Interpretable Machine Learning for Cross‐Cohort Prediction of Motor Fluctuations in Parkinson's Disease

• Published in: Movement disorders Vol. 40; no. 8; pp. 1604 - 1617
• Main Authors: Mangone, Graziella, Khoury, Fouad, Vidailhet, Marie, Corvol, Jean‐Christophe, Krüger, Rejko, Glaab, Enrico, Acharya, Geeta, Aguayo, Gloria, Alexandre, Myriam, Ali, Muhammad, Ammerlann, Wim, Arena, Giuseppe, Bassis, Michele, Batutu, Roxane, Beaumont, Katy, Béchet, Sibylle, Berchem, Guy, Bisdorff, Alexandre, Boussaad, Ibrahim, Bouvier, David, Castillo, Lorieza, Contesotto, Gessica, Dewitt, Brian, Diederich, Nico, Dondelinger, Rene, Ramia, Nancy E., Ferrari, Angelo, Frauenknecht, Katrin, Fritz, Joëlle, Gamio, Carlos, Ghosh, Soumyabrata, Giraitis, Marijus, Goergen, Martine, Lope, Elisa Gómez De, Graas, Jérôme, Graziano, Mariella, Groues, Valentin, Grünewald, Anne, Hammot, Gaël, Hanff, Anne‐Marie, Hansen, Linda, Heneka, Michael, Henry, Estelle, Henry, Margaux, Herzinger, Sascha, Hundt, Alexander, Jacoby, Nadine, Jónsdóttir, Sonja, Klucken, Jochen, Kofanova, Olga, Lambert, Pauline, Landoulsi, Zied, Lentz, Roseline, Lopes, Ana Festas, Lorentz, Victoria, Marques, Tainá M., Marques, Guilherme, Conde, Patricia Martins, Mcintyre, Deborah, Mediouni, Chouaib, Meisch, Francoise, Minelli, Maura, Mittelbronn, Michel, Nehrbass, Ulf, Nickels, Sarah, Nicolai, Beatrice, Nicolay, Jean‐Paul, Pachchek, Sinthuja, Pauly, Claire, Pauly, Laure, Pavelka, Lukas, Perquin, Magali, Pexaras, Achilleas, Rauschenberger, Armin, Rawal, Rajesh, Bobbili, Dheeraj Reddy, Remark, Lucie, Richard, Ilsé, Roland, Olivia, Roomp, Kirsten, Rosales, Eduardo, Sapienza, Stefano, Satagopam, Venkata, Schmitz, Sabine, Severino, Raquel, Sharify, Amir, Soare, Ruxandra, Sokolowska, Kate, Theresine, Maud, Thien, Hermann, Thiry, Elodie, Loo, Rebecca Ting Jiin, Trouet, Johanna, Tsurkalenko, Olena, Vaillant, Michel, Vilas Boas, Liliana, Wilmes, Paul, Wollscheid‐Lengeling, Evi
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.08.2025; Wiley Subscription Services, Inc
• Subjects: Aged; Calibration; Cohort Studies; cross‐cohort analysis; Decision making; Female; Humans; Learning algorithms; longitudinal cohorts; Longitudinal Studies; LRRK2 protein; Machine Learning; Male; Middle Aged; motor fluctuations; Movement disorders; Neurodegenerative diseases; Parkinson Disease - complications; Parkinson Disease - diagnosis; Parkinson Disease - genetics; Parkinson Disease - physiopathology; Parkinson's disease; Predictions; predictive modeling; Prognosis; Quality of life; Tremor - etiology; Tremor - physiopathology; longitudinal cohorts; cross‐cohort analysis; machine learning; motor fluctuations; predictive modeling
• ISSN: 0885-3185; 1531-8257; 1531-8257
• DOI: 10.1002/mds.30223

Background Motor fluctuations are a common complication in later stages of Parkinson's disease (PD) and significantly affect patients' quality of life. Robustly identifying risk and protective factors for this complication across distinct cohorts could lead to improved disease management. Objectives The goal was to identify key prognostic factors for motor fluctuations in PD by using machine learning and exploring their associations in the context of the prior literature. Methods We applied interpretable machine learning techniques for time‐to‐event analysis and prediction of motor fluctuations within 4 years in three longitudinal PD cohorts. Prognostic models were cross‐validated to identify robust predictors, and the performance, stability, calibration, and utility for clinical decision‐making were assessed. Results Cross‐validation analyses suggest the effectiveness of the models in identifying significant baseline predictors. Movement Disorder Society‐Unified Parkinson's Disease Rating Scale parts I and II, freezing of gait, axial symptoms, rigidity, and pathogenic GBA and LRRK2 variants were positively correlated with motor fluctuations. Conversely, motor fluctuations were inversely associated with tremors and late age of onset of PD. Cross‐cohort data integration provides more stable predictions, reducing cohort‐specific bias and enhancing robustness. Decision curve and calibration analysis confirms the models' practical utility and alignment of predictions with observed outcomes. Conclusions Interpretable machine learning models can effectively predict motor fluctuations in PD from baseline clinical data. Cross‐cohort data integration increases the stability of selected predictors. Calibration and decision curve analyses confirm the model's reliability and utility for practical clinical applications. © 2025 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.