A multi-institutional study has identified a genetic variation common in people of African ancestry that is associated with an increased risk of complications from diabetes, including diabetic retinopathy. The results from their analysis of human studies and health records data indicated that the diagnosis of diabetes and treatment needed to prevent diabetes complications may be delayed in people who carry genetic variants leading to G6PD deficiency (G6PDdef) because they are associated with reduced levels of HbA1c, a widely used clinical marker of blood glucose levels. The study highlighted one particular variant, or risk allele, designated, rs1050828.
The team suggests that testing for genetic variations that cause G6PD deficiency could lead to improvements in the way clinicians diagnose and treat diabetes, thereby helping to reduce the long-observed disparity in diabetes complications between individuals of European and African ancestries.
“This discovery could lead to changes in the way diabetes is managed for millions of patients in the U.S. and around the world,” said Todd Edwards, PhD, at Vanderbilt University Medical Center (VUMC). “More needs to be done, such as health economics and policy studies, and clinical trials, to establish the best way to use this knowledge to prevent diabetes complications. Now that process can begin.”
Edwards is co-corresponding author of the team’s published paper in Nature Medicine titled, “Adaptive selection at G6PD and disparities in diabetes complications.” In their paper, the team concluded that their findings “demonstrate that individuals with the G6PDdef risk allele receiving standard-of-care treatment have an increased incidence of diabetes complications compared with individuals without the G6PDdef risk allele.”
Edwards and co-corresponding author Ayush Giri, PhD, are both VUMC faculty members affiliated with the VA Tennessee Valley Healthcare System. The study was collaborative with colleagues at the Million Veteran Program (MVP) of the U.S. Department of Veterans Affairs, Emory University School of Medicine, and the Joseph Maxwell Cleland Atlanta VA Healthcare System.
Diabetic retinopathy results from damage to retinal blood vessels and nerves at the back of the eye, which can cause permanent vision loss. The disorder has previously been linked to single nucleotide polymorphisms (SNPs), but these associations have been studied primarily in individuals of European and Asian ancestry. And as the authors pointed out, “At present, only a small number of genetic loci have been associated with diabetic retinopathy, even though twin-based heritability estimates are between 18% and 52% for diabetic retinopathy and proliferative diabetic retinopathy (PDR).”
There are racial disparities in the prevalence of diabetes and its complications, the authors continued, with one recent study estimating that diabetic retinopathy prevalence rates in the United States are 24.4% in non-Hispanic white people and 34.4% in non-Hispanic Black people. To better understand why some people with diabetes develop retinopathy but others don’t, the researchers conducted a combined-ancestry genome-wide association study (GWAS) of more than 197,000 individuals with diabetes, including more than 68,000 who also had diabetic retinopathy.
The researchers utilized electronic health records (EHR) and genomic data from the MVP, the UK Biobank, VUMC’s BioVU biorepository, the Mass General Brigham Biobank in Boston, and summary statistics from a 2019 study. The resulting study was the largest ancestry-stratified, SNP-based estimation of the heritability of diabetic retinopathy conducted to date, and included an unprecedented number of individuals of non-Hispanic African (NH-AFR) ancestry—more than 46,000. “With an EHR-based algorithm to identify diabetic retinopathy cases and controls with diabetes, we conducted a combined-ancestry GWAS of diabetic retinopathy, totaling 68,169 cases and 129,188 controls,” the authors explained. “Individuals of NH-AFR ancestry are overrepresented in the MVP compared with the US population, allowing us to investigate the genetic architecture of diabetic retinopathy in individuals of NH-AFR ancestry on an unprecedented scale.”
Glucose-6-phosphate dehydrogenase deficiency, which is only common in African and some Asian populations, may have evolved as a protection against severe malaria, the team commented. G6PDdef is associated with a shorter red blood cell lifespan, which lowers HbA1c levels but not blood glucose levels. However, this “mismatch” can mask the true extent of hyperglycemia: in individuals who carry a G6PDdef mutation, HbA1c levels systematically underestimate blood glucose levels. As the authors pointed out, “Diabetes complications occur at higher rates in individuals of African ancestry.”
