Abstract P460: Genetic Risk for Obesity Across the Transition From Young to Older Adulthood: A Pathway Based Analysis

Background: Genetics and lifestyle behaviors are key contributors to obesity. However, few studies explicitly examine the differential genetic association with BMI across critical lifecycle periods between young adulthood and later adulthood. Hypothesis: We anticipate that genetic risk will be directly associated with BMI in young adulthood and there will be a strong indirect association with BMI later in life, as a result the increased risk of BMI associated with young adulthood. Methods: For 7,397 Chinese adult participants (age 18 to 65y) of the China Health and Nutrition Survey (CHNS; 1991-2015), we calculated an obesity polygenic risk score (PRS) based on BMI-related SNPs selected from the largest BMI GWAS in 700,000 individuals of European ancestry from UK biobank and GIANT consortium, weighted by effect size estimated from the GWAS study. We used linear mixed models to estimate BMI across all periods of adulthood, adjusting for calendar time and sex. In a subsample of participants with three measurements across 24 years (n=1220), we used pathway based analysis to estimate the direct effects of genetic risk in association with BMI during young (18-35y), middle (35-45y), and later (45-65y) adulthood as well as the indirect genetic effects of at later adulthood through BMI at each of these earlier periods in life. Results: In the full sample, the linear mixed model-adjusted mean BMI (Standard Error (SE)) was 22.4 (0.04) kg/m 2 at young adulthood, 23.1 (0.04) kg/m 2 at middle adulthood and 23.2 (0.04) kg/m 2 at later adulthood, adjusting for sex and calendar year. In the subsample, a one standard deviation higher obesity PRS was directly associated with a 0.38 (0.07 SE) and 0.21 (0.06 SE) kg/m 2 higher BMI in young and middle adulthood, respectively. We found no evidence of a direct estimated effect of PRS in later adulthood (age 45-64y), but evidence of an indirect association between obesity PRS and BMI in later adulthood (age 45-64y). For example, we found a one standard deviation higher PRS was indirectly associated with 0.31 (0.06 SE) kg/m 2 higher BMI at age 35-45y through BMI at age 18-35y, and a 0.25 (0.05 SE) kg/m 2 higher BMI at age 45-65y through BMI at age 18-35y and subsequently BMI at age 35-45y. Conclusion: We observed a strong direct association between polygenic risk for obesity in young adulthood (age 18-35y), with persistence through age 35-45y and into age 45-65y.