{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T10:04:48Z","timestamp":1791972288928,"version":"4.3.4"},"reference-count":53,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J Med Internet Res"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec sec-type=\"background\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Sharing information derived from individual patient data is restricted by regulatory frameworks due to privacy concerns. Generative artificial intelligence can generate shareable virtual patient populations as proxies for sensitive reference datasets. Explicit demonstration of privacy is demanded.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"objective\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study evaluated whether a privacy-by-design technique called \u201cavatars\u201d can generate synthetic datasets replicating all reported information from randomized clinical trials (RCTs).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"methods\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We generated 2160 synthetic datasets from two phase 3 RCTs for patients with multiple sclerosis (NCT00213135 and NCT00906399; n=865 and 1516 patients) with different configurations to select one synthetic dataset with optimal privacy and utility for each. Several privacy metrics were computed, including protection against distance-based membership inference attacks. We assessed fidelity by comparing variable distributions and assessed utility by checking that all end points reported in the publications had the same effect directions, were within the reported 95% CIs, and had the same statistical significance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"results\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Protection against membership inference attacks was the hardest privacy metric to optimize, but the technique yielded robust privacy and replication of the primary end points (in 72.5% and 80.8% of the 1080 generated datasets). Utility was uneven across the variables and end points, such that information about some end points could not be captured. With optimized generation configurations, we selected one dataset from each RCT replicating all efficacy end points of the placebo and approved treatment arms while maintaining satisfactory privacy (hidden rate: 85.0% and 93.2%).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"conclusions\">\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Generating synthetic RCT datasets replicating primary and secondary efficacy end points is possible while achieving a satisfactory and explicit level of privacy. To show the potential of this method to unlock health data sharing, we released both placebo arms as open datasets.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.2196\/71297","type":"journal-article","created":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T19:42:01Z","timestamp":1759347721000},"page":"e71297-e71297","source":"Crossref","is-referenced-by-count":1,"title":["Privacy-by-Design Approach to Generate Two Virtual Clinical Trials for Multiple Sclerosis and Release Them as Open Datasets: Evaluation Study"],"prefix":"10.2196","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2504-5294","authenticated-orcid":false,"given":"Stanislas","family":"Demuth","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5726-9620","authenticated-orcid":false,"given":"Olivia","family":"Rousseau","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5735-0173","authenticated-orcid":false,"given":"Igor","family":"Faddeenkov","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-6148-4676","authenticated-orcid":false,"given":"Julien","family":"Paris","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7197-7578","authenticated-orcid":false,"given":"J\u00e9r\u00f4me","family":"De S\u00e8ze","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-4229-7873","authenticated-orcid":false,"given":"B\u00e9atrice","family":"Baciotti","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-8375-7530","authenticated-orcid":false,"given":"Marianne","family":"Payet","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6182-2193","authenticated-orcid":false,"given":"Morgan","family":"Guillaudeux","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-2625-4103","authenticated-orcid":false,"given":"Alban-F\u00e9lix","family":"Barreteau","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6113-6938","authenticated-orcid":false,"given":"David","family":"Laplaud","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9641-6734","authenticated-orcid":false,"given":"Gilles","family":"Edan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1131-9554","authenticated-orcid":false,"given":"Pierre-Antoine","family":"Gourraud","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2025,10,1]]},"reference":[{"issue":"1","key":"R1","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1177\/1352458519868982","article-title":"Big