Biogeographic Ancestry in the African Descent and Glaucoma Evaluation Study (ADAGES): Association With Corneal and Optic Nerve Structure

Abstract

Purpose: We determined if quantitative measurements of biogeographic ancestry (BGA) correlate with variations in optic disc area, corneal thickness (CCT), and retinal nerve fiber layer (RNFL) thickness.

Methods: Data were obtained from 656 participants in the African Descent and Glaucoma Evaluation Study (ADAGES) cohort who consented to BGA testing. Data for CCT, optic disc area, and RNFL thickness were obtained from subjects in the ADAGES study who also had participated in the current substudy. A total of 31 ancestry informative markers (AIMs) with large allele frequencies differences between populations was used to calculate admixture proportion (implemented in STRUCTURE). Correlations with BGA adjusted for diagnosis, age, and sex for CCT and optic disc area using the whole group and RNFL thickness adjusted for age and sex for the normal study participants were determined.

Results: The mean percentage of African admixture was 79.6% in the self-described African Descent (AD) group and 3.5% in the European Descent (ED) group. Percent African ancestry was significantly correlated with CCT (ρ = -0.27, P < 0.0001) and disc area (ρ = 0.15, P < 0.0001), but only marginally associated with RNFL thickness (ρ = 0.20, P = 0.092) in adjusted models.

Conclusions: The BGA correlates with variation in ocular features that significantly differ across racial groups and that have been associated with the development of glaucoma. While BGA can provide an objective measurement of the biologic component of self-described race for ocular research, for most nongenetic epidemiologic studies, self-described race may adequately describe the associations with these ocular characteristics. (ClinicalTrials.gov number, NCT00221923.).

Keywords: biogeographic ancestry; corneal thickness; optic disc; race; retinal nerve fiber layer.

Copyright 2015 The Association for Research in Vision and Ophthalmology, Inc.

Figures

Figure 1

Box and whisker plot and histograms showing the distribution of proportion of African ancestry determined by biogeographic testing for the AD (n = 297) and ED (n = 359) groups used in the analysis.

Figure 2

(a) Box and whisker plot of the distribution of corneal thickness for the AD and ED groups. (b) Scatterplot illustrating the association between the proportion of African ancestry and CCT for the entire cohort. Dark dots indicate ED group, light dots indicate AD group. The associations for self-report and BGA were highly significant (P < 0.0001 for both).

Figure 3

(a) Box and whisker plot of the distribution of optic disc area for the AD and ED groups. (b) Scatterplot illustrating the association between the proportion of African ancestry and optic disc area for the entire cohort. Dark dots indicate ED group, light dots indicate AD group. The associations for self-report and BGA were highly significant (P < 0.0001 for both).

Figure 4

(a) Box and whisker plot of the distribution of RNFL thickness for the AD and ED groups. (b) Scatterplot illustrating the association between the proportion of African ancestry and RNFL thickness for the entire normal group RNFL. Dark dots indicate ED group, light dots indicate AD group. In models adjusting for age and sex neither BGA (P = 0.092) nor self-report (P = 0.095) reached statistical significance.