Hematologic Biomarkers and AI in Breast Cancer: A New Frontier for Risk Stratification and Treatment Response Prediction

Hematologic Biomarkers and AI in Breast Cancer: A New Frontier for Risk Stratification and Treatment Response Prediction | 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 Systematic Review Hematologic Biomarkers and AI in Breast Cancer: A New Frontier for Risk Stratification and Treatment Response Prediction Terry B Trent This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6500133/v4 This work is licensed under a CC BY 4.0 License Status: Posted Version 4 posted You are reading this latest preprint version Show more versions Abstract Background Precision oncology for breast cancer increasingly relies on hematologic biomarkers and artificial intelligence (AI) to enhance risk stratification and predict treatment response. Recent advancements in liquid biopsy technologies and machine learning have significantly accelerated progress in this field since 2020 (Bartolomucci et al., 2025). Methods We conducted a comprehensive review of literature published between 2020 and 2025, examining publicly available data on blood-based biomarkers, including complete blood count (CBC) indices, circulating tumor DNA (ctDNA), and circulating microRNAs (miRNAs) in breast cancer. Special emphasis was placed on studies utilizing AI and advanced statistical modeling for risk assessment and prediction of therapy outcomes. Findings from major cohorts and novel pilot studies were synthesized, and an illustrative AI-driven analysis of publicly accessible data was highlighted. Results Evidence increasingly shows that both routine hematologic parameters and advanced liquid biopsy markers have significant prognostic and predictive value. For example, Araujo et al. (2024) demonstrated in a cohort of approximately 400,000 women that machine learning models incorporating age and neutrophil-to-lymphocyte ratio (NLR) effectively stratify breast cancer risk. Elevated NLR has consistently predicted worse survival outcomes (Gao et al., 2023; Xiang et al., 2023), and dynamic changes in NLR during neoadjuvant chemotherapy reliably forecast pathological complete response (Gao et al., 2023). Furthermore, ctDNA has emerged as a sensitive indicator of minimal residual disease and early recurrence, with AI-driven analyses enhancing detection of cancer-specific genomic fragmentation patterns (Parikh et al., 2020). In metastatic breast cancer, shallow whole-genome sequencing combined with Bayesian modeling of ctDNA predicted treatment responses with up to 75% sensitivity, surpassing traditional tumor marker assessments (Beddowes et al., 2025). Additionally, circulating miRNA signatures, especially total circulating miRNA levels, have shown significant prognostic implications for relapse (Ward Gahlawat et al., 2022). Discussion These findings underscore the substantial yet underexplored potential of hematologic biomarkers, especially when integrated with machine learning approaches. Such integration may facilitate non-invasive, cost-effective screening for breast cancer risk and provide real-time monitoring of treatment efficacy. However, challenges remain, particularly in data standardization, prospective validation, and clinical integration of AI-driven methodologies. Conclusion Hematologic biomarkers—ranging from straightforward CBC indices to sophisticated liquid biopsy analytes—are increasingly positioned to complement traditional risk assessment and tissue-based biomarkers. AI-driven analyses offer powerful tools to decode complex biomarker interactions, providing innovative opportunities for personalized breast cancer screening and therapy. Future multidisciplinary research and rigorous clinical trials are essential to validate and incorporate these promising approaches into standard clinical practice, ultimately improving patient outcomes and enabling tailored treatments. Cancer Biology Oncology Internal Medicine Artificial Intelligence and Machine Learning Laboratory Diagnostics breast cancer circulating tumor DNA artificial intelligence liquid biopsy hematologic biomarkers machine learning Figures Figure 1 Figure 2 Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 4 posted You are reading this latest preprint version Show more versions 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-6500133","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":465950063,"identity":"a1b68ddc-053e-478a-b82e-567650dfc75d","order_by":0,"name":"Terry B Trent","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwUlEQVRIiWNgGAWjYPACGwYGCRB9gAi1PGwgMiENSQsbcVoOk6DFXr734ePCH+cT185uf8D44wxDHr98AyFb2I2NZyTcTtx254wBM88NhmLJNoIOY2OT5gFpuZHD/pvhA0PihmOEtbD/5kk4B9SSDnQYUMt+IrSwMfMkHABqSTBgADoscQPBEDuWxizNk5ZsDHQY0C9nJBJnHEvAr4W9+RjjZx4bO1mIw47ZJPY3HyBgDRqQIE35KBgFo2AUjALsAABJcj9QIV8H5AAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0009-0001-0039-771X","institution":"University of Medicine and Health Sciences","correspondingAuthor":true,"prefix":"","firstName":"Terry","middleName":"B","lastName":"Trent","suffix":""}],"badges":[],"createdAt":"2025-04-22 04:23:57","currentVersionCode":4,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6500133/v4","doiUrl":"https://doi.org/10.21203/rs.3.rs-6500133/v4","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":84425814,"identity":"268e545b-10d6-4847-9972-f65a8a8d66f7","added_by":"auto","created_at":"2025-06-11 20:00:20","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":42716,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTop Predictors from an AI-Driven Breast Cancer Risk Model Based on CBC\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e(A)\u003c/em\u003e Ridge regression model coefficients highlighting the top three predictors: age, neutrophil-to-lymphocyte ratio (NLR), and red blood cell (RBC) count. Higher values of age and NLR are associated with increased breast cancer risk, whereas higher RBC counts show an inverse correlation.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e(B)\u003c/em\u003e SHAP summary plot from a LightGBM model illustrating the direction and magnitude of feature impact on breast cancer risk prediction. Pink points represent high feature values contributing to elevated risk (right), and blue points represent lower values or protective factors (left).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAdapted from Araujo et al., 2024, under a Creative Commons license.