data in MS\u2014What can we learn from large international observational studies such as MSBase?","volume":"26","author":"Warnke","journal-title":"Mult Scler"},{"issue":"9","key":"R2","doi-asserted-by":"publisher","first-page":"1132","DOI":"10.1001\/jamaneurol.2020.1568","article-title":"Contribution of relapse-independent progression vs relapse-associated worsening to overall confirmed disability accumulation in typical relapsing multiple sclerosis in a pooled analysis of 2 randomized clinical trials","volume":"77","author":"Kappos","journal-title":"JAMA Neurol"},{"key":"R3","unstructured":"Center for global clinical research data. Vivli. URL: https:\/\/vivli.org\/ [Accessed 04-11-2023]"},{"key":"R4","unstructured":"ClinicalStudyDataRequest.com. URL: https:\/\/clinicalstudydatarequest.com\/ [Accessed 04-11-2023]"},{"issue":"10","key":"R5","doi-asserted-by":"publisher","first-page":"957","DOI":"10.2165\/11537420-000000000-00000","article-title":"No head-to-head trial? Simulate the missing arms","volume":"28","author":"Caro","journal-title":"Pharmacoeconomics"},{"issue":"6","key":"R6","doi-asserted-by":"publisher","first-page":"940","DOI":"10.1016\/j.jval.2012.05.004","article-title":"Matching-adjusted indirect comparisons: a new tool for timely comparative effectiveness research","volume":"15","author":"Signorovitch","journal-title":"Value Health"},{"issue":"6","key":"R7","doi-asserted-by":"publisher","first-page":"1053","DOI":"10.1002\/ana.26626","article-title":"Ravulizumab in aquaporin-4-positive neuromyelitis optica spectrum disorder","volume":"93","author":"Pittock","journal-title":"Ann Neurol"},{"issue":"4","key":"R8","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pmed.1001202","article-title":"Open clinical trial data for all? A view from regulators","volume":"9","author":"Eichler","journal-title":"PLoS Med"},{"key":"R9","unstructured":"Opinion 05\/2014 on anonymisation techniques. Data Protection Working Party. 2014. URL: https:\/\/ec.europa.eu\/justice\/article-29\/documentation\/opinion-recommendation\/files\/2014\/wp216_en.pdf [Accessed 23-11-2023]"},{"issue":"1","key":"R10","doi-asserted-by":"publisher","first-page":"3069","DOI":"10.1038\/s41467-019-10933-3","article-title":"Estimating the success of re-identifications in incomplete datasets using generative models","volume":"10","author":"Rocher","journal-title":"Nat Commun"},{"key":"R11","unstructured":"L\u2019anonymisation de donn\u00e9es personnelles. Commission Nationale de l\u2019informatique et des libert\u00e9s. 2020. URL: https:\/\/www.cnil.fr\/fr\/lanonymisation-de-donnees-personnelles [Accessed 12-12-2023]"},{"issue":"1","key":"R12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41746-023-00927-3","article-title":"Harnessing the power of synthetic data in healthcare: innovation, application, and privacy","volume":"6","author":"Giuffr\u00e8","journal-title":"npj Digit Med"},{"key":"R13","doi-asserted-by":"crossref","unstructured":"Zhu JY Kr\u00e4henb\u00fchl P Shechtman E Efros AA . Generative visual manipulation on the natural image manifold. In: Leibe B Matas J Sebe N Welling M , editors. Comput Vis \u2013 ECCV 2016. Springer International Publishing; 2016:597-613. [doi: 10.1007\/978-3-319-46454-1_36]","DOI":"10.1007\/978-3-319-46454-1_36"},{"issue":"8","key":"R14","doi-asserted-by":"publisher","first-page":"1930","DOI":"10.1038\/s41591-023-02448-8","article-title":"Large language models in medicine","volume":"29","author":"Thirunavukarasu","journal-title":"Nat Med"},{"issue":"3","key":"R15","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1016\/j.neurol.2024.04.004","article-title":"Clinical applications of deep learning in neuroinflammatory diseases: a scoping review","volume":"181","author":"Demuth","journal-title":"Rev Neurol (Paris)"},{"issue":"4","key":"R16","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1109\/MSEC.2020.2992821","article-title":"Seven ways to evaluate the utility of synthetic data","volume":"18","author":"El Emam","journal-title":"IEEE Secur Privacy"},{"key":"R17","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2019.12.136","article-title":"Generation and evaluation of privacy preserving synthetic health data","volume":"416","author":"Yale","journal-title":"Neurocomputing"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.2196\/53008","article-title":"Generative AI in medical practice: in-depth exploration of privacy and security challenges","volume":"26","author":"Chen","journal-title":"J Med Internet Res"},{"key":"R19","unstructured":"Roundtable of G7 data protection and privacy authorities statement on generative AI. G7 Data Protection and Privacy Authorities. 