Medication Treatment Modulates Sleep Architecture and Cognitive-Related Cortical Hemodynamics in Narcolepsy Type 1

This exploratory pilot study aimed to evaluate the associations between combination medication treatment and changes in sleep architecture and cognitive task-related cortical hemodynamics in narcolepsy type 1 (NT1).

A total of 16 drug-naive NT1 patients were enrolled. All participants underwent polysomnography (PSG), multiple sleep latency test (MSLT), and fNIRS monitoring during Verbal Fluency Test (VFT) and Go/NoGo task before and after pharmacological intervention. Paired nonparametric tests were used to compare differences in PSG, MSLT, and fNIRS parameters before and after treatment. Least absolute shrinkage and selection operator (LASSO) regression and stepwise regression were applied to screen predictors of sleep latency.

REM latency and mean sleep latency from MSLT were significantly prolonged, and REM sleep proportion was significantly reduced after medication treatment. During VFT, significant changes in oxyhemoglobin (HbO₂) concentration were observed in 5 channels: increased HbO₂ levels in frontopolar area, middle temporal gyrus, and subcentral area, and decreased HbO₂ levels in retrosplenial cortex (BA 29/30) and dorsolateral prefrontal cortex. No significant HbO₂ changes were found during Go/NoGo task. Regression analysis identified that pre-treatment HbO₂ change of subcentral area and post-treatment HbO₂ change of middle temporal gyrus predicted post-treatment mean sleep latency from MSLT.

In this pilot cohort, combination medication therapy was associated with prolongation of sleep latency and task-specific alterations in cortical hemodynamics in NT1 patients. fNIRS-derived parameters show preliminary potential as non-invasive biomarkers for predicting post-treatment sleep latency, which requires validation in larger independent cohorts.