Objective: The objective of the present prospective study was to conduct latent class analysis (LCA) of a range of motor and nonmotor symptoms of MSA and thereby characterize latent homogenous subgroups of patients with early disease (<3 years from the time of motor symptom onset).

Background: Patients with multiple system atrophy (MSA) are classically divided into MSA-P (prominent parkinsonism) or MSA-C (prominent cerebellar ataxia) on the basis of their dominant motor phenotypes. However, the conventional classification into MSA-P and MSA-C heavily emphasized the initial motor symptoms of disease onset while paying little attention to other nonmotor features. In addition, several novel MSA phenotypes, such as non-motor variant MSA, benign MSA with long survival, “minimal change” MSA-P or MSA-C, MSA with corticobasal presentation, prodromal-premotor MSA, and mono-system atrophy, are recognized but cannot be readily fitted into the classical classification system.

Method: Sixty-one probable or possible MSA patients with disease durations of three years or less were included prospectively. LCAs with a range of MSA motor and nonmotor symptoms were conducted to identify latent MSA clinical subgroups, trialing from one to six latent class solutions based on fit indices and interpretability.

Results: The LCA outcome supported the categorization of at least three latent subgroups of patients with early-stage MSA: the largest class 1, which was labelled as “moderate parkinsonism + high dysautonomia,” included approximately half of our study patients showing moderate parkinsonism with a burden of autonomic dysfunction. The other two classes that were more intriguing were class 2 “high parkinsonism + low dysautonomia” and class 3 “high cerebellar symptoms + low dysautonomia,” which showed prominent core motor symptoms, including parkinsonism or cerebellar symptoms, although their dysautonomia was largely low.

Conclusion: We incorporated a variety of motor and nonmotor symptoms of the early phase of disease and characterized the parkinsonian or cerebellar subtypes with fewer autonomic symptoms. If such these statistically defined classes can be further validated in data-driven replication and the longitudinal studies, these findings will facilitate the definition of proper MSA subtype classification as well as investigations of subtype-specific treatment.

« Back to MDS Virtual Congress 2021

MDS Abstracts - https://www.mdsabstracts.org/abstract/data-driven-subtype-classification-of-patients-with-early-stage-multiple-system-atrophy/