Abstract
Refractive errors, and especially myopia, are one of the leading causes of preventable visual impairment and blindness worldwide. The prevalence of myopia has increased significantly over the last three decades and is projected to affect 50% of the global population by 2050 (1). Even though previous work identified several risk factors, the pathophysiology of refractive errors is not yet fully understood. My work, presented as a thesis incorporating publications, examines associations between genetic and environmental factors and the population’s refractive error and myopia.The molecular mechanisms underlying refractive development were examined in the functional analyses incorporating the results of a meta-analysis of refractive error genome-wide association studies (GWAS) in Europeans. The results of post-GWAS analyses suggested that refractive error was a genetically heterogeneous trait, driven by genes involved in the development of all the different structures of the eye. Genetic factors associated with myopia are associated with neurodevelopment, ion trafficking that regulates cell polarity, but also control circadian rhythm, and are involved in melanin pigmentation and maintenance of intraocular pressure. The loci identified in the meta-analysis currently explain approximately 20% of refractive error heritability and polygenic models that were created as part of this thesis were able to predict myopia with considerable accuracy among Europeans as well as Singaporean Asian populations. In Europeans, identified variants predicted myopia with the area under the curve (AUC) of 0.74-0.75, while in Singaporean adolescent children prediction model incorporating associated SNPs and time outdoors showed AUC values of 0.59 and 0.72 for high and moderate myopia, respectively.
These GWAS common variant findings should not disguise the importance of rare variants that influence myopia as other complex traits. This thesis describes how for the first time rare variants in several distinct loci were associated with refractive error in the general population. Through my work, I identified rare variant associations in eight loci of which five were novel. The two most significant associations were identified for a gene encoding a transcription factor critical to the eye, retina and optic disc development and morphology (SIX6), and a novel candidate regulating photoreceptor specification and expression of many retinal genes (CRX). Other candidates were implicated in neurogenesis and neural signalling and were reported as associated with major psychiatric disorders.
Although GWAS and rare variant gene-based association analyses advanced the understanding of the molecular mechanisms behind lifelong risk to diseases of refraction, they contributed relatively a modest amount of knowledge to the understanding of the age-related specifics of refractive error genes. The age of refractive correction is one of the indicators of myopia progression and associated with the degree of refractive error in adulthood. In this thesis, I describe the largest scale genome-wide time-to-event analyses of self-reported age of first spectacle wear, performed on a sub-sample of UK Biobank participants of European descent. Although the self-reported age of refractive correction was strongly genetically correlated with refractive error, my work identified six loci that were novel and previously not described in relationship to refractive error, as well as showed good replication for many other, previously known loci. My results indicate that the effects of gene loci over phenotypic traits were not necessarily linear. These analyses demonstrated that, despite a trend for genes with higher effects to cause correction during early childhood years, there was considerable variation in the times when genes exerted maximal effects, which was not always proportional to the magnitude of their effect over refractive error.
Refractive errors arise from a mixture of gene effects and environment, while the first is constant and the latter has been changing and driving increasing prevalence rates of myopia. Several relatively well-studied socio-economic and cultural factors have been implicated in refractive error and the changes of their prevalence throughout the past seven decades. However, the structure of morbidity and medication has also changed considerably during the same time frame, in parallel to the same cultural and societal changes. The relationship between refractive errors and pharmacological treatments has been understudied. In this thesis, I describe the observations of the association between refractive error and medication-taking in a large population-based cohort. The most significant associations were found for 18 distinct pharmacological agents. Participants who reported taking anti-glaucoma preparations exhibited more negative, myopic refractive error compared to subjects who were not taking these medications, potentially due to the relationship between elevated intraocular pressure and myopic refraction. Interestingly, the group that was using pain control medications displayed a tendency towards a positive, more hypermetropic refractive error. Causal inference models show that multisite chronic pain, often associated with pain control medications may be at the origin of this relationship with refractive error.
The work presented in this thesis identified several novel factors associated with refractive error. The results described in this thesis, may enable the prediction of individual risk to refractive error and facilitate the development of optimised strategies for myopia prevention and treatment.
| Date of Award | 1 Jul 2022 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Pirro Hysi (Supervisor) & Mario Falchi (Supervisor) |