Résumé
Study Objectives : Cerebrospinal fluid (CSF) hypocretin-1/orexin-A quantification via radioimmunoassay (RIA) is used for diagnosing narcolepsy type 1(NT1), but its limitations require alternative methods. Using liquid chromatography-mass spectrometry (LC-MS), this study aimed to 1) fully characterize CSF orexin-A fragments detected by the gold standard RIA method, and 2) assess diagnostic relevance of measuring the most prevalent fragment, a 16-mer, in patients with NT1, narcolepsy type 2 (NT2), idiopathic hypersomnia (IH), and nonspecified hypersomnolence (NSH). Methods : CSF samples were analyzed using RIA and LC-MS in patients with hypersomnolence disorders evaluated at the French Narcolepsy National Reference Center. Fractionation techniques isolated orexin-A and its fragments, which were identified via targeted MS. Statistical analysis compared LC-MS performance against RIA. Results : CSF samples from 115 patients (54.8% females, mean age 30.4 +/- 15.8 years) including 52 with NT1, 6 NT2, 13 IH, and 44 NSH were analyzed by RIA and LC-MS. A 16-mer orexin-A fragment was identified and quantified by LC-MS, with lower levels in NT1. This fragment correlated strongly with RIA-measured orexin-A (r = 0.83, p < .0001) and demonstrated high sensitivity (98.1%) and specificity (85.7%) for diagnosis of NT1. Receiver operating characteristics analyses confirmed the high performance with an area under the curve of 0.975 and an optimal diagnostic cutoff of 10.67 pg/mL for LC-MS. Conclusions : The 16-mer orexin-A peptide represents a promising biomarker that could in the future help in the diagnosis of central hypersomnolence disorders. The transition from RIA to LC-MS methods for the quantification of 16-mer orexin-A represents a critical advance toward improving diagnostic accuracy and understanding orexinergic dysfunction in sleep disorders.