Large Multicohort Study Repositions Parkinson Disease as a Potential Network Disorder

A large multimodal imaging study published in Nature reframes Parkinson disease (PD) as a disorder of a distributed somato-cognitive action network (SCAN), rather than solely a dysfunction of classic motor circuits. Using a dataset comprising 863 individuals across multiple cohorts and interventions, the study authors report that hyperconnectivity (ie, increased functional coupling across brain regions) between the SCAN and subcortical structures is a central feature of PD pathophysiology and therefore a potential therapeutic target.

The research team analyzed resting-state functional connectivity (ie, the degree to which different brain regions show synchronized fluctuations when an individual is at rest) in a precision imaging PD cohort (n=166 individuals with PD; n=60 healthy controls) and a deep brain stimulation (DBS) sweet spot dataset (n=342), along with additional movement disorder cohorts, including essential tremor (ET; n=45), dystonia (n=42 with dystonia; n=21 controls), and amyotrophic lateral sclerosis (ALS; n=30 with ALS; n=30 controls). Across analyses, the substantia nigra and established DBS targets—the subthalamic nucleus (STN), globus pallidus internus, and ventral intermediate thalamus—were selectively connected to the SCAN rather than to effector-specific motor regions.

Key findings include the following:

• Individuals with PD demonstrated significantly increased resting-state functional connectivity between the SCAN and subcortical nuclei compared with healthy controls (2-tailed independent t=3.2; P=.002).
• SCAN hyperconnectivity correlated with motor severity (MDS-UPDRS-III; r=0.162, P=.037), cognitive performance (MMSE; r=0.161, P=.038), anxiety (HAMA; r=−0.186, P=.017), and depression (HAMD; r=−0.177, P=.023).
• SCAN abnormalities were not observed in ET, dystonia, or ALS cohorts, suggesting disease specificity.
• In a longitudinal STN-DBS cohort (n=4), motor scores improved significantly over 12 months (F=15.71, P<.0001), accompanied by reductions in SCAN–subcortical hyperconnectivity (F=4.25, P=0.006).
• Reduction in SCAN connectivity after DBS correlated with motor improvement (F=6.86, P=.013).
• In a levodopa challenge test cohort (n=21), motor symptoms improved (1-tailed paired t=7.18, P < 0.0001) and SCAN hyperconnectivity decreased (t = 3.58, P = 0.001).
• In a randomized transcranial magnetic stimulation (TMS) trial (n=36), SCAN-targeted stimulation reduced MDS-UPDRS-III scores more than effector-targeted stimulation (−13.48 vs −6.49 points; P=.0003).
• SCAN-targeted TMS also significantly reduced SCAN hyperconnectivity (P=.020).
• In a focused ultrasound cohort (n=10), motor improvement correlated with proximity of the lesion to the SCAN-defined thalamic target (Spearman correlation, ρ=−0.68, P=.031).

Collectively, the findings suggest that PD may reflect dysfunction of a distributed SCAN—and that network-guided neuromodulation could improve targeting and clinical outcomes beyond traditional basal ganglia–centric approaches.

Source: Ren, J., Zhang, W., Dahmani, L. et al. Parkinson’s disease as a somato-cognitive action network disorder. Nature (2026). https://doi.org/10.1038/s41586-025-10059-1