Disease progression and clinical outcomes in latent osteoarthritis phenotypes: Data from the Osteoarthritis Initiative

Disease progression and clinical outcomes in latent osteoarthritis phenotypes: Data from the Osteoarthritis Initiative | 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 Disease progression and clinical outcomes in latent osteoarthritis phenotypes: Data from the Osteoarthritis Initiative Weihua Guo, ZeYu Huang, Zhao Zhang, Mary Bucklin, John Martin This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3855831/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 prevalence of knee osteoarthritis (OA) is widespread and the heterogeneous patient factors and clinical symptoms in OA patients impede developing personalized treatments for OA patients. In this study, we used unsupervised and supervised machine learning to organize the heterogeneity in knee OA patients and predict disease progression in individuals from the Osteoarthritis Initiative (OAI) dataset. We identified four distinct knee OA phenotypes using unsupervised learning that were defined by nutrition, disability, stiffness, and pain (knee and back) and were strongly related to disease fate. Interestingly, the absence of supplemental vitamins from an individual’s diet was protective from disease progression. Moreover, we established a phenotyping tool and prognostic model from 5 variables (WOMAC disability score of the right knee, WOMAC total score of the right knee, WOMAC total score of the left knee, supplemental vitamins and minerals frequency, and antioxidant combination multivitamins frequency) that can be utilized in clinical practice to determine the risk of knee OA progression in individual patients. We also developed a prognostic model to estimate the risk for total knee replacement and provide suggestions for modifiable variables to improve long-term knee health. This combination of unsupervised and supervised data-driven tools provides a framework to identify knee OA phenotype in a clinical scenario and personalize treatment strategies. Health sciences/Risk factors Health sciences/Biomarkers/Prognostic markers Health sciences/Health care/Public health/Epidemiology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Full Text Additional Declarations There is NO Competing Interest. Supplementary Files OAIKneeOASupplementaryInfov3.pdf Legends for supplementary figures and datasets Table1231208.pdf Table 1 SD1usedinputvariables.csv Dataset 1 SD2cleanv25finalcluster4kmeansdirectknn2imp12112020datacleanmarkerdflargeB.csv Dataset 2 SD3outcomerealdateconversionyear.csv Dataset 3 SD4directpredictcluster230109importancedataframe.xlsx Dataset 4 SD5directpredictcluster230109mergescoredataframe.csv Dataset 5 SD6predictorvis230213gathmerge.csv Dataset 6 SD7mergescoredftestmax9999230202.csv Dataset 7 FigS1231207.pdf FigS2231207.pdf FigS3231207.pdf FigS4231207.pdf FigS5231207.pdf OAIKneeOASupplementaryInfov3.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-3855831","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":267260649,"identity":"6b49f3d2-207e-4796-8553-768bcb846313","order_by":0,"name":"Weihua Guo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAmklEQVRIiWNgGAWjYLCCDwwMCaTpYJxBshZmHpK0GJxfY/jYpsYmj4G99/EL4rTceGNsnHMsrZiB57iZBVFazG6cMZPObTic2CCRxmZAvBZL0rSc7zGTZoRoYX5AlBb7G2zFhj3H0hLbeI6xEaWDQbL/8MYHP2psEvvZ25g/EKdHIgFCA61gkyBOC/8BOJNYW0bBKBgFo2CkAQD+IC2l73RvTAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-9580-513X","institution":"Beckman Research Institute at City of Hope","correspondingAuthor":true,"prefix":"","firstName":"Weihua","middleName":"","lastName":"Guo","suffix":""},{"id":267260650,"identity":"d27c398f-2281-4b38-97e5-28526cf20ec1","order_by":1,"name":"ZeYu Huang","email":"","orcid":"https://orcid.org/0000-0002-0456-8379","institution":"Duke University Hospital","correspondingAuthor":false,"prefix":"","firstName":"ZeYu","middleName":"","lastName":"Huang","suffix":""},{"id":267260651,"identity":"535370e2-52bd-4906-9310-ea53f14bf7d9","order_by":2,"name":"Zhao Zhang","email":"","orcid":"","institution":"West China Hospital, West China Medical School, Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Zhao","middleName":"","lastName":"Zhang","suffix":""},{"id":267260652,"identity":"b40593ad-1eff-4aa1-b62b-99abc45b2236","order_by":3,"name":"Mary