2023. URL: https:\/\/www.cnil.fr\/sites\/cnil\/files\/2023-06\/g7roundtable_202306_statement.pdf [Accessed 05-09-2025]"},{"issue":"4","key":"R20","doi-asserted-by":"publisher","first-page":"3367","DOI":"10.1109\/TKDE.2021.3130903","article-title":"Adversarial attacks against deep generative models on data: a survey","volume":"35","author":"Sun","journal-title":"IEEE Trans Knowl Data Eng"},{"issue":"4","key":"R21","doi-asserted-by":"publisher","DOI":"10.1093\/jamiaopen\/ooac083","article-title":"Validating a membership disclosure metric for synthetic health data","volume":"5","author":"El Emam","journal-title":"JAMIA Open"},{"issue":"5","key":"R22","doi-asserted-by":"publisher","first-page":"1699","DOI":"10.1093\/bib\/bby043","article-title":"In silico clinical trials: concepts and early adoptions","volume":"20","author":"Pappalardo","journal-title":"Brief Bioinform"},{"issue":"7","key":"R23","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2018.5474","article-title":"Evaluation of digital breast tomosynthesis as replacement of full-field digital mammography using an in silico imaging trial","volume":"1","author":"Badano","journal-title":"JAMA Netw Open"},{"issue":"Suppl 6","key":"R24","doi-asserted-by":"publisher","first-page":"294","DOI":"10.1186\/s12911-022-02034-x","article-title":"In silico clinical trials for relapsing-remitting multiple sclerosis with MS TreatSim","volume":"22","author":"Sips","journal-title":"BMC Med Inform Decis Mak"},{"key":"R25","unstructured":"Synthetic SDTM sample dataset. GitHub. URL: https:\/\/github.com\/lhncbc\/r-snippets-bmi\/tree\/master\/cdisc\/inst\/extdata\/cdisc01\/csv [Accessed 19-07-2023]"},{"issue":"4","key":"R26","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2020-043497","article-title":"Can synthetic data be a proxy for real clinical trial data? A validation study","volume":"11","author":"Azizi","journal-title":"BMJ Open"},{"key":"R27","doi-asserted-by":"publisher","DOI":"10.2196\/66821","article-title":"Augmenting insufficiently accruing oncology clinical trials using generative models: validation study","volume":"27","author":"El Kababji","journal-title":"J Med Internet Res"},{"key":"R28","doi-asserted-by":"publisher","DOI":"10.2196\/55118","article-title":"Comparison of synthetic data generation techniques for control group survival data in oncology clinical trials: simulation study","volume":"12","author":"Akiya","journal-title":"JMIR Med Inform"},{"issue":"8","key":"R29","doi-asserted-by":"publisher","first-page":"758","DOI":"10.1093\/aje\/kwv254","article-title":"Using big data to emulate a target trial when a randomized trial is not available","volume":"183","author":"Hern\u00e1n","journal-title":"Am J Epidemiol"},{"issue":"1","key":"R30","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-022-30908-1","article-title":"Addressing challenges with real-world synthetic control arms to demonstrate the comparative effectiveness of Pralsetinib in non-small cell lung cancer","volume":"13","author":"Popat","journal-title":"Nat Commun"},{"issue":"S 01","key":"R31","doi-asserted-by":"publisher","first-page":"e19","DOI":"10.1055\/s-0042-1760247","article-title":"Synthetic tabular data evaluation in the health domain covering resemblance, utility, and privacy dimensions","volume":"62","author":"Hernadez","journal-title":"Methods Inf Med"},{"issue":"1","key":"R32","doi-asserted-by":"publisher","DOI":"10.2196\/53542","article-title":"Digital representation of patients as medical digital twins: data-centric viewpoint","volume":"13","author":"Demuth","journal-title":"JMIR Med Inform"},{"issue":"1","key":"R33","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-023-01869-w","article-title":"A method for generating synthetic longitudinal health data","volume":"23","author":"Mosquera","journal-title":"BMC Med Res Methodol"},{"issue":"1","key":"R34","doi-asserted-by":"publisher","first-page":"83","DOI":"10.1038\/s41746-023-00822-x","article-title":"Synthetic electronic health records generated with variational graph autoencoders","volume":"6","author":"Nikolentzos","journal-title":"NPJ Digit Med"},{"issue":"1","key":"R35","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1093\/jamia\/ocaa249","article-title":"Optimizing the synthesis of clinical trial data using sequential trees","volume":"28","author":"Emam","journal-title":"J Am Med Inform Assoc"},{"issue":"9","key":"R36","doi-asserted-by":"publisher","first-page":"375","DOI":"10.3390\/info12090375","article-title":"A review of tabular data synthesis using GANs on an IDS dataset","volume":"12","author":"Bourou","journal-title":"Information"},{"issue":"1","key":"R37","doi-asserted-by":"publisher","first-page":"37","DOI":"10.1038\/s41746-023-00771-5","article-title":"Patient-centric