From Categorical to Continuous: A Symbiotic Human-AI Approach to HER2 Scoring in the Antibody-Drug Conjugates Era

From Categorical to Continuous: A Symbiotic Human-AI Approach to HER2 Scoring in the Antibody-Drug Conjugates Era | 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 Method Article From Categorical to Continuous: A Symbiotic Human-AI Approach to HER2 Scoring in the Antibody-Drug Conjugates Era Min-Hsiang Chang, Hsin-Hsiu Tsai, Chun-Jui Chien, Jian-Chiao Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8372972/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 The advent of antibody-drug conjugates (ADCs) has fundamentally transformed HER2 assessment in breast cancer, shifting therapeutic focus from gene amplification to protein expression levels. This paradigm exposes critical limitations in traditional categorical scoring: inability to preserve expression gradients, poor interobserver agreement at clinically relevant boundaries, and systematic discarding of information within borderline categories (HER2-low and HER2-ultra-low) on which ADC treatment decisions now depend. We developed a continuous HER2 scoring framework (c-score) calculated as the weighted average of categorical proportions, maintaining visual-cognitive correspondence with traditional assessment while enabling quantitative precision. In a proof-of-concept cohort (66 cases spanning all HER2 categories), c-score achieved robust discrimination across clinical decision tasks: unambiguous HER2 3+ identification (AUC=1.00), effective ISH triage (AUC=0.96), and accurate HER2-low detection (AUC=0.98). Critically, c-score revealed substantial heterogeneity within borderline categories, preserving gradient information that categorical boundaries necessarily discard. Multiple independent groups converging on continuous quantification through diverse methodologies suggest this evolution is neither speculative nor optional, but an inevitable response to contemporary therapeutic biology. Validation in outcome cohorts will determine whether preserved expression gradients improve patient selection for ADCs. Pathology Full Text Additional Declarations The authors declare potential competing interests as follows: Authors Hsin-Hsiu Tsai and Chun-Jui Chien were employed by the company Quanta Computer Inc. The all authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results. Supplementary Files WSITinyVitresultsdetailed.csv Dataset 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-8372972","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Method Article","associatedPublications":[],"authors":[{"id":560997959,"identity":"ff9e416e-53f0-41a6-bd36-ba738c0ac6fa","order_by":0,"name":"Min-Hsiang Chang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYJCCAxCK+RiYYmMnVgsPA1saA0MCUAszsVbxMPCYgbUwENKi23784oGPbXYM9hI53x58/LFNno+ZgfHDxxzcWszO5BQcnNmWzMAjkbvdcEbCbcM2ZgZmyZnb8Gg5kJNwmOcMMwOPdO42aZ6E24xALWzMvPi0nH+TcPjPmXqglpxnIC32hLXcSD9wmKHiMEgLG0hLIhFa3jAc7Kk4zsBz/5mZ5Iy028ltzIzN+P1yPv3xhx8G1QzsPYefSXywuW07v7354IePeLQAI8QARNY3IEQYG7CrhAP2BwQUjIJRMApGwYgHAKnOT3dOi4+oAAAAAElFTkSuQmCC","orcid":"","institution":"Li Jen Pathology Clinic","correspondingAuthor":true,"prefix":"","firstName":"Min-Hsiang","middleName":"","lastName":"Chang","suffix":""},{"id":560997960,"identity":"3aeabb2e-d7f1-4889-b9b7-ffd59ee28253","order_by":1,"name":"Hsin-Hsiu Tsai","email":"","orcid":"","institution":"AI Lab, Quanta Computer lnc.","correspondingAuthor":false,"prefix":"","firstName":"Hsin-Hsiu","middleName":"","lastName":"Tsai","suffix":""},{"id":560997961,"identity":"16670375-cd45-4542-ab1d-8ba4cf7426d6","order_by":2,"name":"Chun-Jui Chien","email":"","orcid":"","institution":"AI Lab, Quanta Computer lnc.","correspondingAuthor":false,"prefix":"","firstName":"Chun-Jui","middleName":"","lastName":"Chien","suffix":""},{"id":560997962,"identity":"16ed2e8a-771b-4cc3-8385-c350460a335c","order_by":3,"name":"Jian-Chiao Wang","email":"","orcid":"","institution":"AI Lab, Quanta Computer lnc.","correspondingAuthor":false,"prefix":"","firstName":"Jian-Chiao","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-12-16 07:18:35","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":true,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-8372972/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8372972/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98624358,"identity":"01066cc4-f4d5-4582-8b50-a772ba0e5af4","added_by":"auto","created_at":"2025-12-19 17:08:21","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":541483,"visible":true,"origin":"","legend":"","description":"","filename":"HER2Perspective2024.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8372972/v1_covered_06b06af7-26a6-4ae4-85a9-3940d4a4fe11.pdf"},{"id":98478460,"identity":"4e013965-66e2-4e87-a8a4-2128c6230b0a","added_by":"auto","created_at":"2025-12-18 04:30:03","extension":"csv","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3085,"visible":true,"origin":"","legend":"\u003cp\u003eDataset\u003c/p\u003e","description":"","filename":"WSITinyVitresultsdetailed.csv","url":"https://assets-eu.researchsquare.com/files/rs-8372972/v1/26a74ea9e89415010f05e8b5.csv"}],"financialInterests":"The authors declare potential competing interests as follows: Authors Hsin-Hsiu Tsai and Chun-Jui Chien were employed by the company Quanta Computer Inc. The all authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eFrom Categorical to Continuous: A Symbiotic Human-AI Approach to HER2 Scoring in the Antibody-Drug Conjugates Era\u003c/strong\u003e\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Li Jen Pathology Clinic","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-8372972/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8372972/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe advent of antibody-drug conjugates (ADCs) has fundamentally transformed HER2 assessment in breast cancer, shifting therapeutic focus from gene amplification to protein expression levels. This paradigm exposes critical limitations in traditional categorical scoring: inability to preserve expression gradients, poor interobserver agreement at clinically relevant boundaries, and systematic discarding of information within borderline categories (HER2-low and HER2-ultra-low) on which ADC treatment decisions now depend. We developed a continuous HER2 scoring framework (c-score) calculated as the weighted average of categorical proportions, maintaining visual-cognitive correspondence with traditional assessment while enabling quantitative precision. In a proof-of-concept cohort (66 cases spanning all HER2 categories), c-score achieved robust discrimination across clinical decision tasks: unambiguous HER2 3+ identification (AUC=1.00), effective ISH triage (AUC=0.96), and accurate HER2-low detection (AUC=0.98). Critically, c-score revealed substantial heterogeneity within borderline categories, preserving gradient information that categorical boundaries necessarily discard. Multiple independent groups converging on continuous quantification through diverse methodologies suggest this evolution is neither speculative nor optional, but an inevitable response to contemporary therapeutic biology. Validation in outcome cohorts will determine whether preserved expression gradients improve patient selection for ADCs.\u003c/p\u003e","manuscriptTitle":"From Categorical to Continuous: A Symbiotic Human-AI Approach to HER2 Scoring in the Antibody-Drug Conjugates Era","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-18 04:29:59","doi":"10.21203/rs.3.rs-8372972/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":"deb50e1e-79f3-4683-97c6-4ae54a4c83f4","owner":[],"postedDate":"December 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":59726396,"name":"Pathology"}],"tags":[],"updatedAt":"2025-12-18T04:29:59+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-18 04:29:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8372972","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8372972","identity":"rs-8372972","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}