Understanding the Genetics Behind Susceptibility to PDR

Proliferative diabetic retinopathy (PDR) is the most severe form of diabetic eye disease and a leading cause of preventable blindness among working‑age adults worldwide. Although tight glycemic control and management of systemic risk factors remain the cornerstone of prevention, a striking variability in disease progression exists among individuals with similar metabolic profiles. This heterogeneity has long pointed to an underlying genetic component. Over the past decade, genome‑wide association studies (GWAS) and candidate gene analyses have begun to unravel the hereditary factors that predispose certain patients to PDR. Understanding these genetic drivers not only refines risk stratification but also opens the door to mechanism‑based therapies that could halt or reverse retinal neovascularization.

In this article, we explore the current state of knowledge regarding the genetics of PDR susceptibility. We examine key susceptibility genes, their biological roles in angiogenesis and inflammation, the emerging utility of polygenic risk scores, and how pharmacogenomics may personalize anti‑VEGF treatment. We also address the challenges and ethical considerations that accompany the translation of genetic findings into routine clinical care.

The Role of Genetics in PDR

PDR occurs when retinal ischemia triggered by diabetic microvascular damage stimulates pathological growth of fragile new blood vessels. While hyperglycemia and hypertension are established environmental triggers, twin studies have estimated that heritability accounts for roughly 25–50% of the risk for diabetic retinopathy overall, with even higher heritability for the proliferative stage. Family aggregation studies report that siblings of patients with PDR have a two‑ to threefold increased risk compared with the general diabetic population, independent of glycemic control. These observations provide strong rationale for identifying the specific genetic variants that confer susceptibility.

Genetic predisposition can influence every stage of the disease process: from the rate of capillary closure and hypoxia to the magnitude of the angiogenic response and the degree of inflammatory infiltration. Importantly, many of the genes implicated in PDR are also involved in other microvascular complications, such as diabetic nephropathy and neuropathy, suggesting shared molecular pathways.

Genetic Polymorphisms and PDR Risk

Most genetic risk variants for PDR are single‑nucleotide polymorphisms (SNPs) that occur in non‑coding or regulatory regions of the genome. These SNPs can alter transcription factor binding sites, affect mRNA splicing, or modify enhancer activity, thereby changing the expression level of nearby genes. Several large GWAS have reported significant associations in populations of European, Asian, and African descent, although many findings have not been consistently replicated. Meta‑analyses have been instrumental in identifying robust susceptibility loci.

A landmark 2020 GWAS meta‑analysis from the International Genetics of Diabetic Retinopathy Consortium identified four genome‑wide significant loci for severe diabetic retinopathy, including PDR: one near VEGFA, one upstream of TSHZ2, one in COL1A1, and another in an intergenic region on chromosome 8. These results underscore the polygenic nature of PDR risk, where many small‑effect variants cumulatively modulate susceptibility.

Key Genes Associated with PDR

  • VEGFA (Vascular Endothelial Growth Factor A): The most extensively studied gene in PDR. Common promoter SNPs (e.g., rs2010963, rs3025039) are associated with altered VEGF expression. High VEGF levels drive neovascularization; carriers of high‑expression haplotypes exhibit an earlier onset and more aggressive form of PDR.
  • EPAS1 (Endothelial PAS Domain Protein 1): Encodes HIF‑2α, a master regulator of the hypoxic response. Variants in EPAS1 have been linked to increased VEGF transcription under low oxygen conditions, amplifying angiogenic signaling in the ischemic retina.
  • ACE (Angiotensin‑Converting Enzyme): An insertion/deletion (I/D) polymorphism in the ACE gene influences serum enzyme activity. The D‑allele is associated with higher ACE levels, increased angiotensin II, and enhanced vasoconstriction, promoting retinal microvascular dysfunction and progression to PDR.
  • TNF‑α (Tumor Necrosis Factor‑alpha): Pro‑inflammatory cytokine. Certain promoter alleles (e.g., ‑308 G>A) lead to higher TNF‑α production, perpetuating chronic inflammation that compromises the blood‑retinal barrier and facilitates neovascularization.
  • AKR1B1 (Aldo‑Keto Reductase Family 1 Member B1): Encodes aldose reductase, the first enzyme in the polyol pathway. Genetic variants that increase enzyme activity enhance sorbitol accumulation, oxidative stress, and microvascular damage in diabetic retinas.
  • COL1A1 (Collagen Type I Alpha 1): Polymorphisms affecting collagen synthesis have been reported in GWAS for severe retinopathy. Altered extracellular matrix composition may influence basement membrane thickening and tractional retinal detachment in PDR.

How Genetic Variations Influence Disease Mechanisms

The genes implicated in PDR susceptibility map to three interrelated biological themes: angiogenesis and hypoxia sensing, chronic inflammation, and metabolic stress. Understanding how variants perturb these pathways provides insight into why some patients develop neovascularization despite good glycemic control, while others with poor control are spared.

