SGLT-2 Inhibitors Top Other Diabetes Medications for Eye Safety

STOCKHOLM — The sodium glucose co-transporter 2 (SGLT2) class of diabetes medications may offer more protection against diabetic retinopathy as compared with other hypoglycemic agents, an analysis of a large commercial database suggested.

SGLT2 inhibitors, such as empagliflozin (Jardiance) and dapagliflozin (Farxiga), reduced the risk of sight-threatening retinopathy by 21-39% versus GLP-1 receptor agonists, DPP-4 inhibitors, and sulfonylureas. The other three drug classes were associated with a similar risk of vision-threatening diabetic retinopathy events.

Of note, the study did not show an increased risk of diabetic neuropathy complications in patients treated with GLP-1 agonists as had been suggested in one of the pivotal trials of semaglutide (Ozempic), said Andrew J. Barkmeier, MD, of the Mayo Clinic in Rochester, Minnesota, at the American Society of Retina Specialists (ASRS) meeting.

“The SGLT2 inhibitor use was associated with a lower-risk of sight-threatening retinopathy compared to other classes of glucose-lowering medications,” said Barkmeier. “GLP-1 receptor agonists did not confer increased risk relative to DPP-4 inhibitors and sulfonylureas. The relative inter-class risks were similar at shorter and longer duration of use.”

Following the talk, ASRS session co-moderator David Boyer, MD, of Retina-Vitreous Associates Medical Group in Los Angeles, asked whether the analysis showed any evidence of ischemic optic neuropathy associated with use of GLP-1 receptor agonists, as recently reported.

“We did look at ischemic neuropathy specifically,” said Barkmeier. “Clearly these are powerful medications that affect broadly our systemic condition, and we are in the process of trying to characterize what are the benefits, what are the potential complications. That’s clearly going to be studied much deeper.”

The study to which Boyer referred involved patients who had recently seen a neuro-ophthalmologist, a limitation of the findings’ generalizability, he added.

In response to another question from Boyer, Barkmeier said investigators did not access data on the patient’s hemoglobin A1c levels, which would have permitted analyses of associations between A1c changes and incidence of diabetic retinopathy complications.

“That’s a critically important question with what we know about worsening of retinopathy with improved diabetes control,” he added.

In 2019 the American Diabetes Association (ADA) recommended SGLT2 inhibitors or GLP-1 receptor agonists for patients with type 2 diabetes and established atherosclerotic cardiovascular disease, independent of glycemic control status, Barkmeier noted. Over the past 5 years, the ADA has expanded the recommendations to include patients with multiple cardiovascular risk factors heart failure, chronic kidney disease, and overweight/obesity. Additionally, use of a GLP-1 receptor agonist is preferred prior to starting insulin.

The increased use of the newer classes of antidiabetic medications has led to questions about potential adverse effects, including diabetic retinopathy complications. A meta-analysis published several years ago showed no increased risk of diabetic retinopathy for SGLT2 inhibitors, GLP-1 agonists, or DPP-4 inhibitors versus placebo.

“All of these trials had relatively short follow-up,” said Barkmeier. “They had narrow systemic inclusion criteria looking at patients at high risk for cardiovascular disease. They had very limited retinopathy assessments, and no direct comparisons between medication classes.”

Preclinical studies showed that SGLT2 inhibitors, GLP-1 agonists, and DPP-4 inhibitors have distinct retinal microvascular, inflammatory, and neuroprotective effects that may be independent of glucose lowering, he continued.

“Add that to what we know about inter-class differences in glucose lowering and pleiotropic effects, and it is plausible that there are meaningful inter-class differences in the risk of diabetic retinopathy,” said Barkmeier.

To look for potential inter-class differences in retinopathy risk, investigators analyzed data from the large and diverse OptumLabs patient population. The query included patients initiating treatment with a GLP-1 receptor agonist, a SGLT2 inhibitor, a DPP-4 inhibitor, or sulfonylurea at least a year after enrollment and no history of diabetic macular edema (DME) or proliferative diabetic retinopathy (PDR). The primary outcome was time to treatment for DME or PDR.

The study population included 371,698 patients: 42,265 who initiated treatment with a GLP-1 receptor agonist, 53,476 with an SGLT2 inhibitor 78,444 with a DPP-4 inhibitor, and 197,513 with a sulfonylurea. Mean follow-up by drug class ranged from 669 to 1,263 days.

Overall, the study population had a low 5-year incidence of treatment for DME or PDR: 0.7% for the SGLT2 inhibitor group, 1.0% for the GLP-1 receptor agonist group, 0.9% for the DPP-4 inhibitor group, and 1.2% for the sulfonylurea group.

SGLT-2 inhibitors were associated with the lowest risk of DME and/or PDR, as compared with each of the other drug classes: 0.73 versus GLP-1 receptor agonists (95% CI 0.55-0.97), 0.79 versus DPP-4 inhibitors (95 CI 0.64-0.97), and 0.61 versus sulfonylureas (95% CI 0.50-0.74). All of the differences achieved statistical significance (P<0.001).

GLP-1 receptor agonists did not differ significantly from DPP-4 inhibitors (HR 1.07, 95% CI 0.85-1.35) or sulfonylureas (HR 0.83, 95% CI 0.67-1.03). Sulfonylureas were associated with a significantly higher risk versus DPP-4 inhibitors (HR 1.29, 95% CI 1.17-1.42).

Please enable JavaScript to view the

comments powered by Disqus.