[en] Background
Prolonged levodopa treatment in Parkinson's disease (PD) often leads to motor complications, including levodopa-induced dyskinesia (LID). Despite continuous levodopa treatment, some patients do not develop LID symptoms, even in later stages of the disease.
Objective
This study explores machine learning (ML) methods using baseline clinical characteristics to predict the development of LID in PD patients over four years, across multiple cohorts.
Methods
Using interpretable ML approaches, we analyzed clinical data from three independent longitudinal PD cohorts (LuxPARK, n = 356; PPMI, n = 484; ICEBERG, n = 113) to develop cross-cohort prognostic models and identify potential predictors for the development of LID. We examined cohort-specific and shared predictive factors, assessing model performance and stability through cross-validation analyses.
Results
Consistent cross-validation results for single and multiple cohort analyses highlighted the effectiveness of the ML models and identified baseline clinical characteristics with significant predictive value for the LID prognosis in PD. Predictors positively correlated with LID include axial symptoms, freezing of gait, and rigidity in the lower extremities. Conversely, the risk of developing LID was inversely associated with the occurrence of resting tremors, higher body weight, later onset of PD, and visuospatial abilities.
Conclusions
This study presents interpretable ML models for dyskinesia prognosis with significant predictive power in cross-cohort analyses. The models may pave the way for proactive interventions against dyskinesia in PD by optimizing levodopa dosing regimens and adjunct treatments with dopamine agonists or MAO-B inhibitors, and by employing non-pharmacological interventions such as dietary adjustments affecting levodopa absorption for high-risk LID patients.
Research center :
Luxembourg Centre for Systems Biomedicine (LCSB): Biomedical Data Science (Glaab Group)
Disciplines :
Neurology Life sciences: Multidisciplinary, general & others Human health sciences: Multidisciplinary, general & others Biotechnology
Author, co-author :
Loo, R.T.J.
Tsurkalenko, O.
Klucken, J.
Mangone, G.
Khoury, F.
Vidailhet, M.
Corvol, J.-C.
Krüger, R.
GLAAB, Enrico ; University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > Biomedical Data Science
External co-authors :
yes
Language :
English
Title :
Levodopa-induced dyskinesia in Parkinson's disease: Insights from cross-cohort prognostic analysis using machine learning
Kwon, D.K., Kwatra, M., Wang, J., Ko, H.S., Levodopa-induced dyskinesia in Parkinson's disease: pathogenesis and emerging treatment strategies. Cells, 11(23), 2022, 3736, 10.3390/cells11233736.
Hutny, M., Hofman, J., Klimkowicz-Mrowiec, A., Gorzkowska, A., Current knowledge on the background, pathophysiology and treatment of levodopa-induced dyskinesia - literature review. J. Clin. Med., 10(19), 2021, 4377, 10.3390/jcm10194377.
Tirozzi, A., Modugno, N., Palomba, N.P., Ferese, R., Lonbardi, A., Olivola, E., et al. Analysis of genetic and non-genetic predictors of levodopa induced dyskinesia in Parkinson's disease. Front. Pharmacol., 12, 2021, 10.3389/fphar.2021.640603.
You, H., Mariani, L.L., Mangone, G., Le Febvre de Nailly, D., Charbonnier-Beaupel, F., Corvol, J.C., Molecular basis of dopamine replacement therapy and its side effects in Parkinson's disease. Cell Tissue Res. 373 (2018), 111–135, 10.1007/s00441-018-2813-2.
Cilia, R., Akpalu, A., Sarfo, F.S., Cham, M., Amboni, M., Cereda, E., et al. The modern pre-levodopa era of Parkinson's disease: insights into motor complications from sub-Saharan Africa. Brain 137:10 (2014), 2731–2742, 10.1093/brain/awu195.
Bezard, E., Brotchie, J.M., Gross, C.E., Pathophysiology of levodopa-induced dyskinesia: potential for new therapies. Nat. Rev. Neurosci. 2 (2001), 577–588, 10.1038/35086062.
Keun, J.T.B., Arnoldussen, I.A., Vriend, C., van de Rest, O., Dietary approaches to improve efficacy and control side effects of levodopa therapy in Parkinson's disease: a systematic review. Adv. Nutr. 12:6 (2021), 2265–2287, 10.1093/advances/nmab060.
Hornung, R., Causeur, D., Bernau, C., Boulesteix, A.L., Improving cross-study prediction through addon batch effect adjustment or addon normalization. Bioinformatics 33:3 (2017), 397–404, 10.1093/bioinformatics/btw650.
Freund, Y., Schapire, R.E., A decision-theoretic generalization of on-line learning and an application to boosting. J. Comput. Syst. Sci. 55:1 (1997), 119–139, 10.1006/jcss.1997.1504.
Quinlan, J.R., Improved use of continuous attributes in C4.5. J. Artif. Intell. 4 (1996), 77–90, 10.1613/jair.279.
Tan, Y.S., Singh, C., Nasseri, K., Agarwal, A., Duncan, J., Ronen, O., et al. Fast interpretable greedy-tree sums. Preprint at arXiv preprint, arXiv:2201.11931, 2022, 10.48550/arXiv.2201.11931.
McTavish, H., Zhong, C., Achermann, R., Karimalis, I., Chen, J., Rudin, C., et al. Fast sparse decision tree optimization via reference ensembles. Preprint at arXiv preprint, arXiv:2112.00798 https://doi.org/10.48550/arXiv.2112.00798, 2022.
Friedman, J.H., Greedy function approximation: a gradient boosting machine. Ann. Stat. 29:5 (2001), 1189–1232 http://www.jstor.org/stable/2699986.
