Associations between genetic polymorphisms of insulin-like growth factor axis genes and risk for age-related macular degeneration.

Purpose: To investigate whether insulin-like growth factor (IGF) axis genes, together with a novel dietary risk factor, the dietary glycemic index (dGI), and body mass index (BMI) affect the risk for age-related macular degeneration (AMD).
Methods: This case-control study involved 962 subjects originally recruited through the Age-Related Eye Disease Study (AREDS) Genetic Repository. After those with missing covariates or invalid calorie intake (n = 23), diabetes (n = 59), and non-Caucasian race (n = 16) were excluded, 864 participants were used, including 209 AREDS category 1 participants (control group), 354 category 2 or 3 participants (drusen group), and 301 category 4 participants (advanced AMD group). A total of 25 single-nucleotide polymorphisms (SNPs) selected from IGF-1 (n = 9), IGF-2 (n = 1), IGF binding protein 1 (IGFBP1; n = 3), IGFBP3 (n = 3), acid-labile subunit of IGFBP (IGFALS; n = 2), IGF1 receptor (IGF1R; n = 4), and IGF2R (n = 3) were genotyped. SNP-AMD associations were measured with genotype, allele χ(2) tests and Armitage's trend test. Odds ratios (OR), 95% confidence intervals (CIs), and SNP-exposure interactions were evaluated by multivariate logistic regression.
Results: One SNP (rs2872060) in IGF1R revealed a significant association with advanced AMD (P-allele = 0.0009, P-trend = 0.0008; the significance level was set at 0.05/25 = 0.002 for multiple comparisons). The risk allele (G) in the heterozygous and homozygous states (OR, 1.67 and 2.93; 95% CI, 1.03-2.71 and 1.60-5.36, respectively) suggests susceptibility and an additive effect on AMD risk. Further stratification analysis remained significant for both neovascularization (OR, 1.49 and 2.61; 95% CI, 0.90-2.48 and 1.39-4.90, respectively) and geographic atrophy (OR, 2.57 and 4.52; 95% CI, 0.99-6.71 and 1.49-13.74, respectively). The G allele interaction analysis with BMI was significant for neovascularization (P = 0.042) but not for geographic atrophy (P = 0.47). No significant interaction was found with dGI.
Conclusions: These data suggest a role of IGF1R on the risk for advanced AMD in this group of subjects.