OBJECTIVE: The objective of this study is to present a report from the Big Multiple Sclerosis Data (BMSD) statistics workshop (Bari - Italy, June 2023) which focused on advanced statistical approaches for real-world data (RWD) analyses in multiple sclerosis (MS). The report emphasises the application of these approaches in predicting individual treatment response, assessing comparative effectiveness and safety of therapies and their sequences, and harmonizing data for large-scale federated analyses. METHODS: The BMSD network, comprising five national registries and the international MSBase database (> 350,000 total patients), convened in June 2023 in Bari (Italy) to review methodological advances in RWD analysis. Experts discussed strengths, limitations, and regulatory implications of frequentist, Bayesian, and machine learning (ML) approaches, with case studies on treatment response modelling, comparative effectiveness, safety surveillance, and Common Data Model (CDM)-based federated learning. RESULTS: Bayesian and ML techniques, integrated with causal inference frameworks, can improve personalized predictions of treatment benefit and risk by using high-dimensional longitudinal data. Propensity score-based methods and marginal structural models remain essential for minimizing confounding in comparative analyses, but require rigorous diagnostics and sensitivity analyses. Adoption of a CDM facilitates harmonization of heterogeneous datasets, while federated learning enables privacy-preserving, multi-jurisdictional collaboration. Together, these innovations address key challenges in studying treatment sequences, rare adverse events, and underrepresented patient groups. CONCLUSIONS: This workshop report highlights how advanced statistical and computational methodologies enhance the robustness, interpretability, and regulatory relevance of MS RWD studies. By promoting the integration of complementary statistical and computational approaches within harmonized data infrastructures, the BMSD network is positioned to accelerate the translation of real-world evidence into precision medicine for MS.

CORe Department of Medicine University of Melbourne Melbourne 3000 Australia

CORESEARCH Center for Outcomes Research and Clinical Epidemiology Pescara Italy

Department of Clinical Neuroscience Karolinska Institute Stockholm Sweden

Department of Neurology and Centre of Clinical Neuroscience 1st Faculty of Medicine Charles University Prague and General University Hospital Prague Czech Republic

Department of Neurology Danish Multiple Sclerosis Center Copenhagen University Hospital Rigshospitalet 2100 Copenhagen Denmark

Department of Neuroscience Central Clinical School Monash University Melbourne VIC Australia

Department of Translational Biomedicine and Neurosciences DiBraiN University of Bari Aldo Moro Bari Italy

Department of Translational Biomedicine and Neurosciences DiBrain University of Bari Aldo Moro Piazza Umberto 1 70121 Bari Italy

Eugène Devic EDMUS Foundation Against Multiple Sclerosis State Approved Foundation 69677 Bron France

INSERM 1028 et CNRS UMR 5292 Observatoire Français de la Sclérose en Plaques Centre de Recherche en Neurosciences de Lyon 69003 Lyon France

Institute for Clinical Medicine University of Copenhagen Copenhagen Denmark

Medical Physics Section Department of Biomedicine and Prevention University of Rome Tor Vergata Italy

Neuroimmunology Centre Department of Neurology Royal Melbourne Hospital Melbourne 3000 Australia

ReMuS Registry ReMuS Endowment Fund Prague Czech Republic

Service de Neurologie sclérose en plaques pathologies de la myéline et neuro inflammation Hospices Civils de Lyon Hôpital Neurologique Pierre Wertheimer 69677 Bron France

The Danish Multiple Sclerosis Registry Copenhagen University Hospital Rigshospitalet Copenhagen Denmark

Unit of Biostatistics Fondazione IRCCS Casa Sollievo Della Sofferenza San Giovanni Rotondo Italy

Université de Lyon Université Claude Bernard Lyon 1 69000 Lyon France

Zobrazit více v PubMed

Basch E, Schrag D (2019) The evolving uses of “real-world” data. JAMA 321(14):1359–1360. https://doi.org/10.1001/jama.2019.4064 PubMed DOI

U.S. food and drug administration. Framework for FDA’s real-world evidence program. (2018); https://www.fda.gov/media/120060/download?attachment ; European medicines agency, Accessed 14 Jan, 2025

Real-world evidence framework to support EU regulatory decision-making: Report on the experience gained with regulator-led studies from september 2021 to february 2023. (2023); https://www.ema.europa.eu/en/documents/report/real-world-evidence-framework-support-eu-regulatory-decision-making-report-experience-gained-regulator-led-studies-september-2021-february-2023_en.pdf , Accessed 14 Jan, 2025

Glaser A, Butzkueven H, van der Walt A et al (2024) Big multiple sclerosis data network: an international registry research network. J Neurol 271(6):3616–3624. https://doi.org/10.1007/s00415-024-12303-6 PubMed DOI PMC

Trojano M, Tintore M, Montalban X et al (2017) Treatment decisions in multiple sclerosis - insights from real-world observational studies. Nat Rev Neurol 13(2):105–118. https://doi.org/10.1038/nrneurol.2016.188 PubMed DOI

