Objective: To determine the clinical subtypes of MSA using data-driven approach and to compare the differences of the subtypes on survival and bed-ridden rates.

Background: A high heterogeneity in the presentation and prognosis of Multiple system atrophy (MSA) was observed in clinical practice, the subtyping of MSA using now may not be sufficient to present underlying mechanisms and future survival of the disease.

Method: A cohort of 193 MSA patients were recruited from Xuanwu hospital Capital Medical University as part of the CHINA cohort. A wide range of motor and nonmotor symptoms were evaluated at baseline. Data on demographics, motor symptoms and nonmotor manifestations were analyzed with k-means cluster analysis. After a mean period of 31.87 months, the Schwab and England Activities of Daily Living Scale (ADL), bedridden condition, and survival of 95 patients were followed by telephone. Kaplan-Meier analysis was used to calculate survival and cox regression was performed to identify factors associated with survival. The univariate and multivariable logistic regression analyses were used to identify factors associated with bedridden state in MSA patients.

Results: We identified four distinct subtypes: sleep disorder dominant, gait disturbance dominant, motor disorder dominant and cerebellar symptom dominant. Patients in sleep disorder dominant and cerebellar symptom dominant presented with milder motor features, whereas patients in gait disturbance dominant and motor disorder dominant had similar motor disability, but different non-motor presence. Median survival from symptom onset to death determined by Kaplan-Meier analysis was 6.25 years（95% CI 4.74-7.76). The multivariate cox analysis showed that cluster of gait disturbance dominant (hazard ratio [HR] 7.018, 95%CI 1.57–31.36; p=0.011) and motor disorder dominant (HR 10.01,95%CI 2.19–45.80; p=0.003) were associated with poorer survival, compared with sleep disorder dominant. After adjusting for disease duration and age, multivariable logistic regression showed motor disorder dominant had a higher risk to be bedridden, compared with cerebellar symptom dominant (odds ratio [OR]=19.03, 95%CI 1.74- 208.59; p=0.016).

Conclusion: We identified four different clinical clusters by cluster analysis in a large MSA cohort. There were significant differences between four new clinical subtypes in terms of prognosis, which might indicate differences in the underlying mechanism.

« Back to 2023 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/data-driven-subtypes-of-multiple-system-atrophy-and-the-implication-on-prognosis/