Development and Validation of a Nocturnal Sleep Signature for the Diagnosis of Idiopathic Hypersomnia

Abstract Idiopathic hypersomnia (IH) is a chronic, disabling central nervous system disorder of hypersomnolence that typically begins in adolescence. Progress in understanding IH pathophysiology and identifying effective treatments is stymied by the commonly used diagnostic test, the multiple sleep latency testing (MSLT). This test is very reliable in classic narcolepsy but has poor validity and reliability in people with IH. In addition to this poor clinical performance, the current diagnostic use of the MSLT also hinders clinical trials by enrollment of clinically heterogeneous IH cohorts with more variable outcomes. Thus, there is a critical need to develop and validate an objective IH biomarker signature to improve diagnostic accuracy. The most striking IH features occur in the nocturnal sleep period. Most IH patients report non-restorative sleep and profound difficulty waking from sleep in the morning despite normal to long sleep durations. In our prior work, we showed IH is characterized by shorter (less stable) bouts of NREM 3 (N3) sleep (the most restorative sleep stage), longer bouts of less refreshing NREM 2 (N2) sleep, long sleep times, and high sleep efficiency. These findings may reflect proposed mechanisms of IH including heightened GABA receptor potentiation or prolonged circadian period. As a multi-disciplinary team with expertise in modeling neurophysiological processes, biomarker discovery, statistics, and hypersomnia disorders, we hypothesize that measures of nocturnal sleep stability will define the IH signature and more accurately identify IH than the current MSLT method. In the R61 grant, our objectives are to: (Aim 1) identify nocturnal sleep features of hypersomnia patients using polysomnograms retained in the National Sleep Research Resource; (Aim 2) develop and test a sleep biomarker signature from polysomnogram signals (IH signature) for IH diagnosis in a clinical population across 5 sites using current diagnostic criteria and cutoff scores on validated IH symptom scales as the “gold standard” IH classification. In meeting our robust “go criteria” showing high sensitivity and specificity of our IH signature for IH diagnosis, we will proceed with the R33 phase. In this next grant phase, we will validate the IH signature in a separate and new clinical cohort of patients (Aim 3) and test the reliability of the IH signature (Aim 4). Through this work, we will identify an accurate and stable biomarker signature that will provide more accurate diagnosis of IH using nocturnal PSG testing. This will improve diagnostic certainty in the clinic and reduce clinical heterogeneity for research, enabling study of underlying causal mechanisms and future drug development.