Angiogenesis and the VEGF Pathway

VEGF is the primary driver of pathological neovascularization in PDR. Genetic variation in the VEGFA gene itself, as well as in its regulatory partners, dictates the magnitude of VEGF secretion in response to hypoxia. For instance, the promoter SNP rs833061 has been linked to altered HIF‑1α binding, leading to differential VEGF induction. Similarly, variants in KDR (encoding VEGFR‑2) affect receptor density on endothelial cells, modulating the sensitivity of the retina to VEGF. Patients carrying high‑response alleles may require higher cumulative doses of anti‑VEGF therapy and may experience faster recurrence of macular edema.

Beyond VEGF, the hypoxia‑inducible factor (HIF) pathway is central. Variants in EPAS1 (HIF‑2α) and VHL (von Hippel‑Lindau protein, which degrades HIF) can shift the threshold for HIF stabilization, leading to prolonged expression of Angpt‑2 and PDGF‑B, growth factors that contribute to pericyte dropout and vascular leakage. Recent studies also highlight polymorphisms in PEDF (pigment epithelium‑derived factor), an endogenous anti‑angiogenic factor; low‑expression variants may tip the balance toward pro‑angiogenic signaling.

Inflammatory Response Genes

Chronic low‑grade inflammation is a hallmark of diabetic retinopathy. Genetic variants that amplify cytokine release can accelerate disruption of the blood‑retinal barrier. Beyond TNF‑α, polymorphisms in IL‑1β (rs1143627) and IL‑6 (rs1800795) have been associated with higher serum cytokine levels and increased risk of PDR. The inflammasome component NLRP3 is also implicated; gain‑of‑function variants promote IL‑1β and IL‑18 maturation, fuelling sterile inflammation. Variants in ICAM1 (intercellular adhesion molecule 1) affect leukocyte adhesion to retinal endothelium, a key early step in vascular damage.

A growing literature suggests that genetic regulation of complement factors (e.g., CFH) may modify PDR risk, mirroring findings in age‑related macular degeneration. The interplay between complement activation and VEGF signaling is complex but represents a promising intersection for future therapeutics.

Oxidative Stress and Metabolic Pathways

Hyperglycemia‑induced oxidative stress is magnified by genetic variation in antioxidant defense enzymes. Polymorphisms in SOD2 (manganese superoxide dismutase), CAT (catalase), and GPX1 (glutathione peroxidase 1) have been linked to reduced free radical scavenging capacity, leading to retinal cell apoptosis and endothelial injury. The polyol pathway gene AKR1B1 is one of the most consistently replicated susceptibility genes. The intronic variant rs759853 is associated with increased aldose reductase activity and higher sorbitol accumulation, which in turn depletes NADPH and generates reactive oxygen species. This mechanism explains why some diabetic patients develop more rapid neuroretinal degeneration and capillary loss.

Additionally, variation in genes involved in advanced glycation end‑product (AGE) metabolism, such as RAGE (receptor for AGEs), influences the accumulation of vascular‑damaging cross‑links. The 82G/S polymorphism in RAGE increases ligand binding and promotes inflammatory signaling, further accelerating microvascular pathology.

Genetic Testing and Risk Prediction

As the inventory of confirmed PDR‑associated variants grows, the possibility of integrating genetic testing into routine diabetes care becomes more tangible. However, the predictive power of any single variant is too low to guide clinical decisions. Instead, researchers are increasingly focusing on polygenic risk scores (PRS) that aggregate the effects of hundreds to thousands of SNPs across the genome.

Current State of Genetic Screening

Commercially available panels for diabetic retinopathy are limited, and most are used in research settings. A few direct‑to‑consumer platforms include PDR‑related SNPs, but without clinical validation or established risk thresholds. The American Diabetes Association currently does not recommend routine genetic testing for retinopathy risk. Challenges include the lack of high‑quality prospective studies, ethnic variation in effect sizes (most GWAS have been performed in East Asian or European populations), and the need for scalable, cost‑effective genotyping platforms in low‑resource settings where diabetes prevalence is highest.

Nevertheless, targeted testing for high‑penetrance variants, such as those in VEGFA or AKR1B1, could be used to stratify patients after they have already developed moderate non‑proliferative retinopathy, helping identify those most likely to progress to PDR within a given time window. Early data suggest that combining a PRS with clinical variables (HbA1c, duration of diabetes, blood pressure) improves risk discrimination compared with clinical factors alone.

Polygenic Risk Scores in Action

Several recent studies have developed and internally validated PRS for PDR. For example, a 2023 analysis from the UK Biobank and a South Asian cohort derived a PRS based on 150 validated variants that yielded an area under the curve (AUC) of approximately 0.75 for PDR, compared with 0.60 for clinical variables alone. A similar PRS in a multi‑ethnic cohort—including Black, Hispanic, and Asian participants—showed consistent predictive ability across groups, though with attenuated effect sizes in African‑ancestry individuals, highlighting the need for ancestral‑diverse GWAS.

Integration of PRS into electronic health records could enable automated risk alerts, prompting earlier referral to retinal screening or more frequent follow‑up. However, widespread adoption requires rigorous clinical utility studies showing that PRS‑guided screening reduces the incidence of advanced PDR or blindness.