The team’s newly reported study detected nine previously unreported chromosome loci that were associated with diabetic retinopathy, including an evolutionarily adaptive genetic variant, rs1050828 that potentially may explain some of the racial differences in diabetes complications, they suggested. “We detected nine previously unreported loci associated with diabetic retinopathy, including an evolutionarily adaptive genetic variant with potential impacts on racial disparities in diabetes complications in individuals of NH-AFR ancestry,” they wrote.
As part of their analyses, the team investigated the effect of G6PDdef on the risk of diabetes complications among participants in the NIH-sponsored Action to Control Cardiovascular Risk in Diabetes (ACCORD) clinical trial, which assessed the impact of tight diabetes control on cardiovascular events in more than 10,000 adults with type 2 diabetes.
Their analyses found that individuals of non-Hispanic African ancestry with G6PDdef had a significantly higher likelihood of two diabetes complications—diabetic retinopathy and diabetic nephropathy—compared to individuals without the variant, despite receiving standard-of-care treatment to lower HbA1c levels. “A composite outcome of diabetic retinopathy or diabetic nephropathy as well as two other diabetes complications were significantly associated with the G6PDdef risk allele in the ACCORD clinical trial data, regardless of medication regimen,” they further reported.
Outlining the results of their analysis of MVP data, the team also noted, “For men, we estimated a 2.39% (total effect, P < 2.0 x 10-16) increase in diabetic retinopathy risk due to the G6PDdef risk allele.” And while most MVP participants are men, they identified similar results in women.
Based on the prevalence of this genetic variant, the researchers estimated that more than 250,000 men and 500,000 women of non-Hispanic African ancestry in the United States who have diabetes may have some level of G6PD deficiency. “With an allele frequency between 8% and 12%, approximately 1.6 to 2.4 million men and 3.1 to 4.5 million women of NH-AFR ancestry carry this allele,” the team wrote. “Considering the prevalence of diabetes in NH-AFR ancestry individuals is as much as 20% in the U.S., we estimate that there are approximately 270,000 men and 510,000 women of NH-AFR ancestry with diabetes and some level of G6PDdef derived from the rs1050828 genotype.”
Their figures, they suggested, are broadly consistent with a previous study, which estimated that diabetes might be diagnosed late, or remain undiagnosed, in as many as 650,000 people of non-Hispanic African ancestry in the United States due to the rs1050828 G6PDdef variant. “With excess risk of diabetic retinopathy for the G6PDdef risk allele of 118 per 1,000 individuals of NH-AFR ancestry with diabetes, we estimate that nearly 12% of all diabetic retinopathy cases in individuals of NH-AFR ancestry in the U.S. could be attributed to the G6PDdef risk allele.”
HbA1c works well as a marker for hyperglycemia in most people, but not as well in individuals with genetic disorders that affect enzyme function such as G6PD deficiency, noted Giri, who is assistant professor of obstetrics and gynecology at VUMC. “Were HbA1c not used widely for screening and management of diabetes, we likely would not observe such a finding,” he said
“While this discovery may impact how millions of individuals manage their diabetes, it also highlights the importance of including diverse populations in biomedical research,” said the paper’s first author, Joseph Breeyear, PhD, a postdoctoral research fellow at the National Institute of Environmental Health Sciences of the National Institutes of Health (NIH). “By including underrepresented individuals, we can identify genetic variations that affect health outcomes.”
The researchers concluded, “With comprehensive screening … and subsequent standard-of-care treatment, possibly aimed at glucose rather than HbA1c targets, nearly 12% of diabetic retinopathy cases and 9% of diabetic neuropathy cases in individuals of non-Hispanic African ancestry could be avoided in the U.S. alone.” They estimate that screening as few as 229 individuals of HN-AFR ancestry would prevent one case of diabetic retinopathy. “Alternatively, if only NH-AFR people who had been diagnosed with diabetes were screened, with the optimistic assumption that the effects of the G6PDdef risk allele were completely mitigated by adjusted treatment after diagnosis, screening as few as 44 people could prevent one case of diabetic retinopathy.”