Potential Fluid Biomarkers and a Prediction Model for Better Recognition Between Multiple System Atrophy-Cerebellar Type and Spinocerebellar Ataxia

The prediction model that was developed based on laboratory and demographic variables obtained from ataxia patients at admission to the hospital might help improve the ability to differentiate MSA-C from spinocerebellar ataxia clinically.

We established a retrospective cohort with 125 ataxia patients from southwest China between April 2018 and June 2020. Demographic and laboratory variables obtained at the time of hospital admission were screened using Least Absolute Shrinkage and Selection Operator (LASSO) regression and logistic regression to construct a diagnosis score. The receiver operating characteristic (ROC) and decision curve analyses were performed to assess the accuracy and net benefit of the model. Also, independent validation using 25 additional ataxia patients was carried out to verify the model efficiency. Then the model was translated into a visual and operable web application using the R studio and Shiny package.

This study screened potential fluid biomarkers and developed a prediction model based on the easily obtained information at initial inspection to identify ataxia patients more likely to have multiple system atrophy-cerebellar type (MSA-C).

From 47 indicators, five variables were selected and integrated into the prediction model, including the age of onset (AO), direct bilirubin (DBIL), aspartate aminotransferase (AST), eGFR, and synuclein-alpha. The prediction model exhibited an area under the curve (AUC) of 0.929 for the training cohort and an AUC of 0.917 for the testing cohort. The decision curve analysis (DCA) plot displayed a good net benefit for this model, and external validation confirmed its reliability. The model also was translated into a web application that is freely available to the public.