Implications for Personalized Treatment

The ultimate promise of PDR genetics is not simply risk prediction but the tailoring of prevention and treatment to each patient’s molecular profile. Personalized medicine in ophthalmology is in its infancy, but pharmacogenomic insights are beginning to reshape anti‑VEGF therapy.

Pharmacogenomics of Anti‑VEGF Therapy

Response to intraocular anti‑VEGF agents (e.g., ranibizumab, aflibercept, bevacizumab) varies substantially among PDR patients. Genetic differences in the VEGF‑VEGFR axis help explain this variability. Several studies have reported that carriers of the VEGFA rs3025039 variant (a 3′‑UTR SNP affecting mRNA stability) require more injections over a two‑year period and have a higher likelihood of persistent neovascularization. Likewise, polymorphisms in VEGFR‑2 (KDR) and LPA (lipoprotein(a), which binds to VEGFR‑2) have been associated with reduced drug efficacy.

Emerging data suggest that a genotype‑guided treatment algorithm could optimize dosing schedules. For instance, patients with high‑expression VEGFA haplotypes might benefit from higher loading doses or shorter injection intervals, while those with low‑expression variants might be managed with a less intensive regimen. Additionally, genetic testing could identify patients who are less likely to respond to VEGF inhibition alone and may require combination therapy with a corticosteroid or an anti‑inflammatory agent. These strategies are being tested in small observational trials; larger randomized studies are needed to confirm cost‑effectiveness and outcomes.

Future Targeted Therapies

Beyond anti‑VEGF, the identified genetic pathways offer novel druggable targets. Aldose reductase inhibitors (e.g., epalrestat), which have been used for diabetic neuropathy, are being re‑evaluated for retinopathy in patients carrying high‑risk AKR1B1 variants. HIF‑2α antagonists (e.g., belzutifan, approved for VHL‑associated cancers) are being explored in preclinical models of retinal neovascularization. Anti‑inflammatory agents that block IL‑1β (e.g., canakinumab) or TNF‑α (e.g., adalimumab) may find a niche in PDR patients with high inflammatory genotypes. Finally, gene‑editing approaches, such as CRISPR‑based disruption of VEGFA in retinal cells, are being researched but remain years away from clinical translation.

The development of these targeted therapies will be greatly accelerated by the inclusion of genetic stratification in clinical trial design. Enriching trial populations for patients who harbor the relevant risk variants can increase effect sizes, reduce sample size requirements, and bring effective drugs to market faster.

Research Challenges and Ethical Considerations

Despite significant progress, several hurdles remain before PDR genetics can be routinely applied. First, reproducibility of genetic associations across diverse populations is a critical issue. Most published variants were discovered in cohorts of European or East Asian ancestry; their transferability to individuals of African, South Asian, or Native American descent is uncertain. Large‑scale, multi‑ethnic collaborative consortia are essential to develop robust PRS that work equitably.

Second, the interplay between genetics and environmental factors—such as diet, exercise, and smoking—is poorly quantified. Epigenetic modifications (e.g., DNA methylation at hypoxia‑responsive genes) may mediate some of these interactions, but the field lacks comprehensive studies that integrate multi‑omics data (genomics, epigenomics, metabolomics) with longitudinal clinical outcomes.

Third, the clinical utility of genetic testing must be demonstrated through prospective randomized trials. Informing a patient that they carry a high‑risk genotype could cause anxiety or lead to fatalism, while false reassurance could delay essential screening. Clear genetic counseling protocols and decision‑support tools are needed.

Finally, ethical considerations regarding privacy, insurance discrimination, and return of results in asymptomatic populations must be addressed. GINA (Genetic Information Nondiscrimination Act) provides protections in the United States against employer and health insurer discrimination, but gaps remain for life insurance and disability coverage. Any implementation framework should include safeguards and informed consent processes that respect patient autonomy.

Conclusion

The genetics of susceptibility to proliferative diabetic retinopathy has advanced from candidate gene studies to well‑powered GWAS and polygenic risk scores. Key genes in the VEGF, HIF, inflammatory, and polyol pathways have been robustly linked to disease risk and progression. These discoveries are beginning to enable more precise risk stratification and to guide personalized anti‑VEGF therapy. In the coming decade, continued efforts to diversify genetic studies, integrate multi‑omics data, and conduct pragmatic clinical trials will determine whether genetic insights translate into measurable reductions in vision loss from PDR.

For clinicians, staying informed about these developments is essential, as the first clinical‑grade genetic tests for diabetic retinopathy are likely to emerge within the next few years. While genetics will never replace the need for tight glucose and blood pressure control, it can identify those who need intensified surveillance and carve out new avenues for intervention. The ultimate goal—a future in which fewer patients progress to the proliferative stage of retinopathy and those who do receive precisely targeted therapies—is now within sight.

For further reading: Genome‑wide association study of proliferative diabetic retinopathy in an East Asian population (PMID: 32805155); Polygenic risk scores in diabetic retinopathy (Diabetes Care, 2021); Pharmacogenetics of anti‑VEGF therapy in diabetic macular edema (PMCID: PMC8145638).