Purpose
Common genetic variants have recently been associated with myopia and axial length. CREAM and 23andMe identified 42 genes associated with refractive error, axial length, and/or myopia using genome-wide association studies. The age of onset is currently unknown for these genes. We aimed to compare the effect of the individual risk variants on ocular biometry in young children versus adults.
Methods
In this study, we evaluated the effect sizes of the 42 genes on axial length (AL) and corneal curvature (CC) in the Generation R birth cohort (mean age 6 (±0.5 yrs), and in the population-based Rotterdam Studies(RSI-III; 45+ years). In the Generation R cohort, data were available on AL (N=3880) and corneal curvature (N=3795). In RSI-III, data were available on SE (N=10,774), AL (N=2,471) and CC (N=10,753). Linear regression was performed using each SNP as a determinant and AL and CC as outcomes, adjusting for age, sex and height. We compared effect sizes (betas) between the two study populations.
Results
For AL, seven of the previously identified genes (LAMA2, GJD2, C3orf26, ZC3H11B, CACNA1D, BMP4 and RSPO1) were significant in both study cohorts (P<0.05). The effect sizes of these seven genes were larger in adults (Figure 1). Four genes (KCNQ5, SHISA6, DLX1, DLG2) were only significant in Generation R (P<0.05). For CC, three genes (RSPO1, BMP4 and BICC1) were significantly associated in both study populations (figure 2). Effect size for RSPO1 increases, for BMP4 decreases and for BICC1 reverses.
Conclusions
Some genes associated with refractive error in adults already have a significant effect on ocular biometry in young children. The genes SHISA6 and, DLG2 appear to have all their effect on AL early in life. Most genes (LAMA2, GJD2) have an early effect which progresses with age, and some genes (CD55, CHRNG) have no effect at all yet at 6 years. Our results provide insights on the age period in which myopia genes exert their effect.
Keywords: 605 myopia •
533 gene/expression •
497 development