Next Generation Sequencing in Diabetic Macular Edema and Proliferative Diabetic Retinopathy Patients Reveals Distinct Genetic Architectures: The DRGen Study

Abstract

Purpose : Our prior retrospective study has demonstrated a distinct phenotypic dichotomy within diabetic retinopathy (DR), with the majority of patients exhibiting either Proliferative Diabetic Retinopathy (PDR) without macular edema or Diabetic Macular Edema (DME) without any neovascularization. This observation suggests that PDR and DME may be distinct disease processes driven by independent molecular mechanisms and genetic factors.

Methods : We explored two distinct patient cohorts: Group 1 "PDR only" (without signs of DME) cohort (n = 182), and Group 2 "DME only" cohort (n = 77) (without signs of neovascularization). Ultrawide field retinal images were graded for DR severity at a masked central reading center by ETDRS standard grading protocol. We integrated data from both Whole Exome Sequencing (WES) and Whole Genome Sequencing (WGS) across both cohorts, enabling comprehensive exome-wide analyses of rare variants. We also utilized a newly developed Combined Functional Prediction (CFP) algorithm was used to evaluate the impact of coding variants in the cohort and integrated it with the phenotype-dependent gene variant prioritization utilizing Fisher's exact test.

Results : Analysis of "PDR only" and "DME only" patients for high-impact rare variants identified distinct genetic architecture between the two groups. The rare variant burden in PDR was significant compared to the DME cohort. Variants in the "PDR only" cohort were significantly enriched for pathways linked to CTL-mediated apoptosis, LDL Oxidation in Atherogenesis, and Interferon Gamma Signaling. Notably, phenotype-dependent gene variant prioritization analysis identified variants in genes such as CD36, MMRN2, EPHB2, PLAUR, and RAPGEF3 as angiogenesis phenotype modifiers (p<0.001). Similar prioritization analysis in the DME cohort identified gene variants such as MEOX2 and PTGER3 as significant phenotype risk modifiers (p<0.001).

Conclusions : Overall, these results suggest that several rare variants may modify the genetic architecture of the phenotypic differentiation of DR into DME or PDR. If validated in future studies, these factors could represent new biomarkers and therapeutic targets for DR.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.