American Society of Retina Specialists 42nd Annual Scientific Meeting

The Syfovre balancing act: The retinal vasculitis risk is real, but research also points to benefits

Last year’s annual meeting of the ASRS in Seattle, Washington, was rocked by news that a small but notable number of patients treated with Syfovre (pegcetacoplan), the first-ever FDA-approved treatment for geographic atrophy, developed retinal vasculitis, a serious complication that can lead to severe, permanent vision loss. The ASRS Research and Safety in Therapeutics Committee provided an almost day-by-day timeline and details on what was then a handful of cases.

At this year’s meeting, Syfovre’s risks and benefits were still a featured topic, but the sense of urgency mellowed into the calmer territory more typical of medical conferences: discrepant findings and differing emphasis on risk and benefits but also points of agreement.

The combined results of two real-world evidence studies presented at the meeting put the risk of developing retinal vasculitis at about 1 in every 1,330 patients. David Eichenbaum, M.D., a retinal specialist at the Retina Vitreous Associates of Florida in southwest Florida and a consultant and speaker for Apellis Pharmaceuticals, the maker of Syfovre, put the risk at 1 in every 4,000 first injections.

“I think it’s still a great interest, and so that’s why we’ll continue to do our real-world study to look at this,” said Samuel Minaker, M.D., M.Sc., an assistant professor at RushRush University Medical Center in Chicago and a physician Illinois Retina Associates, who presented findings about the retinal vasculitis from one of the real-world studies. But Minaker said he uses both Syfovre and its rival, Izervay (avacincaptad pegol), to treat geographic atrophy, noting that there might be a trade-off with Syfovre being more effective but having more documented risks.

Practice patterns vary, he said, with some retinal specialists treating patients with Syfovre and others electing not to and possibly referring patients to specialists who do. “Patients just need to be aware that something is available, but it requires some more nuanced discussion because of these additional risks,” he said in an interview after his presentation.

The organizers of the expert panel session on Syfovre — the meeting schedule used the generic name, pegcetacoplan, not the brand name, in the title — seemed conscious of striking a balance. Eichenbaum and Sunir J. Garg, M.D., a retina specialist at the Wills Eye Hospital in Philadelphia, presented data supporting Syfovre’s efficacy. Minaker and Janani Singaravelu, M.D., M.S., a vitreoretinal surgery fellow at the Retina Group of Washington, in Washington, D.C., presented real-world study data that focused on the complications, particularly retinal vasculitis.

ASRS annual meeting attendees during a break between sessions. The meeting was held in Stockholm, Sweden, at Stockholmsmässan, the city's main convention center.

Eichenbaum presented data from an open-label extension study of Syfovre showing the efficacy of monthly and every-other-month injections of Syfovre relative to sham treatment. There has been some debate about the strength of the evidence for Syfovre having a positive effect on vision. Eichenbaum shared results that showed monthly injections of Syfovre reduced the risk of severe visual impairment by
38% relative to the sham treatment, although there was only a 12% difference for every-other-month injections. Severe visual impairment was defined as sustained reduction in sight, as measured by two standard eye chart exams (the ETDRS and Snellen charts) for four months or more without recovery.

Eichenbaum did not present retinal vasculitis data in detail but said after a total of 250,000 injections, it appears that retinal vasculitis occurs after 1 in every 4,000 first injections. One of the points of agreement about Syfovre and retinal vasculitis is that the risk seems limited to the first injections of the drug. “Once patients make it past that first injection, it is not something that they need to worry about,” said Minaker.

“As we see the pegcetacoplan risk profile stay stable, we are seeing the clinical utility of pegcetacoplan come into clearer focus,” Eichenbaum summed up for the ASRS audience.

Microperimetry is a test that measures people’s field of vision by asking them whether they can see spots of light. Garg presented data showing that patients who had been treated for 36 months with Syfovre scored better on prespecified microperimetry end points than those who had been treated for only 12 months. In the written summary of his presentation, he said this was the first time a treatment for geographic atrophy had shown a positive visual function benefit according to a prespecified end point.

Minaker’s presentation showed data pulled from the electronic medical records from Retinal Consultants of America, a nationwide group that includes 265 retina specialists, and Mid Atlantic Retina, the retinal specialist group based at Wills to which Garg belongs. The review covered the period from Feb. 1, 2023, through March 30, 2024 (the FDA approved Syfovre in February 2023). During that period, 6,525 patients were treated with Syfovre and 32,080 injections were administered. Minaker said it was the largest data set on the experience with Syfovre so far.

Minaker’s results showed that treatment of 853 of the 8,987 (9.49%) eyes treated with Syfovre was halted during the period of his study. His tally of side effects included 296 new cases of choroidal neovascularization; 29 cases of mild intraocular inflammation, which he said worked out to 0.4% of all treated patients; and four cases of retinal vasculitis, which Minaker calculated to be a risk of 0.06% per patient.

Singaravelu’s study included 1,091 patients treated with Syfovre at the three retina specialist groups that make up the PRISM Vision Group between February 2023 and October 2023. Ocular hypertension (defined as increase in intraocular pressure greater than 10 mm Hg from baseline) occurred in 76 patients (5.25%) and was the most common ocular side effect seen, Singaravelu reported.