synthetic data generation, no reason to risk re-identification in biomedical data analysis","volume":"6","author":"Guillaudeux","journal-title":"NPJ Digit Med"},{"key":"R38","unstructured":"Zhao Z Kunar A Scheer H Birke R Chen LY . CTAB-GAN: effective table data synthesizing. arXiv. Preprint posted online on Feb 16, 2021. [doi: 10.48550\/arXiv.2102.08369]"},{"key":"R39","doi-asserted-by":"publisher","DOI":"10.18637\/jss.v074.i11","article-title":"Synthpop: bespoke creation of synthetic data in R","volume":"74","author":"Nowok","journal-title":"J Stat Softw"},{"issue":"5","key":"R40","doi-asserted-by":"publisher","first-page":"416","DOI":"10.1056\/NEJMoa0902533","article-title":"A placebo-controlled trial of oral cladribine for relapsing multiple sclerosis","volume":"362","author":"Giovannoni","journal-title":"N Engl J Med"},{"issue":"7","key":"R41","doi-asserted-by":"publisher","first-page":"657","DOI":"10.1016\/S1474-4422(14)70068-7","article-title":"Pegylated interferon \u03b2-1a for relapsing-remitting multiple sclerosis (ADVANCE): a randomised, phase 3, double-blind study","volume":"13","author":"Calabresi","journal-title":"Lancet Neurol"},{"issue":"1","key":"R42","doi-asserted-by":"publisher","DOI":"10.18637\/jss.v025.i01","article-title":"FactoMineR: an R package for multivariate analysis","volume":"25","author":"L\u00ea","journal-title":"J Stat Softw"},{"issue":"3","key":"R43","doi-asserted-by":"publisher","DOI":"10.18637\/jss.v045.i03","article-title":"Mice: multivariate imputation by chained equations in R","volume":"45","author":"van Buuren","journal-title":"J Stat Softw"},{"key":"R44","unstructured":"Octopize\/saiph: a projection package. GitHub. URL: https:\/\/github.com\/octopize\/saiph [Accessed 08-07-2024]"},{"issue":"6","key":"R45","doi-asserted-by":"publisher","first-page":"819","DOI":"10.1177\/1352458518771875","article-title":"Efficacy of cladribine tablets in high disease activity subgroups of patients with relapsing multiple sclerosis: a post hoc analysis of the CLARITY study","volume":"25","author":"Giovannoni","journal-title":"Mult Scler"},{"key":"R46","doi-asserted-by":"crossref","unstructured":"Venables WN Ripley BD . Modern applied statistics with S. In: Stat Comput. Springer; 2002. URL: http:\/\/link.springer.com\/10.1007\/978-0-387-21706-2 [Accessed 05-09-2025] [doi: 10.1007\/978-0-387-21706-2]","DOI":"10.1007\/978-0-387-21706-2"},{"key":"R47","doi-asserted-by":"crossref","unstructured":"Davison AC Hinkley DV . Bootstrap Methods and Their Application. Cambridge University Press; 1997. URL: https:\/\/www.cambridge.org\/core\/product\/identifier\/9780511802843\/type\/book [Accessed 05-09-2025] [doi: 10.1017\/CBO9780511802843]","DOI":"10.1017\/CBO9780511802843"},{"issue":"85","key":"R48","first-page":"2825","volume":"12","author":"Pedregosa","journal-title":"J Mach Learn Res"},{"key":"R49","unstructured":"Hello from Octopize Docs. Octopize. URL: https:\/\/docs.octopize.io\/ [Accessed 23-05-2024]"},{"key":"R50","unstructured":"Privacy-by-design generation of two virtual clinical trials in multiple sclerosis and their release as open datasets. Figshare. URL: https:\/\/figshare.com\/s\/ba49ed0550fd069567e6 [Accessed 16-08-2024]"},{"key":"R51","unstructured":"Privacy-by-design generation of two virtual clinical trials in multiple sclerosis. GitLab. URL: https:\/\/gitlab.com\/stanislas.demuth\/avatars-for-randomized-clinical-trials\/ [Accessed 16-08-2024]"},{"issue":"10","key":"R52","doi-asserted-by":"publisher","DOI":"10.2196\/20891","article-title":"Federated learning on clinical benchmark data: performance assessment","volume":"22","author":"Lee","journal-title":"J Med Internet Res"},{"issue":"1","key":"R53","doi-asserted-by":"publisher","first-page":"11540","DOI":"10.1038\/s41598-023-38457-3","article-title":"A comparison of synthetic data generation and federated analysis for enabling international evaluations of cardiovascular health","volume":"13","author":"Azizi","journal-title":"Sci Rep"}],"container-title":["Journal of Medical Internet Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jmir.org\/2025\/1\/e71297","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:00:34Z","timestamp":1782842434000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jmir.org\/2025\/1\/e71297"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,1]]},"references-count":53,"URL":"https:\/\/doi.org\/10.2196\/71297","relation":{"has-preprint":[{"id-type":"doi","id":"10.21203\/rs.3.rs-4958414\/v1","asserted-by":"object"},{"id-type":"doi","id":"10.21203\/rs.3.rs-4958414\/v2","asserted-by":"object"}]},"ISSN":["1438-8871"],"issn-type":[{"value":"1438-8871","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,1]]},"article-number":"v27i1e71297"}}