Association of change on insulin-like growth factor (IGF)-i and IGF-binding protein 3 with genetic markers after a month of growth hormone (GH) therapy on Chinese children born with GH deficiency

Methods This phase IV open-label interventional study was performed on samples from 205 GHD children (175 subjects born AGA and 30 subjects SGA) of Chinese Han origin, recruited at 8 centers. All the subjects were given r-hGH for 4 weeks (0.033mg/kg/d).1536 SNPs were selected from 100 Candidate Genes involved in the GHIGFI axes, growth plate and other short stature-related diseases. We genotyped Single nucleotide polymorphism (SNP) using Illumina GoldenGateTM Assays. Linear regression was used to identify single SNPs significantly associated with changes, from baseline to week 4, in serum IGF-I SDS and IGFBP3 SDS. Gestational age (GA) was also included in the model. The significance threshold was set as a corrected P-value<0.05. Multiple linear regressions with interaction effect were used to identify significantly associated genes with changes, from baseline to week 4, in serum IGF-I SDS and IGFBP3 SDS. The significance threshold was set as a corrected P-value<0.1. Results (1) 6/19 SNPs which correspond to 4/17 genes were significantly associated with IGF-I SDS change/IGFBP3 SDS change separately from baseline to Week 4 in all samples (AGA+SGA) with a corrected P-value<0.05. Among them, 3/10 SNPs showed significant interaction effects, suggesting that the pattern of SNPs associated with IGF-I SDS change/IGFBP3 SDS change was different between two groups. (2) 14/14 genes which significantly associated with IGF-I SDS change/IGFBP3 SDS change at Week 4 were identified in all samples (AGA+SGA). Among them, 6/ 12 genes showed significant interaction effects, suggesting that IGF-I SDS change/IGFBP3 SDS change have different effect on these genes expression between two groups. We also found that 5 genes in common associated with both IGF-I SDS change and IGFBP-3 SDS change.