Privacy-preserving synthetic data from a national CIDP registry: validation against real-world evidence

@article{signori2026privacypreserving,
  title = {{Privacy-preserving synthetic data from a national CIDP registry: validation against real-world evidence}},
  author = {Alessio Signori and Erika Schirinzi and Fiore Manganelli and Dario Cocito and Yuri Matteo Falzone and Chiara Briani and Anna Mazzeo and Angelo Schenone and Vincenzo Di Stefano and Giuseppe Cosentino and Girolama Alessandra Marfia and Luana Benedetti and Marinella Carpo and Massimiliano Filosto and Luca Leonardi and Marco Luigetti and Sabrina Matà and Giuseppe Piscosquito and Tiziana Rosso and Marta Lucchetta and Gabriele Siciliano and Giuseppe Lauria Pinter and Maurizio Inghilleri and Teresa Anna Cantisani and Francesca Notturno and Dario Ricciardi and Francesco Habetswallner and Elisa Vegezzi and Federica Ginanneschi and Antonio Todisco and Roberta Brusa and Stefano Sotgiu and Paolo Solla and Alberto De Lorenzo and Claudia Lozi and Alessandro Salvalaggio and Adele Ratti and Luca Gentile and Giorgia Mataluni and Filippo Brighina and Barbara Risi and Loris Poli and Francesca Forcina and Francesca Vitali and Federica Moret and Fabrizio Canale and Emanuele Cassano and Eleonora Dalla Bella and Stefania Casali and Vincenzo Di Lazzaro and Alessandro Cesare Prelle and Alessandra Carta and Filippo De Mattei and Giuseppe De Biasi and Eduardo Nobile-Orazio and Pietro Emiliano Doneddu},
  journal = {Frontiers in Neurology},
  year = {2026},
  month = oct,
  doi = {10.3389/fneur.2026.1839538},
  url = {https://www.frontiersin.org/articles/10.3389/fneur.2026.1839538/full},
  note = {CC BY 4.0},
}

TY  - JOUR
TI  - Privacy-preserving synthetic data from a national CIDP registry: validation against real-world evidence
AU  - Alessio Signori
AU  - Erika Schirinzi
AU  - Fiore Manganelli
AU  - Dario Cocito
AU  - Yuri Matteo Falzone
AU  - Chiara Briani
AU  - Anna Mazzeo
AU  - Angelo Schenone
AU  - Vincenzo Di Stefano
AU  - Giuseppe Cosentino
AU  - Girolama Alessandra Marfia
AU  - Luana Benedetti
AU  - Marinella Carpo
AU  - Massimiliano Filosto
AU  - Luca Leonardi
AU  - Marco Luigetti
AU  - Sabrina Matà
AU  - Giuseppe Piscosquito
AU  - Tiziana Rosso
AU  - Marta Lucchetta
AU  - Gabriele Siciliano
AU  - Giuseppe Lauria Pinter
AU  - Maurizio Inghilleri
AU  - Teresa Anna Cantisani
AU  - Francesca Notturno
AU  - Dario Ricciardi
AU  - Francesco Habetswallner
AU  - Elisa Vegezzi
AU  - Federica Ginanneschi
AU  - Antonio Todisco
AU  - Roberta Brusa
AU  - Stefano Sotgiu
AU  - Paolo Solla
AU  - Alberto De Lorenzo
AU  - Claudia Lozi
AU  - Alessandro Salvalaggio
AU  - Adele Ratti
AU  - Luca Gentile
AU  - Giorgia Mataluni
AU  - Filippo Brighina
AU  - Barbara Risi
AU  - Loris Poli
AU  - Francesca Forcina
AU  - Francesca Vitali
AU  - Federica Moret
AU  - Fabrizio Canale
AU  - Emanuele Cassano
AU  - Eleonora Dalla Bella
AU  - Stefania Casali
AU  - Vincenzo Di Lazzaro
AU  - Alessandro Cesare Prelle
AU  - Alessandra Carta
AU  - Filippo De Mattei
AU  - Giuseppe De Biasi
AU  - Eduardo Nobile-Orazio
AU  - Pietro Emiliano Doneddu
PY  - 2026
DA  - 2026/10/13
JO  - Frontiers in Neurology
DO  - 10.3389/fneur.2026.1839538
UR  - https://www.frontiersin.org/articles/10.3389/fneur.2026.1839538/full
AB  - Background: Research in chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is constrained by low prevalence, limited data sharing, and privacy regulations. Disease registries mitigate some of these barriers, yet the reuse of individual-level data remains challenging. Artificial intelligence– and regression-based synthetic data (SD) generation may offer a privacy-preserving solution, provided analytical validity is maintained. We aimed to validate SD derived from the Italian CIDP Registry using a published case–control study as a reference framework.  Methods: We analyzed a matched case–control dataset including 195 CIDP cases and 195 controls to examine lifestyle and dietary factors associated with CIDP risk. SD were generated using sequential conditional models stratified by sex. Fidelity was assessed by comparing real and SD in terms of marginal distributions, data distinguishability (propensity score mean squared error), and preservation of associations with CIDP risk using univariable and multivariable logistic regression.  Results: SD preserved the original case–control balance and sex distribution. Marginal distributions showed high concordance between real and SD, with minor smoothing in a small number of multi-level dietary variables. Distinguishability analyses indicated low separability between real and synthetic records. Univariable associations with CIDP risk were closely replicated, including the protective effects of higher fish and rice intake. Multivariable models demonstrated highly concordant effect estimates between datasets.  Conclusion: SD derived from the Italian CIDP Registry accurately reproduces key analytical properties of real-world data while preserving patient privacy. This approach enables reproducible and shareable CIDP research and may support future applications, including synthetic standard-of-care comparator cohorts and precision medicine–oriented investigations.
SN  - 1664-2295
ER  -