Bucklin","email":"","orcid":"https://orcid.org/0000-0003-3893-2818","institution":"Rush University","correspondingAuthor":false,"prefix":"","firstName":"Mary","middleName":"","lastName":"Bucklin","suffix":""},{"id":267260653,"identity":"001dedee-8afc-4583-8f82-9cc7d61eaa22","order_by":4,"name":"John Martin","email":"","orcid":"","institution":"Duke University","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"","lastName":"Martin","suffix":""}],"badges":[],"createdAt":"2024-01-12 05:50:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3855831/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3855831/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":50226711,"identity":"49383255-50d1-4640-85e2-1678f1c44cf7","added_by":"auto","created_at":"2024-01-26 18:17:44","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":349213,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of the experiment design. Osteoarthritis Initiative (OAI) data was organized and cleaned with 4,669 subjects (patients) and 737 variables. Unsupervised clustering was used to stratify the patients into four clusters. The detailed characteristics of each cluster were investigated with cluster annotation and survival analysis. A web-based clinical tool was developed to predict the cluster new patient could belong to with required information. Based on the most accurate WOMAC total score (WOMTS) prediction from an artificial intelligence model, and OA care guideline was also provided for translational usage.\u003c/p\u003e","description":"","filename":"Fig1231207.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/3d0c64ee2671652057e8221e.jpg"},{"id":50225748,"identity":"9a03fa46-bd89-49ad-960b-a4c406d65df5","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":404320,"visible":true,"origin":"","legend":"\u003cp\u003eCluster characteristics of OAI. (A) Four clusters on UMAP. (B) Top 10 numeric variables of each cluster. Kruskal-Wallis test was used to determine the statistics between the cluster of interests and all the other clusters together. Benjamini \u0026amp; Hochberg method was used to adjust the p-value. The numerical variables with adjusted p-values \u0026lt;0.05 were ranked by the log2 fold changes (log2FC) to select the top 10 of each cluster. (C) Top 10 categorical variables of each cluster. Fisher’s exact test was used to determine the statistics between the cluster of interests and all the other clusters together. Benjamini \u0026amp; Hochberg method was used to adjust the p-value. The categorical variables with adjusted p-values \u0026lt;0.05 were ranked by the Pearson’s chi-squared statistics to select the top 10 of each cluster. (D)~(I) Key variables categorized into demographic (V00AGE, age; P02RACE, race; P02SEX, gender; P01BMI, BMI at baseline), medical record (V00COMRB, Charlson Comorbidity Index; V00HSPSS, Short Form 12 Physical Summary Score), pain evaluation (V00WOMKPL/R, WOMAC pain score of left/right knee), diet \u0026amp; nutrition (V00VITCCV, Vitamin C single vitamin, how often taken in past\u003c/p\u003e\n\u003cp\u003e23\u003c/p\u003e\n\u003cp\u003e12 months; V00SUPVITC, average daily Vitamin C supplement, mg), psychological evaluation (V00CESD, Center for Epidemiology Studies Depression Index; V00HSMSS, Short Form 12, Mental Summary Score), socioeconomic status (V00INCOME, annual personal income; V00EDCV, education level) on UMAP.\u003c/p\u003e","description":"","filename":"Fig2231211.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/bf8f150c9628f3e56ef616ca.jpg"},{"id":50225755,"identity":"30ede413-d333-41a9-96c5-d4284875d09f","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1005527,"visible":true,"origin":"","legend":"\u003cp\u003ePrognostic values of 4 OA phenotypes. Kaplan-Meier plots (first and third from left) and forest plots (second and fourth from left) considering good knee health cluster as reference of KL grade (A), joint space width (B), WOMAC total score (C),WOMAC pain score (D), WOMAC stiffness score (E), WOMAC function score (F), and total knee replacement (G) were shown in a table format for both left (left two columns) and right (right two columns) knees. Log-rank p-value was shown in the KM plots. A univariant cox regression model for each outcome variable and each knee was built with the Good Knee group as the reference group and visualized in the forest plots.