Evidence that emmetropization buffers against both genetic and environmental risk factors for myopia

Abstract

Purpose : The rise in the prevalence of myopia must be caused by altered exposure to one or more environmental risk factors. However genetics explains much of the inter-individual variation in refractive error, hence both genes and environment seem to play important roles. We tested the hypothesis that the effects of specific genetic and environmental risk factors vary between individuals who maintain emmetropia by the time they reached young adulthood vs. those who develop ametropia.

Methods : Participants of the population-based ALSPAC cohort were assessed using noncycloplegic autorefraction at ages 7-15 years-old. The number of myopic parents (0, 1 or 2), time spent reading and time spent outdoors were ascertained at age 8 years-old (parental questionnaire). A polygenic risk score was generated for 146 genetic variants associated with refractive error, identified by Tedja et al. (Nature Genetics 2018:50;834-848). The “effect size” associated with each risk factor was examined across quantiles of the refractive error distribution using conditional quantile regression (minimum N=2273).

Results : As expected given the increasing prevalence of myopia with age during childhood, all risk factors had larger effects when assessed at older ages. However, effect sizes were much smaller in individuals who remained emmetropic over childhood than those who developed myopia or hyperopia. This was true for both genetic and environmental risk factors. For instance at age 15, a high genetic risk (defined as a 1 standard deviation increase in the polygenic risk score) was associated with a -0.65 D (95% CI -0.50 to -0.80) more negative refractive error in individuals in the 0.05^th quantile, i.e. those in the cohort who developed most myopia, vs. a -0.09 D (95% CI -0.07 to -0.11) more negative refractive error in individuals in the 0.50^th quantile, i.e. those who remained emmetropic during childhood. The equivalent values for participants classified as spending a high vs. low amount of time reading were: 0.05^th quantile = -1.13 D (95% CI -0.61 to -1.64); 0.50^th quantile = -0.06 D (95% CI -0.02 to -0.10). Full results are shown in the Figure.

Conclusions : (1) These findings suggest emmetropization can buffer against both genetic and environmental risk factors for myopia. (2) There is a huge degree of inter-individual difference in susceptibility to risk factors for myopia to which conventional analysis methods are blind.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.