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"f1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6500133/v4/ae5da31ef50e69501321b9dc.jpg"},{"id":84425815,"identity":"8dfdf15d-0839-4647-afb4-3b59d5997f4c","added_by":"auto","created_at":"2025-06-11 20:00:20","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":31027,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ectDNA Dynamics in Breast Cancer Monitoring and Early Relapse Detection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis schematic demonstrates a hypothetical patient’s tumor burden trajectory over time. After treatment initiation (green area), clinical remission is achieved; however, minimal residual disease (MRD) persists (blue zone). Conventional imaging (dotted line) fails to detect recurrence until tumor burden increases significantly. ctDNA (solid curve) detects molecular recurrence earlier, potentially allowing preemptive therapy adjustments. The vertical arrow indicates the early detection window.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAdapted from Bartolomucci et al., 2025, under a Creative Commons license.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"f2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6500133/v4/148edcce9ec8d07f7fc3bbfd.jpg"},{"id":84427125,"identity":"f4e2c477-3999-415c-896e-0e0eadaccadf","added_by":"auto","created_at":"2025-06-11 20:32:22","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":641859,"visible":true,"origin":"","legend":"","description":"","filename":"HematologicBiomarkersAIBreastCancerRevisedv6.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6500133/v4_covered_0c4039ad-628d-40cf-b675-0dad9d1baa03.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"Hematologic Biomarkers and AI in Breast Cancer: A New Frontier for Risk Stratification and Treatment Response Prediction","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University of Medicine and Health Sciences","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":"breast cancer, circulating tumor DNA, artificial intelligence, liquid biopsy, hematologic biomarkers, machine learning","lastPublishedDoi":"10.21203/rs.3.rs-6500133/v4","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6500133/v4","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003ePrecision oncology for breast cancer increasingly relies on hematologic biomarkers and artificial intelligence (AI) to enhance risk stratification and predict treatment response. Recent advancements in liquid biopsy technologies and machine learning have significantly accelerated progress in this field since 2020 (Bartolomucci et al., 2025).\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003e We conducted a comprehensive review of literature published between 2020 and 2025, examining publicly available data on blood-based biomarkers, including complete blood count (CBC) indices, circulating tumor DNA (ctDNA), and circulating microRNAs (miRNAs) in breast cancer. Special emphasis was placed on studies utilizing AI and advanced statistical modeling for risk assessment and prediction of therapy outcomes. Findings from major cohorts and novel pilot studies were synthesized, and an illustrative AI-driven analysis of publicly accessible data was highlighted.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eEvidence increasingly shows that both routine hematologic parameters and advanced liquid biopsy markers have significant prognostic and predictive value. For example, Araujo et al. (2024) demonstrated in a cohort of approximately 400,000 women that machine learning models incorporating age and neutrophil-to-lymphocyte ratio (NLR) effectively stratify breast cancer risk. Elevated NLR has consistently predicted worse survival outcomes (Gao et al., 2023; Xiang et al., 2023), and dynamic changes in NLR during neoadjuvant chemotherapy reliably forecast pathological complete response (Gao et al., 2023). Furthermore, ctDNA has emerged as a sensitive indicator of minimal residual disease and early recurrence, with AI-driven analyses enhancing detection of cancer-specific genomic fragmentation patterns (Parikh et al., 2020). In metastatic breast cancer, shallow whole-genome sequencing combined with Bayesian modeling of ctDNA predicted treatment responses with up to 75% sensitivity, surpassing traditional tumor marker assessments (Beddowes et al., 2025). Additionally, circulating miRNA signatures, especially total circulating miRNA levels, have shown significant prognostic implications for relapse (Ward Gahlawat et al., 2022).\u003c/p\u003e\u003ch2\u003eDiscussion\u003c/h2\u003e \u003cp\u003eThese findings underscore the substantial yet underexplored potential of hematologic biomarkers, especially when integrated with machine learning approaches. Such integration may facilitate non-invasive, cost-effective screening for breast cancer risk and provide real-time monitoring of treatment efficacy. However, challenges remain, particularly in data standardization, prospective validation, and clinical integration of AI-driven methodologies.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eHematologic biomarkers\u0026mdash;ranging from straightforward CBC indices to sophisticated liquid biopsy analytes\u0026mdash;are increasingly positioned to complement traditional risk assessment and tissue-based biomarkers. AI-driven analyses offer powerful tools to decode complex biomarker interactions, providing innovative opportunities for personalized breast cancer screening and therapy. Future multidisciplinary research and rigorous clinical trials are essential to validate and incorporate these promising approaches into standard clinical practice, ultimately improving patient outcomes and enabling tailored treatments.\u003c/p\u003e","manuscriptTitle":"Hematologic Biomarkers and AI in Breast Cancer: A New Frontier for Risk Stratification and Treatment Response Prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":4,"date":"2025-06-11 20:00:16","doi":"10.21203/rs.3.rs-6500133/v4","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}},{"code":3,"date":"2025-06-03 15:04:54","doi":"10.21203/rs.3.rs-6500133/v3","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}},{"code":2,"date":"2025-05-15 19:15:35","doi":"10.21203/rs.3.rs-6500133/v2","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}},{"code":1,"date":"2025-04-23 11:16:49","doi":"10.21203/rs.3.rs-6500133/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":"a2faef3d-b08d-4794-8903-34ff1b446ee0","owner":[],"postedDate":"June 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":49467758,"name":"Cancer Biology"},{"id":49467759,"name":"Oncology"},{"id":49467760,"name":"Internal Medicine"},{"id":49467761,"name":"Artificial Intelligence and Machine Learning"},{"id":49467762,"name":"Laboratory Diagnostics"}],"tags":[],"updatedAt":"2025-05-30T06:45:25+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-11 20:00:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v4","identity":"rs-6500133","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6500133","identity":"rs-6500133","version":["v4"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}