\u003c/p\u003e","description":"","filename":"Fig3231203.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/8b0100f37bfbb668231f7b43.jpg"},{"id":50225754,"identity":"7ecf2ec4-8b6d-4afb-a0f0-ed4d3cb7b782","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":518830,"visible":true,"origin":"","legend":"\u003cp\u003ePrognostic values of four OA phenotypes within baseline cohorts. (A) Relative distribution of 4 OA phenotypes within each baseline cohort. Kaplan-Meier plots for WOMAC total score (B), KL grades (C), joint space width (D), total knee replacement (E) were shown in a table format for both left (left first and third columns) and right (left second and fourth columns) knees within incidence cohort (left two columns) and progression cohort (right two columns). Log-rank p-value was shown in the KM plots.\u003c/p\u003e","description":"","filename":"Fig4231203.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/7ad2277d0a8d98c520f67e6f.jpg"},{"id":50225758,"identity":"071c8fe5-2737-4648-94ae-b78c6e676d17","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":480280,"visible":true,"origin":"","legend":"\u003cp\u003ePrediction accuracies of cluster labels. A) Screening the optimal number of input\u003c/p\u003e\n\u003cp\u003e24\u003c/p\u003e\n\u003cp\u003evariables (from 2 to 50) with different machine learning models (lr=linear regression, logistic regression model; rf10/20/40/60/80/100tree = random forest model with 10/20/40/60/80/100 trees; svclinear/poly/rbf/sigmoid = supporting vector classifier with linear/polynomial/radial basis function/sigmoid kernels). As a multi-class prediction problem, three accuracy metrics were used, i.e., accuracy (relative correct prediction numbers), roc_auc_ovo (area under curve of receiver operating characteristic curve, one vs one), and roc_auc_ovr (area under cuve of receiver operating characteristic curve, one vs rest). B) Detailed screening of the optimal number of input variables (from 5 to 15). The dot represents the mean of corresponding metric and the error bar represents the standard error of the mean from the cross-validation.\u003c/p\u003e","description":"","filename":"Fig5231203.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/4ed48fedeb8a53dd59267e98.jpg"},{"id":50225753,"identity":"ce82275f-e4e4-4069-b709-8b7c5f522e5f","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":766691,"visible":true,"origin":"","legend":"\u003cp\u003ePrediction accuracy of WOMAC total score at fourth and eighth year. A) Correlation coefficients between WOMTS of all visiting years and baseline variables. The top 10 baseline variables were colored based on the average correlation coefficients crossing all the visiting years. W. = WOMAC, K.=KOOS. B) Screening the optimal number of input variables (from 2 to 50) with different machine learning models (linear, linear regression model; rft10/20/40/60/80/100 = random forest regressor with 20/40/60/80/100 trees; svrlinear/poly/rbf/sigmoid = supporting vector regressor with linear/polynomial/rbf/sigmoid kernels, ann = artificial neural network). As a regression problem, two accuracy metrics were used, i.e., RMSE (root mean squared error) and r (correlation coefficient between prediction and measurements). D) Top 5 variables with highest occurrences from the top 10000 most accurate prediction tests. We randomly selected 25 input variables from the cluster markers and used these variables to train an ANN model with the same settings with cross-validation.\u003c/p\u003e\n\u003cp\u003e25\u003c/p\u003e\n\u003cp\u003eThe above procedure was repeated 10,000 times. The top 1000 most accurate tests were extracted and the relative occurrence of each variable to these 1000 tests was calculated. The top 5 with highest relative occurrences for WOMTS of both left and right knees at fourth and eighth year were selected to visualize here. The dot represents the relative occurrences. W. = WOMAC, K.=KOOS\u003c/p\u003e","description":"","filename":"Fig6231211.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/ec9e00dfa31258289c8b3a1a.jpg"},{"id":71235104,"identity":"98a52e87-d528-46ed-97c0-eb87abce60ad","added_by":"auto","created_at":"2024-12-12 11:39:02","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3911512,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript231214.