Abstract
Purpose :
The structural features of the eye with the greatest impact on refractive error are corneal curvature and axial length. Emmetropic eyes range in size both within and across species, yet possess an exquisite balance between corneal curvature and axial length; longer eyes have flatter corneas. Studies in humans, mice and chicks suggest this careful scaling of corneal curvature and axial length is under genetic control. However when myopia develops, the scaling goes awry; myopes typically have axially elongated eyes (1 mm lengthening being associated with ~3 Dioptres (D) myopia). Evidence that eye size prior to onset is a risk factor for myopia is conflicting.
Methods :
We conducted two genome-wide association studies (GWAS), firstly for corneal curvature in the emmetropic eyes of 22,180 UK Biobank participants, and secondly for refractive error in 95,505 unselected Biobank participants. Refractive error GWAS summary statistics were meta-analysed with those from the CREAM consortium (n=44,192). Mendelian randomisation analyses tested for a causal effect of eye size on refractive error. A genetic risk score comprised of the lead variants from the GWAS for corneal curvature was tested for association with corneal curvature and axial length in an independent sample of 307 unrelated emmetropic children aged 15 years (ALSPAC study).
Results :
The GWAS for corneal curvature identified 32 genetic loci (P<5.00E-08). A genetic risk score comprised of the 32 lead variants explained 2.3% (P=7.68E-03) and 2.7% (P=2.32E-03) of the variance in corneal curvature and axial length, respectively, in the independent ALSPAC sample, confirming these variants are predictive of eye size in emmetropes. By Mendelian randomisation analysis, the causal effect of eye size on refractive error was estimated as a +1.41 D (95% CI. 0.65 to 2.16) less myopic refractive error per mm longer axial length (P<0.001). Sensitivity analysis suggested no evidence of bias from directional pleiotropy.
Conclusions :
Our results do not support the hypothesis that a larger eye size is a risk factor for myopia. The findings also imply that the genetic determinants of normal eye size are not shared with those influencing susceptibility to myopia (e.g. those identified in a GWAS for refractive error).
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.