Age-Related Differences in the Association between REM Sleep and the Polygenic Risk for Parkinson's Disease

Parkinson's disease (PD) is one of the rare diseases in which sleep alteration is a true marker of disease outcome. Yet, how the association between sleep and PD emerges over the healthy lifetime is not established. We examined the association between the polygenic risk score (PRS) for PD and the variability in the electrophysiology of rapid eye movement (REM) sleep in 433 younger (18-31 years) healthy individuals and 85 late-midlife (50-69 years) healthy individuals.

In this prospective cross-sectional study, in-lab electroencephalography recordings of sleep were recorded to extract REM sleep metrics. PRS was computed using SBayesR approach.

Generalized additive model for location, scale, and shape analysis showed significant association of REM duration (p_corrected = 0.03) and theta energy in REM (p_corrected = 0.004) with PRS for PD in interaction with the age group. In the younger subsample, REM duration and theta energy were positively associated with PD PRS. In contrast, in the late-midlife subsample, the same associations were negative (although only qualitatively for REM theta energy) and may differ between men and women.

REM sleep is associated with the PRS for PD in early adulthood, 2 to 5 decades before typical symptoms onset. The association changes from positive in younger individuals, presumably free of alpha-synuclein, to negative in late-midlife individuals, possibly because of the progressive presence of alpha-synuclein aggregates or of the repeated increased oxidative metabolism imposed by REM sleep. Our findings may unravel core associations between PD and sleep and may contribute to novel intervention targets to prevent or delay PD. ANN NEUROL 2026;99:922-934.