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1_covered_b94287ca-d5c4-4df2-a255-30adf44d27cf.pdf"},{"id":50226443,"identity":"b00de952-5cc9-4905-b463-fd9c5965e13f","added_by":"auto","created_at":"2024-01-26 18:09:44","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":69207,"visible":true,"origin":"","legend":"Legends for supplementary figures and datasets","description":"","filename":"OAIKneeOASupplementaryInfov3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/bde117b8acf23ae15e3daeb1.pdf"},{"id":50226272,"identity":"f0448703-57bb-4c00-a649-e9b9cd6f521f","added_by":"auto","created_at":"2024-01-26 18:01:44","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":108787,"visible":true,"origin":"","legend":"Table 1","description":"","filename":"Table1231208.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/93b8b74edae04c352ef6f7a7.pdf"},{"id":50225757,"identity":"a60a6cd2-a238-46ff-be05-cf7ac3e04878","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"csv","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":126389,"visible":true,"origin":"","legend":"Dataset 1","description":"","filename":"SD1usedinputvariables.csv","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/afbeca8364e2de2eee3be037.csv"},{"id":50225749,"identity":"21daa154-e7c4-452d-9e74-5038d491497b","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"csv","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":539166,"visible":true,"origin":"","legend":"\u003cp\u003eDataset 2\u003c/p\u003e","description":"","filename":"SD2cleanv25finalcluster4kmeansdirectknn2imp12112020datacleanmarkerdflargeB.csv","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/e73ad7e9f6ada5b2f66421e8.csv"},{"id":50225756,"identity":"15948985-bd48-4060-be0f-33376ee5e4b9","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"csv","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":231655,"visible":true,"origin":"","legend":"Dataset 3","description":"","filename":"SD3outcomerealdateconversionyear.csv","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/df9b5d7b761ef3a7999730e9.csv"},{"id":50225760,"identity":"a0306a8e-0af7-4d5c-a243-fbb2d791eeb3","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":91312,"visible":true,"origin":"","legend":"Dataset 4","description":"","filename":"SD4directpredictcluster230109importancedataframe.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/75eb8a858de7d4eb4d1612ea.xlsx"},{"id":50225759,"identity":"94068e6e-8eb8-4e72-becd-04ccffa1ce5f","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"csv","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":3623917,"visible":true,"origin":"","legend":"Dataset 5","description":"","filename":"SD5directpredictcluster230109mergescoredataframe.csv","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/bad1d5598362125742ca11f0.csv"},{"id":50225761,"identity":"59f5c987-5658-4e78-aa7e-51b25b2b0589","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"csv","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":1177322,"visible":true,"origin":"","legend":"Dataset 6","description":"","filename":"SD6predictorvis230213gathmerge.csv","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/9ac475dde58c7e157eedc0f8.csv"},{"id":50226277,"identity":"2a9a24bb-e67f-4178-8282-d80d41f421f2","added_by":"auto","created_at":"2024-01-26 18:01:44","extension":"csv","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":29495925,"visible":true,"origin":"","legend":"\u003cp\u003eDataset 7\u003c/p\u003e","description":"","filename":"SD7mergescoredftestmax9999230202.csv","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/7ff038901d3bdfc8790d40a0.csv"},{"id":50226444,"identity":"f7bf2f4b-6bf3-48f3-a822-ab893e3335b5","added_by":"auto","created_at":"2024-01-26 18:09:44","extension":"pdf","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":5306710,"visible":true,"origin":"","legend":"","description":"","filename":"FigS1231207.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/901988b21c8cf0669419ef27.pdf"},{"id":50225762,"identity":"7b21db33-fc4d-451f-be63-2672d48cb1c0","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"pdf","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":158636,"visible":true,"origin":"","legend":"","description":"","filename":"FigS2231207.