The High-Risk Type 1 Diabetes HLA-DR and HLA-DQ Polymorphisms Are Differentially Associated With Growth and IGF-I Levels in Infancy: The Cambridge Baby Growth Study.

Abstract

This study explored the link between HLA polymorphisms that predispose to type 1 diabetes and birth size, infancy growth, and/or circulating IGF-I in a general population-based birth cohort. The Cambridge Baby Growth Study is a prospective observational birth cohort study that recruited 2,229 newborns for follow-up in infancy. Of these, 612 children had DNA available for genotyping single nucleotide polymorphisms in the HLA region that capture the highest risk of type 1 diabetes: rs17426593 for <i>DR4</i>, rs2187668 for <i>DR3</i>, and rs7454108 for <i>DQ8</i>. Multivariate linear regression models at critical ages (cross-sectional) and mixed-effects models (longitudinal) were performed under additive genetic effects to test for associations between HLA polymorphisms and infancy weight, length, skinfold thickness (indicator of adiposity), and concentrations of IGF-I and IGF-binding protein-3 (IGFBP-3). In longitudinal models, the minor allele of rs2187668 tagging <i>DR3</i> was associated with faster linear growth (<i>P</i> = 0.007), which was more pronounced in boys (<i>P</i> = 3 × 10^-7) than girls (<i>P</i> = 0.07), and was also associated with increasing IGF-I (<i>P</i> = 0.002) and IGFBP-3 (<i>P</i> = 0.003) concentrations in infancy. Cross-sectionally, the minor alleles of rs7454108 tagging <i>DQ8</i> and rs17426593 tagging <i>DR4</i> were associated with lower IGF-I concentrations at age 12 months (<i>P</i> = 0.003) and greater skinfold thickness at age 24 months (<i>P</i> = 0.003), respectively. The variable associations of <i>DR4</i>, <i>DR3</i>, and <i>DQ8</i> alleles with growth measures and IGF-I levels in infants from the general population could explain the heterogeneous growth trajectories observed in genetically at-risk cohorts. These findings could suggest distinct mechanisms involving endocrine pathways related to the HLA-conferred type 1 diabetes risk.