003csup\u003e+\u003c/sup\u003e T cells, which associate with immunotherapy response. Our results present a machine learning framework for analysing multimodal data for immune-oncology discovery.\u003c/p\u003e","manuscriptTitle":"Immunotherapy drug target identification using machine learning and patient-derived tumour explant validation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-05 06:03:32","doi":"10.21203/rs.3.rs-5499857/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-machine-intelligence","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"natmachintell","sideBox":"Learn more about [Nature Machine Intelligence](http://www.nature.com/natmachintell/)","snPcode":"","submissionUrl":"","title":"Nature Machine Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"2e9d281c-59cd-4e77-a5f1-95edf32e24af","owner":[],"postedDate":"December 5th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":41099705,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"},{"id":41099706,"name":"Biological sciences/Cancer/Cancer microenvironment"},{"id":41099707,"name":"Biological sciences/Cancer/Tumour immunology"}],"tags":[],"updatedAt":"2026-02-04T09:28:45+00:00","versionOfRecord":[],"versionCreatedAt":"2024-12-05 06:03:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5499857","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5499857","identity":"rs-5499857","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}