Leveraging human genetics to overcome complex diagnostic challenges, evaluation of pan-ancestry polygenic scores to reduce misdiagnosis of narcolepsy and circadian rhythm sleep wake disorders.

点击翻译按钮获取中文摘要

Project Summary/Abstract One of the biggest challenges in rare disease is accuracy and timeliness of patient diagnosis. On average it takes 6 years for an accurate diagnosis, delaying treatment and creating substantial burden at the levels of individual, familial and healthcare systems with an estimated $750 billion spent on unnecessary procedures. Misdiagnosis also widens inequalities further, as misdiagnosis is more common among women and people of color. Therefore, there is an urgent need to improve diagnosis. Genetic risk predictions can improve diagnosis, with particular clinical utility in the specific setting where large barriers to diagnosis exist, such as rare sleep disorders. The rare sleep disorders narcolepsy and circadian rhythm sleep wake disorders face large hurdles to diagnosis, where 20% of primary care physicians are unaware that sleep medicine exists as a specialty, routine screening for sleep disorders is nearly absent, diagnostic tests require overnight visits to a limited number of specialized clinics, and insurance coverage for common diagnostic tests is non-existent. In contrast to the current diagnostic landscape for sleep disorders, genetic risk prediction is relatively inexpensive and easily accessible. In order to integrate genetic risk into the diagnosis and treatment pipeline, we must first have genetic predictors of risk applicable across multiple ancestry groups, or we risk widening inequities in healthcare further. To address the challenge of timely patient diagnosis in sleep disorders, we propose to leverage large exome sequence repositories to generate rare variant risk scores, expand the current known common polygenic scores to multiple genetic ancestries, and ultimately test the ability of both the rare and common polygenic scores to predict risk of rare sleep and circadian disorders in a large-scale hospital database with the goal to integrate flags in patient records.