Five cases (0.34% of the 1,451 treated eyes) of intraocular inflammation occurred, three of which were anterior uveitis and one a mild nonocclusive case of retinal vasculitis that was treated with topical steroids. The fifth case developed into hemorrhagic occlusive retinal vasculitis nine days after the first injection of Syfovre. Eventually the affected eye was removed (enucleation).

Coverage for remote diabetic retinopathy screening is declining, creating disparities

Although diabetic retinopathy is a leading cause of blindness in adults, fewer than half of Americans with diabetes have annual eye exams, Glenn C. Yiu, M.D., Ph.D., director of teleophthalmology and professor in the Department of Ophthalmology at University of California Davis Health, said during a presentation at the ASRS meeting.

Teleophthalmology can fill in the gaps in screening, but those who would benefit most often have limited insurance coverage for telehealth.

Glenn C. Yiu, M.D., Ph.D.

“Teleophthalmology increased but payments declined, and lower reimbursements disproportionately impacted vulnerable populations, including [elderly patients], women, Black [patients] and lower-income patients,” Yiu said.

UC Davis’s teleophthalmology program was launched in 2018 at one institution and was able to increase screening by 15%, Yiu said. Image quality was good, and the center was able to identify not only possible retinopathy but also glaucoma, macular degeneration and hypertensive retinopathy. Ultimately, he said, the program referred about one-third of those screened for follow-up.

But only half of the screenings were paid for by insurance, with most denials coming from Medicare, he said. Yiu and his colleagues analyzed use of teleophthalmology screenings and payment for patients newly diagnosed with type 2 diabetes. They used the OptumLabs Data Warehouse, which is a database with about a 14% market share.

They reviewed claims data from January 2011 to December 2020 for three teleophthalmology codes. Researchers then assessed trends in claims paid, factors associated with insurance payments and time to first eye exam.

What they found was that over that decade, remote screening increased but reimbursements declined. “We noticed that the insurance coverage for teleophthalmology codes [has] been drifting down over the last decade, starting from 90% in 2011, when the codes were developed, to less than 50% by 2020,” Yiu said.

Additionally, the amount paid per claim has been stagnant over the last decade, he said.

Insurance coverage was also lower for those who were older, who were women, who were Black, who had lower income and who had less education. “This is especially true for those [with] Medicare Advantage, which correlates with this drop in insurance claim coverage for teleophthalmology,” he said.

Teleophthalmology, Yiu’s review found, allows for screening about a month and half earlier for those newly diagnosed with diabetes. “We found that about one-third of these were the same day as the diabetes diagnosis, which suggests that the advantage of teleophthalmology screening is when they are in the primary care office,” he said.

Researchers also assessed time to in-person follow-up after a diabetic retinopathy screening. Fewer than half saw an ophthalmologist, with non-English-speaking patients and those with comorbidities and diabetes complications being the least likely to follow up.

Last year, UC Davis received a $2 million grant for an interdisciplinary program, the Collaborative UC Teleophthalmology Initiative, or CUTI. The project is in partnership with medical centers at UC San Diego, UC San Francisco and UC Los Angeles to expand eye care access for patients with diabetes, especially underserved populations.

As part of this program, Yiu will conduct a four-year study, collecting retinal images in a centralized database for research using artificial intelligence, which may be able to identify eye disease even earlier.

Cialis for AMD: The possible good, the possible bad, and it may not make a difference

There’s reason to believe that phosphodiesterase 5 (PDE5) inhibitors, such as Cialis (tadalafil) and Viagra (sildenafil), might have an unintended effect on the progression of age-related macular degeneration (AMD).

The PDE5 inhibitors increase blood flow in the choroid, a middle layer of eye tissue sandwiched between the sclera and the retina that is filled with blood vessels. Increased choroidal blood flow could, in theory, be beneficial for people with dry, nonexudative AMD, which is characterized by a breakdown of the cells of the retina and thinning out of the macula, the center of the retina. The increased perfusion might provide the retina with additional oxygen and nutrients.But the extra perfusion might worsen wet, exudative AMD, causing the leaky blood vessels to leak even more.

Tracy Z. Lang, a medical student at the University of Southern California Keck School of Medicine in Los Angeles, and colleagues presented findings of a study looking at the double-edged sword of PDE5 inhibitors at the ASRS meeting. More specifically, they looked for evidence of whether Cialis was associated with a worsening of early or intermediate AMD to advanced AMD, both the dry and wet types of the disease.

They conducted their study using Optum’s Clinformatics Data Mart database of approximately 32 million patients. They sifted through the records to identify patients with a diagnosis code indicating a diagnosis of early or intermediate AMD between January 2015 and December 2020. After a couple of other siftings, they homed in on 1,151 patients with an insurance claim record indicative of a prescription for tadalafil, which is marketed under the brand name Cialis as a treatment for benign prostatic hyperplasia and erectile dysfunction and as Adcirca as a treatment for pulmonary arterial hypertension. They matched those patients with those who didn’t have a tadalafil prescription. Lang and her colleagues also grouped the tadalafil patients into low and high cumulative dosages groups.

