A Prodromal Brain‐Clinical Pattern of Cognition in Synucleinopathies

Abstract

Objective Isolated (or idiopathic) rapid eye movement sleep behavior disorder (iRBD) is associated with dementia with Lewy bodies (DLB) and Parkinson's disease (PD). Biomarkers are lacking to predict conversion to a dementia or a motor‐first phenotype. Here, we aimed at identifying a brain‐clinical signature that predicts dementia in iRBD. Methods A brain‐clinical signature was identified in 48 patients with polysomnography‐confirmed iRBD using partial least squares between brain deformation and 27 clinical variables. The resulting variable was applied to 78 patients with iRBD followed longitudinally to predict conversion to a synucleinopathy, specifically DLB. The deformation scores from patients with iRBD were compared with 207 patients with PD, DLB, or prodromal DLB to assess if scores were higher in DLB compared to PD. Results One latent variable explained 31% of the brain‐clinical covariance in iRBD, combining cortical and subcortical deformation and subarachnoid/ventricular expansion to cognitive and motor variables. The deformation score of this signature predicted conversion to a synucleinopathy in iRBD (p = 0.036, odds ratio [OR] = 2.249; 95% confidence interval [CI] = 1.053–4.803), specifically to DLB (OR = 4.754; 95% CI = 1.283–17.618, p = 0.020) and not PD (p = 0.286). Patients with iRBD who developed dementia had scores similar to clinical and prodromal patients with DLB but higher scores compared with patients with PD. The deformation score also predicted cognitive performance over 1, 2, and 4 years in patients with PD. Interpretation We identified a brain‐clinical signature that predicts conversion in iRBD to more severe/dementing forms of synucleinopathy. This pattern may serve as a new biomarker to optimize patient care, target risk reduction strategies, and administer neuroprotective trials. ANN NEUROL 2021;89:341–357

Identification of REM Sleep Behavior Disorder by Magnetic Resonance Imaging and Machine Learning

“…The deformation of putamen and thalamus as well as the thinning of caudal middle frontal and paracentral cortices best distinguished between iRBD patients and controls. This is in line with previous studies performed on the same cohorts but with a single morphometric measure (12,15,16). Putamen dopamine depletion underlies parkinsonism (45) and the putamen is a region that is already structurally and functionally abnormal in iRBD (12,46–50), as observed by reduced dopamine reuptake imaging (5,47) and perfusion/metabolism imaging (48–50).…”

“…They were also subjected to the same exclusion criteria as iRBD patients. All these participants were part of previous studies on neuroimaging in iRBD (9,12,15,16,32).…”

“…Inclusion criteria were: (1) a diagnosis of idiopathic PD as the likeliest cause (30), (2) being 45-85 years old, (3) disease duration ≤10 years, (4) Hoehn & Yahr stage ≤3, (5) absence of dementia or any major psychiatric disorder, (6) respiratory event index ≤20, and (7) absence of a history of head injury, stroke, brain tumor, cerebrovascular disease, and chronic obstructive pulmonary disease or abnormal EEG features suggesting epilepsy. These patients were part of previous studies on neuroimaging in PD (16,33).…”

“…However, this study has a small sample size and no distinction of the cognitive profile. Given the ability of brain changes, mainly those associated with gray matter alterations, to identify MCI and predict phenoconversion in iRBD (15,16), an assessment of morphometric changes, along with a thorough evaluation of the discriminative power of single versus multiple brain region(s) in the identification and subtyping of iRBD, is of relevance. Moreover, machine learning has been applied to the extraction of relevant pre-clinical and clinical features, and can be used in inter-modality registration of different types of data (17,27), potentially enabling further improvements on the identification of iRBD.…”

Parkinson disease-associated cognitive impairment

“…Studies have identified a specific brainclinical pattern that identifies people with RBD who developed rapid cognitive decline and DLB rather than PD. Based on rou tine MRI using partial least squares, atrophy in the basal ganglia, thalamus, amygdala, and frontotemporal grey and white matter as well as the expansion of CSFfilled spaces predicted cognitive decline in both RBD and PD 262 . In addition to imaging, CSF and EEG biomarkers for cognitive impairment, there is an increasing focus on exploring αsynuclein and other biomarkers in other biofluids and tissues such as skin, colon, submandibular gland, CSF, saliva and blood 263 .…”