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/6a2863a741065cf3cb6c951a.pdf"},{"id":50226278,"identity":"89b3deda-a50c-4468-b0bf-19817db6130a","added_by":"auto","created_at":"2024-01-26 18:01:44","extension":"pdf","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":977073,"visible":true,"origin":"","legend":"","description":"","filename":"FigS3231207.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/ec22a6c583bde1bbd816a748.pdf"},{"id":50225768,"identity":"e475c849-49fb-4aab-bdc3-0adc88c6b64a","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"pdf","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":2022753,"visible":true,"origin":"","legend":"","description":"","filename":"FigS4231207.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/1fc069da8d7f8897bcb6ca0b.pdf"},{"id":50225766,"identity":"680bc3c6-6f23-4a9f-9f49-eb8edccb5afd","added_by":"auto","created_at":"2024-01-26 17:53:44","extension":"pdf","order_by":14,"title":"","display":"","copyAsset":false,"role":"supplement","size":1311422,"visible":true,"origin":"","legend":"","description":"","filename":"FigS5231207.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/9f76f616a11f1a68b80e9af9.pdf"},{"id":50226275,"identity":"6e7d43ba-eb83-4eba-a71d-6d8a9918a8be","added_by":"auto","created_at":"2024-01-26 18:01:44","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"supplement","size":69207,"visible":true,"origin":"","legend":"","description":"","filename":"OAIKneeOASupplementaryInfov3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3855831/v1/8e1aaa20843c29c145139afe.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Disease progression and clinical outcomes in latent osteoarthritis phenotypes: Data from the Osteoarthritis Initiative","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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-3855831/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3855831/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The prevalence of knee osteoarthritis (OA) is widespread and the heterogeneous patient factors and clinical symptoms in OA patients impede developing personalized treatments for OA patients. In this study, we used unsupervised and supervised machine learning to organize the heterogeneity in knee OA patients and predict disease progression in individuals from the Osteoarthritis Initiative (OAI) dataset. We identified four distinct knee OA phenotypes using unsupervised learning that were defined by nutrition, disability, stiffness, and pain (knee and back) and were strongly related to disease fate. Interestingly, the absence of supplemental vitamins from an individual’s diet was protective from disease progression. Moreover, we established a phenotyping tool and prognostic model from 5 variables (WOMAC disability score of the right knee, WOMAC total score of the right knee, WOMAC total score of the left knee, supplemental vitamins and minerals frequency, and antioxidant combination multivitamins frequency) that can be utilized in clinical practice to determine the risk of knee OA progression in individual patients. We also developed a prognostic model to estimate the risk for total knee replacement and provide suggestions for modifiable variables to improve long-term knee health. This combination of unsupervised and supervised data-driven tools provides a framework to identify knee OA phenotype in a clinical scenario and personalize treatment strategies.","manuscriptTitle":"Disease progression and clinical outcomes in latent osteoarthritis phenotypes: Data from the Osteoarthritis Initiative","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-26 17:53:39","doi":"10.21203/rs.3.rs-3855831/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":"ec04435e-a180-49f2-8357-98bbd4ed919b","owner":[],"postedDate":"January 26th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":28155775,"name":"Health sciences/Risk factors"},{"id":28155776,"name":"Health sciences/Biomarkers/Prognostic markers"},{"id":28155777,"name":"Health sciences/Health care/Public health/Epidemiology"}],"tags":[],"updatedAt":"2024-12-12T11:30:47+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-26 17:53:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3855831","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3855831","identity":"rs-3855831","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}