GLP-1 receptor agonists are hypothesized to reduce obesity in part through appetite modulation, but the precise mechanisms remain unclear. We examined common variants from genes involved in GLP-1 signaling to assess their contribution to preadolescent appetitive behavior. Among 247 children aged 9-12 years of European ancestry from two studies with genotyping data, we selected 33 common single nucleotide polymorphisms (SNPs) from: 1) GCG (encoding GLP-1), 2) PCSK1 (GCG post-translational processing), 3) DPP4 (GLP-1-degrading enzyme), and 4) GLP-1R (GLP-1 receptor). We computed a composite score for parent-reported appetitive traits influencing food intake from the Child Eating Behavior Questionnaire (CEBQ) and applied quantitative multifactor dimensionality reduction (QMDR) to search for one- through four-way combinations among the candidate SNPs related to this outcome, adjusting for demographic and genetic covariates. We fit adjusted linear regression models to quantify variance in appetitive traits explained by the optimal one- through four-way combination models selected by QMDR. Permutation tests assessed this variance against a null distribution. The optimal genetic model from QMDR was a four-way combination (testing t = 3.95) including one SNP from each of GCG (rs13429709), PCSK1 (rs156003), DPP4 (rs17574), and GLP-1R (rs2235868). Adjusted linear regression identified substantial variance explained in the composite appetite score (adjusted R2 = 0.212). Under the permutation test, this was consistent with type I error < 0.05. These findings support the involvement of GLP-1 signaling in the endogenous control of appetite and are consistent with the clinical effects of GLP-1R agonists.