Pellegrini F et al (2019) A proof-of-concept application of a novel scoring approach for personalised medicine in multiple sclerosis. Mult Scler J. https://doi.org/10.1177/1352458519849513 DOI

Kappos L, Bar-Or A, Cree BAC et al (2018) Siponimod versus placebo in secondary progressive multiple sclerosis (EXPAND): a double-blind, randomised, phase 3 study. Lancet 391:1263–1273 PubMed DOI

Diouf I, Malpas CB, Sharmin S et al (2023) Variability of the response to immunotherapy among sub-groups of patients with multiple sclerosis. Eur J Neurol. https://doi.org/10.1111/ene.15706 PubMed DOI

Kalincik T, Manouchehrinia A, Sobisek L et al (2017) Towards personalised therapy for multiple sclerosis: prediction of individual treatment response. Brain 140:2426–2443 PubMed DOI

Claggett B, Tian L, Castagno D, Wei LJ (2015) Treatment selections using risk-benefit profiles based on data from comparative randomised clinical trials with multiple endpoints. Biostatistics 16:60–72 PubMed DOI

Bovis F, Kalincik T, Lublin F et al (2021) Treatment response score to glatiramer acetate or interferon beta-1a. Neurology 96:e214–e227 PubMed DOI PMC

Steyerberg EW, Harrell FE (2015) Prediction models need appropriate internal, internaleexternal, and external validation. J Clin Epidemiol. https://doi.org/10.1016/j.jclinepi.2015.04.005 PubMed DOI PMC

Steyerberg EW et al (2010) Assessing the performance of prediction models: a framework for some traditional and novel measures. Epidemiology 21(1):128–138 PubMed DOI PMC

Chalkou K et al (2023) Decision curve analysis for personalised treatment choise between multiple options. Med Decis Making 43(3):337–349 PubMed DOI

Chen T, Guestrin C (2016) Xgboost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. pp 785–794

Breiman L (2001) Random forests. Mach Learn 45(1):5–32 DOI

Katzman JL et al (2018) DeepSurv: personalised treatment recommender system using a cox proportional hazards deep neural network. BMC Med Res Methodol 18(1):24 PubMed DOI PMC

Ranganath R et al (2016) Deep survival analysis. arXiv 1608:02158

Lundberg SM, Lee SI (2017) A unified approach to interpreting model predictions. Adv Neural Inf Process Syst 30:4765–4774

Conti A et al (2023) An interpretable machine learning model to predict cortical atrophy in multiple sclerosis. Brain Sci 13(2):198 PubMed DOI PMC

Treaba CA et al (2021) Cortical and phase rim lesions on 7 T MRI as markers of multiple sclerosis disease progression. Brain Commun 3(3):fcab134 PubMed DOI PMC

Caruana R (1997) Multitask learning. Mach Learn 28(1):41–75 DOI

Zhang Y, Yang Q (2021) A survey on multi-task learning. IEEE Trans Knowl Data Eng. https://doi.org/10.1109/TKDE.2021.3070203 PubMed DOI PMC

Verhaeghe J et al (2024) Conformal prediction for dose-response models with continuous treatments. arXiv 240920412. https://doi.org/10.48550/arXiv.2409.20412 DOI

Athey S, Imbens GW (2016) Recursive partitioning for heterogeneous causal effects. Proc Natl Acad Sci 113(27):7353–7360 PubMed DOI PMC

Wager S, Athey S (2018) Estimation and inference of heterogeneous treatment effects using random forests. J Am Stat Assoc 113(523):1228–1242 DOI

Chernozhukov V et al (2018) Double/debiased machine learning for treatment and structural parameters. Economet J 21(1):C1–C68 DOI

Simoneau G, Pellegrini F, Debray T et al (2022) Recommendations for the use of propensity score methods in multiple sclerosis research. Mult Scler J 28(9):1467–1480 DOI

Lefort M, Sharmin S, Andersen JB et al (2022) Impact of methodological choices in comparative effectiveness studies: application in natalizumab versus fingolimod comparison among patients with multiple sclerosis. BMC Med Res Methodol 22(1):155. https://doi.org/10.1186/s12874-022-01623-8 PubMed DOI PMC

Roos I, Diouf I, Sharmin S et al (2023) Comparative effectiveness in multiple sclerosis: a methodological comparison. Mult Scler J 29(3):326–332 DOI

Iaffaldano P, Lucisano G, Butzkueven H et al (2021) Early treatment delays long-term disability accrual in RRMS: results from the BMSD network. Mult Scler J 27(10):1543–1555 DOI

Spelman T, Magyari M, Piehl F et al (2021) Treatment escalation vs immediate initiation of highly effective treatment for patients with relapsing-remitting multiple sclerosis: data from two national strategies. JAMA Neurol 78(10):1197–1204 PubMed DOI

