Photoreceptor Structure in Patients with Increased Genetic Risk for Age-Related Macular Degeneration

Abstract

Purpose: Utilize high-resolution imaging to examine retinal anatomy in patients with increased genetic risk for developing AMD.

Methods: Thirty asymptomatic subjects (average age = 61.7 years, range = 51-67) were recruited. Genetic AMD relative risk (RR) scores were developed by comparing 1,021 AMD cases (all ages) from 439 controls (individuals over age 70 with recent comprehensive eye exams that did not note AMD). The final algorithm incorporated 5 of the 24 tested SNPs (rs10737680 and rs1329424 in CFH; rs3793917 in ARMS/HTRA1; rs641153 in C2/CFB; and rs493258 in LIPC) and distinguishes cases from controls with an AUC of 0.67. Patients in the high-risk quartile had a RR above 0.744, while those in the low risk quartile had a RR below 0.492. Fundus photographs were graded using the AREDS scale, except a grade of 0 was used if there was no evidence of AMD. SD-OCT was used to assess the integrity of retinal layers, and adaptive optics scanning light ophthalmoscopy (AOSLO) was used to evaluate the photoreceptor mosaic. Patients assigned a grade of zero were used as controls for comparing cone density measurements. All imaging was done masked to the genetic risk profile.

Results: Of the 30 patients, 17 were in the high-risk quartile, 12 were in the low-risk quartile, and one was in neither the high-risk nor low-risk quartiles. In the high-risk quartile, 10 had a fundus grade of 0, 5 were grade 1, one was grade 2, and one was grade 3. In the low-risk quartile, 9 were grade 0, and 3 were grade 1. In 12 subjects AOSLO imaging revealed photoreceptor changes ranging from variability in reflectivity to decreased density. Three subjects with a grade 1 fundus displayed areas of hyporeflectivity on the edges of drusen, likely due to altered waveguiding. There was no significant correlation between RR and fundus grade (Fisher’s exact test, p=0.45) or presence of cone mosaic abnormality (Fisher’s exact test, p=0.11).

Conclusions: Our data indicate the utility of adaptive optics in understanding early pathology in AMD. At this early stage of the disease, RR did not correlate with AMD-associated fundus changes or disruption of the photoreceptor mosaic on AOSLO. Since our patient population is nearly a decade younger than when AMD is clinically diagnosed, continued monitoring of these patients is needed to better understand the relationship between genetic risk and early anatomical changes.

Keywords: 412 age-related macular degeneration • 550 imaging/image analysis: clinical • 648 photoreceptors