乌鸦传媒

SEATTLE -- Diabetic patients treated with GLP-1 agonists had significantly more progression of existing diabetic eye disease as compared with those treated with SGLT2 inhibitors, a matched cohort study showed.

Treatment with GLP-1 agonists almost doubled the likelihood of progression from nonproliferative diabetic retinopathy (NPDR) to proliferative diabetic retinopathy (PDR) after 3 years. A statistically significant difference emerged after just 3 months of follow-up. GLP-1 agonists also were associated with significantly higher rates of progression to diabetic macular edema (DME) beginning at 6 months and continuing through 3 years.

In addition, GLP-1 agonists had a significant association with the need for more interventions for diabetic eye disease, reported Ehsan Rahimi, MD, of Stanford University in California, at the American Society of Retina Specialists (ASRS) annual meeting.

"Future directions are to better describe the phenotypic profile of patients who may be at risk of progression and to assess the potentially synergistic impact for patients that we all see in our clinics who are already receiving intravitreal therapy when they start these meds," said Rahimi.

"Finally, we have a real-world descriptive analysis underway of a separate medication that's related in this class, tirzepatide (Mounjaro), which is a dual-mechanism drug. This drug has been kind of all the rage, and you've probably seen it in the media," he added.

Background and Key Findings

Use of GLP-1 agonists and SGLT2 inhibitors has increased dramatically, with limited evaluation of the drugs' potential effects on the eyes. Rahimi and colleagues performed a retrospective cohort study based on data from the health research network.

Using two propensity score-matched cohorts, the researchers evaluated the frequency of progression to vision-threatening complications according to type of diabetes therapy. Secondarily, they evaluated the need for interventions for diabetic eye disease, including intravitreal injections, photocoagulation, and surgical vitrectomy.

The analysis comprised 7,889 patients treated with GLP-1 agonists and 5,663 patients treated with SGLT2 inhibitors. All patients had existing NPDR. After propensity score matching, the GLP-1 cohort had slightly higher HbA1c (8.52% vs 8.37%, P<0.01), and a higher proportion on insulin (61.1% vs 55.1%, P<0.01).

The GLP-1 group was almost twice as likely to progress to PDR after 3 years (RR 1.585, 95% CI 1.337-1.881, P<0.0001). Rates of progression separated after just 3 months of follow-up (P<0.01). Beginning at 6 months and continuing through 3 years, patients on GLP-1 agonists were significantly more likely to progress to DME (RR 1.283, 95% CI 1.194-1.379, P<0.0001).

From 12 to 36 months of follow-up, the GLP-1 cohort required more intravitreal injections (RR 1.307, 95% CI 1.142-1.496, P<0.0001). Rates of panretinal photocoagulation (RR 1.409, 95% CI 1.03-1.927, P<0.01) and pars plana vitrectomy (RR 1.513, 95% CI 1.011-2.263, P<0.01) also were higher in the GLP-1 cohort at 36 months.

During a discussion that followed the presentation, Rahimi was asked to speculate about a potential mechanistic explanation for the findings.

"We see these patients in our clinics all the time," he said. "They go on these medicines, and their HbA1c crashes, goes down very quickly. That rapid reduction is thought to play some role. But if you look at the basic science literature, it's suggested that there are direct effects of these medications on the retina. That being said, it's also been suggested that there may be a protective effect on the retina. We're getting a lot of mixed signals."

VEGF Inhibitors and Systemic Adverse Events

A second presentation during the same ASRS session confirmed a recent report linking anti-vascular endothelial growth factor (VEGF) drugs to an increased risk of systemic adverse events (AEs) in diabetic patients with eye disease. Treatment with VEGF inhibitors almost doubled the likelihood of any prespecified systemic event and significantly increased the risk of individual types of events: myocardial infarction, cerebrovascular disease, and kidney disease, reported Roomasa Channa, MD, of the University of Wisconsin in Madison.

A multivariable analysis identified four factors that independently increased the risk of systemic events in patients receiving intravitreal anti-VEGF therapy: tobacco use, greater comorbidity burden, existing PDR, and severe NPDR.

As with the drug treatment reviewed by Rahimi, use of anti-VEGF therapy to treat diabetic eye disease has increased dramatically, without accumulation of data regarding the drugs' potential impact on the risk of systemic AEs, already increased by virtue of diabetes, said Channa. One previous study showed among patients with DME treated with anti-VEGF therapy versus laser therapy during 2 years of follow-up. A second study showed among patients with DME treated with anti-VEGF therapy versus laser therapy or steroids, albeit with only 6 months of follow-up.

"As I think about the results from this study, I think about taking a moment to consider our patients' cumulative exposure to anti-VEGF injections, as the indications for these injections expand to patients with diabetes, thinking about how many injections they're getting and for how long and identifying those at high risk," said Channa.

"I'm thinking about how we can decrease the anti-VEGF treatment burden in these high-risk patients, perhaps by switching them to steroids or laser or even considering defer-and-extend in the case of stability," she added.

During the discussion that followed her presentation, Channa said the researchers had not looked at the relationship between the number of intravitreal injections and frequency of AEs, but they did find an association between duration of treatment and AEs.

Marc Peden, MD, of Retina Associates of Florida in Tampa, urged caution when interpreting the findings.

"We should be very cautious about causation versus correlation," he said. "We're big advocates against undertreatment and we have been speaking up against that for a long time. These are sicker patients, that's why we are treating them with anti-VEGF. We have to be very cautious when we make claims about this. It's a great study, and I think it's very important to pursue and look more into [this]."

"But I would take this data with a grain of salt and I wouldn't necessarily let it discourage you from treating patients with anti-VEGF, because that sometimes becomes more of a take-home message," he continued. "We just want to make sure that we're treating our patients appropriately, but to your point, we're going to make sure that we're very safe."