Can genetic associations change with age? CFH and age-related macular degeneration

• Published in: Human molecular genetics 2012-12, Vol.21 (23), p.5229-5236
• Main Authors: ADAMS, Madeleine K. M, SIMPSON, Julie A, HOPPER, John, ROBMAN, Liubov D, BAIRD, Paul N, RICHARDSON, Andrea J, GUYMER, Robyn H, WILLIAMSON, Elizabeth, CANTSILIERIS, Stuart, ENGLISH, Dallas R, ZAW AUNG, Khin, MAKEYEVA, Galina A, GILES, Graham G
• Format: Article
• Language: English
• Subjects: Age Factors; Aged; Aged, 80 and over; Alleles; Biological and medical sciences; Case-Control Studies; Complement Factor H - genetics; Female; Fundamental and applied biological sciences. Psychology; Gene Frequency; Genetics of eukaryotes. Biological and molecular evolution; Genotype; Humans; Macular Degeneration - genetics; Male; Medical sciences; Middle Aged; Molecular and cellular biology; Odds Ratio; Ophthalmology; Retinopathies
• ISSN: 0964-6906; 1460-2083
• DOI: 10.1093/hmg/dds364

Genetic variation in the gene encoding complement factor H (CFH) on chromosome 1q31 has repeatedly been associated with an increased risk of age-related macular degeneration (AMD); however, previous studies have had inadequate numbers of participants across a sufficiently wide age range to determine whether the association varies by age. We conducted a genetic case-control study using data from 2294 cases and 2294 controls selected from the Melbourne Collaborative Cohort Study, matched on age, sex and region of origin. Four consistently replicated CFH single-nucleotide polymorphisms (SNPs) were genotyped: rs1061170 (Y402H), rs2274700, rs393955 and rs800292; their relationship with AMD prevalence was determined across the age range 48-86. A difference in genotype frequencies was seen across age groups, where the low-risk homozygote prevalence rose with each increasing age group. Associations with early AMD were strongly modified by age for three of the four SNPs (interaction P-value: 0.01-0.00003). An inverse association between the high-risk homozygote for each SNP and early AMD was observed in the younger age groups [odds ratios (OR) range 0.37-0.48 for age 75). The direction of associations for this gene change was from inverse to risk with increasing age. These findings have important implications for predictive models for AMD and potentially other age-related diseases which extrapolate risks from older cohorts, as they assume homogeneity of association by age, which might not exist.