Increased body mass index is causally associated with pediatric MS onset: A Mendelian randomization study (P1.375)

Milena Gianfrancesco(University of California, Berkeley), Xiaorong Shao(University of California, Berkeley), Brooke Rhead(University of California, Berkeley), Jennifer Graves, Amy Waldman(Children's Hospital of Philadelphia), Timothy Lotze(Texas Children's Hospital), Teri Schreiner(University of Colorado Denver), Anita Belman(The Huntington Library, Art Museum, and Botanical Gardens), Benjamin Greenberg(Southwestern Medical Center), Bianca Weinstock‐Guttman(Jacobs Institute), Gregory Aaen(Loma Linda University), Jan‐Mendelt Tillema(Mayo Clinic in Arizona), Janace Hart, Jayne Ness, Yolanda Harris, Jennifer Rubin, Meghan Candee(University of Utah), Lauren Krupp(NYU Langone Health), Mark Gorman, Leslie Benson(Boston Children's Museum), Moses Rodriguez(Mayo Clinic in Arizona), Tanuja Chitnis(New England Institute of Art), Soe Mar(Washington University in St. Louis), Ilana Kahn(Children's National), John Rose, Shelly Roalstad, Charlie Casper, Ling Shen, Hong Quach, Catherine Metayer(University of California, Berkeley), Cathy Schaefer, Emmanuelle Waubant, Lisa F. Barcellos(Berkeley College)
Cited by 2

Abstract

Objective:We utilized Mendelian randomization to estimate the causal relationship between increasing body mass index (BMI) and pediatric MS susceptibility using a weighted BMI genetic risk score (BMI GRS). Background:Childhood and adolescent obesity are associated with a two-fold increased risk of pediatric and adult-onset MS. However, the relationship between obesity and MS may be confounded by lifestyle and/or socioeconomic factors that have not been appropriately adjusted for in previous studies. Methods:The BMI GRS incorporates the cumulative effect of 97 variants associated with BMI identified from the largest genome-wide association study (GWAS), to date (Locke et al, 2015). A split-sample instrumental variable (IV) analysis was used by summing all risk alleles multiplied by the estimated effect of each risk allele on the phenotype derived from the GWAS. Participants included non-Hispanic white pediatric MS cases and controls from over 15 sites across the U.S. (total sample size: 394 MS cases, 10,875 controls). Results:A significant association between BMI GRS and pediatric MS was demonstrated (causal odds ratio [OR] = 1.20, 95[percnt] CI 1.06, 1.36; p=0.004) after adjusting for sex, ancestry, HLA-DRB1*15:01 (the strongest genetic predictor of MS) and a weighted genetic risk score comprised of 110 non-HLA MS risk variants. Evidence of interaction between BMI GRS and HLA-DRB1*15:01 was also present (p-interaction=0.04); individuals carrying 1-2 DRB1* 15:01 risk alleles demonstrated a stronger association (OR=1.39, 95[percnt] CI 1.16, 1.37) compared to non-carriers (OR=1.05, 95[percnt] CI 0.88, 1.26). Conclusions:For the first time, we provide evidence supporting increased BMI is causally associated with pediatric MS and interacts with HLA-DRB1*15:01 status. Further, pediatric MS risk conferred by obesity may involve predisposing genetic factors for increased BMI, suggesting that specific inflammatory mechanisms involved in the obesity pathway may mediate disease onset.


Related Papers

No related papers found

Powered by citation graph analysis