The neurophysiological brain-fingerprint of Parkinson’s disease

Abstract

Brain-fingerprinting is a neuroimaging approach that is expanding the neuroscientific perspective on inter-individual diversity in health and disease. In the present study, we used brain-fingerprinting to advance the neurophysiological characterization of Parkinson's disease (PD). We derived the brain-fingerprints of patients with PD and age-matched healthy controls from the rhythmic and arhythmic spectral features of brief and task-free magnetoencephalography recordings. Using this approach, the individual differentiation of patients against healthy controls is 81% accurate, with the differentiability of patients scaling with the severity of their cognitive and motor symptoms. We show that between-patient differentiation is more challenging (77% accurate) than between healthy controls (90%) because the neurophysiological spectral features of patients with PD are less stable over time. The most distinctive features for differentiating healthy controls map to higher-order regions in the brain functional hierarchy. In contrast, the most distinctive features for patient differentiation map to the somatosensori-motor cortex. We also report that patient brain-fingerprints coincide with the cortical topography of the neurotransmitter systems affected in PD. We conclude that Parkinson's disease affects the spectral brain-fingerprint of patients with remarkable heterogeneity between individuals, and increased variability over short periods of time, compared to age-matched healthy controls. Our study demonstrates the relevance of neurophysiological fingerprinting to clinical neuroscience, and highlights its potential in terms of patient stratification, disease modeling, and the development and evaluation of personalized interventions.

Alterations of Cortical Structure and Neurophysiology in Parkinson's Disease Are Aligned with Neurochemical Systems

“…Substantial evidence [64][65][66] supports the role of mu-opioid and endocannabinoid systems in depression and anxiety, indicating that a shift toward decreases in activity of these systems (relative to GABAergic and glutamatergic systems) might represent a compensatory or neuroprotective effect against psychiatric symptoms in PD. Our group has also shown that the neurophysiological fingerprint of patients with PD is less distinct in regions rich in mu-opioid and cannabinoid systems, 67 indicating that the moment-to-moment variability of neurophysiology may be contributing to this finding.…”

Genetic Foundations of Neurophysiological and Behavioural Variability Across the Lifespan

“…The discrepancies were particularly pronounced in the alpha band (8)(9)(10)(11)(12)(13), where MZ twins matched with an accuracy of 47.8% ([33.3, 66.7]) compared to only 1.2% for DZ twins ([0.0, 5.0]). In the beta band (13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30), accuracies were 52.1% for MZ twins ([33.3, 70.0]) versus 7.1% for DZ twins ([0.0, 20.0]).…”

“…Leveraging intra-class correlation (ICC) statistics (2,3,29) (see Methods), we found that cortical activity in the theta (4-8 Hz; average ICC = 0.74), alpha (8)(9)(10)(11)(12)(13) Hz; ICC = 0.83), and beta (13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) Hz; ICC = 0.74) frequency bands are the most distinctive traits of neurophysiological profiles between individuals (Figure 1C). These traits are heritable, with Falconer's method showing a mean heritability (H 2 ) of 0.85 for theta-band traits, 0.76 for alpha-band traits, and 0.77 for beta-band traits across the temporal, frontal, and occipital cortex (Figure S4A).…”

Associations between neuromelanin depletion and cortical rhythmic activity in Parkinson's disease

“…Novel neuromelanin-sensitive MRI measurements of SN and LC integrity hold promise for the non-invasive monitoring of the degeneration of dopaminergic and noradrenergic systems in PD 6,7 in relation to the respective hallmark motor and cognitive features of the disease 3,7 . A knowledge gap remains between these neurochemical observations and the well-documented alterations of rhythmic and arrhythmic neurophysiological activity in PD [8][9][10] , and whether any such associations relate to the cognitive and motor features of PD.…”

“…In particular, cortical alpha rhythms (8)(9)(10)(11)(12) are altered in PD, and, akin to the deterioration of the LC, relate to mild cognitive decline and PD dementia [11][12][13][14][15] . Whether the PD alterations of alpha activity are related to LC degeneration is an open question.…”