Agarwal, A., Tan, Y.S., Ronen, O., Singh, C., Yu, B., Hierarchical Shrinkage: improving the accuracy and interpretability of tree-based methods. Preprint at arXiv preprint, arXiv:2202.00858, 2022, 10.48550/arXiv.2202.00858.
Bertsimas, D., Dunn, J., Gibson, E., Orfanoudaki, A., Optimal survival trees. Mach. Learn. 111 (2022), 2951–3023, 10.1007/s10994-021-06117-0.
He, K., Li, Y., Zhu, J., Liu, H., Lee, J.E., Amos, C.I., et al. Component-wise gradient boosting and false discovery control in survival analysis with high-dimensional covariates. Bioinformatics 32:1 (2016), 50–57, 10.1093/bioinformatics/btv517.
Ishwaran, H., Kogalur, U.B., Blackstone, E.H., Lauer, M.S., Random survival forests. Ann. Appl. Stat. 2:3 (2008), 841–860, 10.1214/08-AOAS169.
Lundberg, S.M., Lee, S.I., A unified approach to interpreting model predictions. Proceedings of the Advances in Neural Information Processing Systems, 2017, 4765–4774.
Sun, X., Xu, W., Fast implementation of DeLong's algorithm for comparing the areas under correlated receiver operating characteristic curves. IEEE Signal Process. Lett. 21:11 (2014), 1389–1393, 10.1109/LSP.2014.2337313.
Kang, L., Chen, W., Petrick, N.A., Gallas, B.D., Comparing two correlated C indices with right-censored survival outcome: a one-shot nonparametric approach. Stat. Med. 34:4 (2015), 685–703, 10.1002/sim.6370.
Piovani, D., Sokou, R., Tsantes, A.G., Vitello, A.S., Bonovas, S., Optimizing clinical decision making with decision curve analysis: insights for clinical investigators. Healthcare, 11(16), 2023, 2244, 10.3390/healthcare11162244.
Severson, K.A., Chahine, L.M., Smolensky, L.A., Dhuliawala, M., Frasier, M., Ng, K., et al. Discovery of Parkinson's disease states and disease progression modelling: a longitudinal data study using machine learning. The Lancet Digital Health 3:9 (2021), e555–e564, 10.1016/S2589-7500(21)00101-1.
Santos-Lobato, B.L., Schumacher-Schuh, A.F., Rieder, C.R.M., Hutz, M.H., Borges, V., Ferraz, H.B., et al. Diagnostic prediction model for levodopa-induced dyskinesia in Parkinson's disease. Arq. Neuro. Psiquiatr. 78:4 (2020), 206–216, 10.1590/0004-282x20190191.
Chen, Z., Li, G., Liu, J., Autonomic dysfunction in Parkinson's disease: implications for pathophysiology, diagnosis, and treatment. Neurobiol. Dis., 134, 2020, 104700, 10.1016/j.nbd.2019.104700.
Tan, A.H., Chuah, K.H., Beh, Y.Y., Schee, J.P., Mahadeva, S., Lim, S.Y., Gastrointestinal dysfunction in Parkinson's disease: neuro-gastroenterology perspectives on a multifaceted problem. Journal of Movement Disorders 16:2 (2023), 138–151, 10.14802/jmd.22220.
Navailles, S., De Deurwaerdère, P., Contribution of serotonergic transmission to the motor and cognitive effects of high-frequency stimulation of the subthalamic nucleus or levodopa in Parkinson's disease. Mol. Neurobiol. 45 (2012), 173–185, 10.1007/s12035-011-8230-0.
Luca, A., Monastero, R., Baschi, R., Cicero, C.E., Mostile, G., Davi, M., et al. Cognitive impairment and levodopa induced dyskinesia in Parkinson's disease: a longitudinal study from the PACOS cohort. Sci. Rep., 11, 2021, 867, 10.1038/s41598-020-79110-7.
Chung, S.J., Yoo, H.S., Lee, H.S., Lee, Y.H., Baik, K., Jung, J.H., et al. Baseline cognitive profile is closely associated with long-term motor prognosis in newly diagnosed Parkinson's disease. J. Neurol. 268 (2021), 4203–4212, 10.1007/s00415-021-10529-2.
Thanprasertsuk, S., Phowthongkum, P., Hopetrungraung, T., Poorirerngpoom, C., Sathirapatya, T., Wichit, P., et al. Levodopa-induced dyskinesia in early-onset Parkinson's disease (EOPD) associates with glucocerebrosidase mutation: a next-generation sequencing study in EOPD patients in Thailand. PLoS One, 18(10), 2023, 0293516, 10.1371/journal.pone.0293516.
Zhang, Z., Liu, G., Wang, D., Chen, H., Su, D., Kou, W., et al. Effect of onset age on the levodopa threshold dosage for dyskinesia in Parkinson's disease. Neurol. Sci. 43 (2022), 3165–3174, 10.1007/s10072-021-05694-1.
Fröhlich, H., Bontridder, N., Petrovska-Delacréta, D., Glaab, E., Kluge, F., Yacoubi, M.E., et al. Leveraging the potential of digital technology for better individualized treatment of Parkinson's disease. Front. Neurol., 13, 2022, 788427, 10.3389/fneur.2022.788427.
Gerke, S., Minssen, T., Cohen, G., Ethical and legal challenges of artificial intelligence-driven healthcare. Artificial Intelligence in Healthcare, 2020, 295–336, 10.1016/B978-0-12-818438-7.00012-5.
Luo, Y., Chen, H., Gui, M., Radiomics and hybrid models based on machine learning to predict levodopa-induced dyskinesia of Parkinson's disease in the first 6 years of levodopa treatment. Diagnostics, 13(15), 2023, 2511, 10.3390/diagnostics13152511.
