Contribution of a genetic risk score to clinical prediction of hepatic steatosis in obese children and adolescents.

Nonalcoholic fatty liver disease (NAFLD) is the commonest liver disease in children and adolescents in Western countries. Complex traits arise from the interplay between environmental and genetic factors in the pathogenesis of NAFLD. We examined the association between NAFLD and eleven single nucleotide polymorphisms (SNPs) at genetic loci potentially associated with liver damage (GCKR, MBOAT7, GPR120), oxidative stress (SOD2), lipid metabolism (PNPLA3, TM6SF2, LPIN1, ELOVL2, FADS2, MTTP) and fibrogenesis (KLF6) in a paediatric population. A genetic risk score (GRS) was performed taking into account both these SNPs and clinical risk factors. We recruited a cohort of 514 obese children and adolescents (mean age [±SD]: 11.2 ± 2.8 years, z-BMI 3.3 ± 0.8). NAFLD was identified by ultrasonography. Genotyping was performed by TaqMan-based RT-PCR system. The overall prevalence of NAFLD was 67.5% (347 patients). Among the eleven genotyped SNPs, the genetic variants in TM6SF2 rs58542926 (OR = 4.13, p = 0.002), GCKR rs1260326 (OR = 1.53, p = 0.003), PNPLA3 rs738409 (OR = 1.58, p = 0.004) and ELOVL2 rs2236212 (OR = 1.34, p = 0.047) were significantly associated with a higher risk of NAFLD. Addition of a 11-polymorphism GRS to established clinical risk factors significantly (albeit modestly) improved the discriminatory capability of the regression model for predicting the risk of NAFLD (with SNPs C-statistic 0.81 [95%CI 0.75–0.88] vs. 0.77 [0.70–0.84] without SNPs; p = 0.047). NAFLD was strongly associated with three genetic variants, TM6SF2 rs58542926, PNPLA3 rs738409 and GCKR rs1260326, and more slightly with ELOVL2 rs2236212, in obese children and adolescents. Addition of a 11-polymorphism GRS to clinical risk factors improved the predictability of NAFLD. [ABSTRACT FROM AUTHOR]