Hillert J, Magyari M, Soelberg Sørensen P et al (2021) Treatment switching and discontinuation over 20 years in the Big Multiple Sclerosis Data Network. Front Neurol 12:647811 PubMed DOI PMC

Hrnciarova T, Drahota J, Spelman T et al (2023) Does initial high efficacy therapy surpass escalation treatment strategy? Mult Scler Relat Disord 76:104803 PubMed DOI

Spelman T, Herring WL, Acosta C et al (2023) Comparative effectiveness and cost-effectiveness of natalizumab and fingolimod in rapidly evolving severe relapsing-remitting multiple sclerosis in the United Kingdom. J Med Econ. https://doi.org/10.1080/13696998.2023.2293379 PubMed DOI

Muñoz J, Efthimiou O, Audigier V et al (2024) Multiple imputation of incomplete multilevel data using Heckman selection models. Stat Med 43(3):514–533 PubMed DOI

Iaffaldano P, Lucisano G, Manni A et al (2022) Risk of getting COVID-19 in people with multiple sclerosis: a case-control study. Neurol Neuroimmunol Neuroinflamm 9(2):e1141 PubMed DOI PMC

Sormani MP, Schiavetti I, Carmisciano L, MuSC-19 Study Group et al (2021) COVID-19 severity in multiple sclerosis: putting data into context. Neurol Neuroimmunol Neuroinflamm 9(1):e1105. https://doi.org/10.1212/NXI.0000000000001105 PubMed DOI PMC

Zeitlin J, Philibert M, Estupiñán-Romero F et al (2023) Developing and testing a protocol using a common data model for federated collection and analysis of national perinatal health indicators in Europe. Open Res Europe 3:54. https://doi.org/10.12688/openreseurope.15701.2 DOI

Sikora A, Keats K, Murphy DJ et al (2024) A common data model for the standardization of intensive care unit medication features. JAMIA Open 7(2):ooae033. https://doi.org/10.1093/jamiaopen/ooae033 PubMed DOI PMC

Wilkinson MD, Dumontier M, Aalbersberg IJJ et al (2016) The FAIR guiding principles for scientific data management and stewardship. Sci Data 3:160018. https://doi.org/10.1038/sdata.2016.18 PubMed DOI PMC

Overhage JM, Ryan PB, Reich CG, Hartzema AG, Stang PE (2012) Validation of a common data model for active safety surveillance research. J Am Med Inform Assoc 19(1):54–60. https://doi.org/10.1136/amiajnl-2011-000376 PubMed DOI

Stang PE, Ryan PB, Racoosin JA et al (2010) Advancing the science for active surveillance: rationale and design for the observational medical outcomes partnership. Ann Intern Med 153(9):600–606. https://doi.org/10.7326/0003-4819-153-9-201011020-00010 PubMed DOI

Drahota J et al. Common data model in multiple sclerosis collaborative research: proved concept and lessons learned in the big MSData network. [Manuscript in preparation]

Brown EG, Wood L, Wood S (1999) The medical dictionary for regulatory activities (MedDRA). Drug Saf 20(2):109–117. https://doi.org/10.2165/00002018-199920020-00002 PubMed DOI

Li R, Romano JD, Chen Y, Moore JH (2024) Centralized and federated models for the analysis of clinical data. Annu Rev Biomed Data Sci 7(1):179–199. https://doi.org/10.1146/annurev-biodatasci-122220-115746 PubMed DOI PMC

McMahan B, Moore E, Ramage D, Hampson S, Arcas BA (2017) Communication-efficient learning of deep networks from decentralized data. In: Singh A, Zhu J (eds) Proceedings of the 20th international conference on artificial intelligence and statistics. PMLR, Cambridge, pp 1273–1282

Teo ZL, Jin L, Li S et al (2024) Federated machine learning in healthcare: a systematic review on clinical applications and technical architecture. Cell Rep Med 5(2):101419. https://doi.org/10.1016/j.xcrm.2024.101419 PubMed DOI PMC

Crowson MG, Moukheiber D, Arévalo AR et al (2022) A systematic review of federated learning applications for biomedical data. PLoS Digit Health 1(5):e0000033. https://doi.org/10.1371/journal.pdig.0000033 PubMed DOI PMC

Wolfson M, Wallace SE, Masca N et al (2010) DataSHIELD: resolving a conflict in contemporary bioscience–performing a pooled analysis of individual-level data without sharing the data. Int J Epidemiol 39(5):1372–1382. https://doi.org/10.1093/ije/dyq111 PubMed DOI PMC

Forsberg L, Hillert J, Mouresan EF, Glaser A, Magyari M, Pontieri L (2023) Assessing likelihood of secondary progressive multiple sclerosis in Sweden and Denmark: a federated learning analysis unveiling modestly higher odds in Sweden. In: MSMilan2023 – Paper Poster – Session 2. Mult Scler. 29(3_suppl):394–649. https://doi.org/10.1177/13524585231196194