The Pathophysiological Correlates of Parkinson’s Disease Clinical Subtypes

Daniele Belvisi, A. Fabbrini, M.I. De Bartolo, M. Costanzo, N. Manzo, G. Fabbrini, G. Defazio, A. Conte and A. Berardelli

Movement Disorders, Vol. 36, No. 2, 2021

Selezionato dal Lettore: Federico Emanuele Pozzi – Specializzando in Neurologia all’Università degli Studi Milano-Bicocca – Ospedale San Gerardo, Monza

MOTIVATION: Parkinson disease (PD) is a highly heterogenous condition, which has recently divided in different subtypes using a data-driven clustering techniques; the most recent studies have identified two of these subtypes at disease onset - a mild motor-predominant subtype and a diffuse malignant subtype (characterized by the coexistence of severe motor and nonmotor symptoms), the latter showing a faster progression. No study so far have elucidated whether the neurophysiological abnormalities described in PD differ between these forms; therefore, the authors assessed neurophysiological parameters previously used to evaluate sensorimotor circuits in 100 de novo PD patients, namely primary motor cortex excitability and plasticity with TMS, kinematic analysis of voluntary movement, somatosensory temporal discrimination threshold (STDT) and STDT movement-induced modulation, with the hypothesis that clinical heterogeneity could reflect different pathophysiological mechanisms in early stages of the disease.

The diagnosis was made based on clinical criteria, with the risk of misdiagnosis (58% of patients had a diagnosis of only clinically probable PD); hierarchical cluster analysis differentiated between a mild motor-predominant subtype and a diffuse malignant cluster (with older patients with worse cognition, motor scores as expressed by MDS-UPDRS).

The authors found that M1 excitability was higher in the diffuse malignant subtype, with steeper input/output curves (which represent the increase in MEP amplitude with increasing TMS intensity, which correlates with far or less excitable neurons recruitment); moreover, intermittent theta burst stimulation (iTBS) response (in terms of MEP amplitude changes) was significantly reduced in this group. This last parameter, which resembles long-term potentiation mechanisms and reflects M1 plasticity, was found to be the best discriminant between the two subtypes in multivariate logistic regression analysis. Finally, the author reported slower execution of voluntary movement (tested by repetitive fast index finger abduction) in the diffuse malignant subtype. Sensory function and sensorimotor integration didn’t differ between the two groups. The author argued that these findings reflect a greater extent of dopaminergic loss and its effect on the cortico-basal-thalamic-cortical motor loop in the diffuse malignant subtype; increase excitability of M1 would be due to a compensatory mechanism for defective basal ganglia function, while reduced plasticity would be related to loss of dopamine-promoted NMDA-dependent LTP due to the effects of dopaminergic denervation of M1. Finally, both parameters correlate with bradykinesia features, which are more pronounced in case of greater dopamine striatal depletion, linking bradykinesia and M1 abnormalities to the same underlying pathophysiological mechanism in the early stages of disease in the diffuse malignant subtype. Therefore, the authors suggest that different degrees of dysfunction of the aforementioned motor loop account for PD heterogeneity.

However, the study has some limitations, namely the risk of misdiagnosis, the fact that other systems involvement (such as the autonomic nervous system) was not studied, and that no DATSCAN or FDG PET was executed. With these caveats, this study provides the first evidence that PD subtyping is not just a clinical classification, but that specific pathophysiological mechanism and neurophysiological abnormalities underlie this division, which could help in the search for new targeted therapies. Moreover, they demonstrated that motor and sensory systems have different roles in the early stages of PD, the latter not being involved in the beginning. Finally, this study gives to basis for further longitudinal studies, which will help understanding how these mechanisms work in the more advanced stages of PD.