0661 Assessment Of A Genetic Risk Score For Prediction Of Restless Legs Syndrome In A Cohort Of Women. (12th April 2019)

• Record Type: Journal Article
• Authors: Daghlas, Iyas; Saxena, Richa; Chasman, Daniel I
• Is Part Of: Sleep. Volume 42(2019)Supplement 1
• Page Start: A263
• Page End: A264
• Publication Date: 2019-04-12
• Subjects: Sleep -- Physiological aspects -- Periodicals; Sleep disorders -- Periodicals; Sommeil -- Aspect physiologique -- Périodiques; Sommeil, Troubles du -- Périodiques; Sleep disorders; Sleep -- Physiological aspects; Sleep -- physiological aspects; Sleep Wake Disorders; Psychophysiology; Electronic journals; Periodicals; 616.8498
• DOI: 10.1093/sleep/zsz067.659
• Languages: English
• ISSNs: 0161-8105

Abstract

Abstract: Introduction: Restless legs syndrome (RLS) is a movement disorder that negatively impacts sleep and quality of life. Genome-wide association studies have identified numerous single-nucleotide-polymorphisms (SNPs) robustly associated with RLS. We aimed to characterize the association of an RLS genetic risk score (GRS) with RLS, and searched for environmental modifiers of genetic risk. Methods: Using questionnaire-defined RLS in the Women's Genome Health Study (WGHS), we tested the association of a weighted 20-SNP GRS with all cases of RLS (n=3, 254 cases/19, 173 controls) self-reported at 9 or 10 year follow-up, and with cases that were persistently reported across both questionnaires (n=833 cases/19, 173 controls). Putative RLS risk factors including body mass index (BMI), hormone replacement therapy, and smoking were investigated for GRS-environment interactions. Predictive metrics were derived from logistic models of persistent RLS using either the GRS or epidemiologic risk factors. Results: A 1-standard deviation (SD) increase in the RLS GRS had a strong effect on risk of any RLS (OR=1.38, 95% CI 1.33-1.43, p<2E -16 ), and risk of persistent RLS (OR=1.55, 1.45-1.66, p<2E -16 ). Compared to the bottom decile, the top decile of the RLS GRS was associated with a 3.5-fold increase in risk of RLS (OR=3.46, 2.18-4.14), and a 5.6-fold increase in risk of persistent RLS (OR=5.56, 3.82-8.10). A 1-SD increase in BMI attenuated genetic risk of RLS (pinteraction =.0093) such that amongst non-obese participants, the OR for RLS was 1.42 (1.35-1.48), and amongst obese participants the OR was 1.27 (1.18-1.37). The AUC for prediction of persistent RLS using the GRS alone was 0.63 [0.61-0.64], exceeding the AUC of a model combining age and epidemiologic risk factors [0.58, 0.56-0.60]. Results were unchanged when removing individuals with potential secondary causes of RLS. Conclusion: Genetic liability measured through a GRS is robustly associated with RLS, particularly for persistent cases and non-obese individuals, and outperforms epidemiologic risk factors in disease prediction. These results inform the use of genetics for personalized sleep medicine. Support (If Any): R01DK105072 (RS), R01DK107859 (RS), R21NS09296 (DC), R21NS104398 (DC)