The result? After two years, they didn’t see any statistically discernible difference between those with a tadalafil prescription claim and those without one as far as progression to advanced AMD. They also didn’t see any difference when comparing patients with high cumulative dosage with those with a relative one.

The good news/bad news implication is that while the PDE5 inhibitor wasn’t associated with a worsening of wet AMD, it also wasn’t associated with any protection against the progression of dry AMD.

In an interview at the ASRS meeting, Lang said the result may show that while tadalafil does, in fact, affect choroidal perfusion, the difference in blood flow may not be clinically significantly. Another possibility, Lang said, is that the two-year period of the study wasn’t long enough to detect tadalafil’s effect on dry and wet AMD and worsening to the advanced forms of the disease.

High-dose Eylea not associated with elevated IOP risk

At 70 microliters, the volume of the intravitreal injection of the high-dose, 8-milligram (mg) version of Eylea (aflibercept) is larger than the 50-microliter volume of the original 2-mg version of the anti-vascular endothelial growth factor (VEGF) drug used to treat several common retinal diseases, including wet age-related macular degeneration, diabetic macular edema and diabetic retinopathy. That higher volume has caused some concern that the high-dose version of the injections, although given less frequently, might cause harmful increases in pressure within the eye, known as intraocular pressure (IOP), and possibly glaucoma-related treatment.

Results of a post hoc analysis of the PHOTON trial presented at the ASRS may allay those worries. Positive results from the PHOTON study led to FDA approval of high-dose Eylea for diabetic macular edema and diabetic retinopathy.

Jordana Fein, M.D., M.S., a retinal specialist at the Retina Group of Washington in Washington, D.C., and the nearby areas of Maryland and Virginia, shared data showing that in patients with diabetic macular edema there was little difference in IOP in all the treatment groups from the start of the 48-week PHOTON trial to its end. The PHOTON trial had three treatment groups: patients received either the standard 2-mg version; the high-dose, 8-mg version every 12 weeks; or that higher dose every 16 weeks. Fein noted patients could receive the 2-mg version in the fellow eye, according to the discretion of investigators. At the start of the study, the mean IOP in the 2-mg, 8-mg every 12 weeks, and 8-mg every 16 weeks groups were 15.9 mm Hg, 15.3 mm Hg and 14.9 mm Hg, respectively. At week 48, the IOP levels for those three groups were
15.7 mm Hg, 15.1 mm Hg and 14.8 mm Hg, respectively. The abstract of Fein’s presentation noted that across all three groups, the mean changes from baseline to the end of the trial, a span of 48 weeks, did not exceed plus or minus 1 mm Hg. She noted that changes in IOP were similar in the treated and the untreated fellow eyes, “suggesting that there was no drift in IOP over time.”

Three eyes were treated with a glaucoma procedure, anterior chamber paracentesis, which involves removing aqueous humor from the eye. Two of the eyes were in the 8-mg study arms and one was in the 2-mg fellow study arm, Fein said.

A small percentage, 8%, of the patients in the PHOTON study were classified as having a glaucoma-related history, which meant having glaucoma or glaucoma suspect, a term that ophthalmologists use for patients with signs that suggest they are at high risk for developing the disease. Of those patients, 7.7% of those receiving 8-mg injections of Eylea were treated with IOP-lowering drops, compared with 16.7% of those receiving 2-mg injections.

Two other analyses of the large amounts of data generated in the PHOTON study were presented at session at the meeting devoted to findings about high-dose Eylea and Vabysmo (faricimab), an anti-VEGF agent that is FDA approved for treating retinal diseases.

Diana V. Do, M.D., of the Byers Eye Institute at Stanford University, presents findings from the PHOTON study of high-dose Eylea (aflibercept).

One compared patients with poor vision (best-corrected visual acuity [BCVA] 20/50 or worse) with those with better vision (BCVA 20/40 or better). The results shared by Diana V. Do, M.D., vice chair for clinical affairs and a professor of ophthalmology at the Byers Eye Institute at the Stanford University School of Medicine, showed that vision improved among both those with poor and better vision but there were larger gains among those who started with worse vision. A measure of the anatomical effects of Eylea, the change in central retinal thickness, was also more favorable in those with poor vision.

The other study compared patients whose diabetic macular edema had been treated before they enrolled in the PHOTON trial with those who hadn’t been treated. While generally comparable, the results shown by Dennis Marcus, M.D., of Southeast Retina Center in Augusta, Georgia, showed that those who hadn’t been treated experienced the greater gains in BCVA and in central retinal thickness than those who had. “This might be expected,” Marcus told the ASRS audience. “At baseline these eyes [without treatment] had slightly greater severity of retinopathy, leaving more room for improvement.” Marcus noted that the group of patients who had been previously treated with 8-mg Eylea at 16-week intervals experienced lower gains in vision than others in the study.

SGLT2s edge out GLP-1s

The sodium-glucose cotransporter-2 (SGLT2) inhibitors are associated with a lower risk of retinal complications more than the glucagon-like peptide-1 (GLP-1) receptor agonist drugs, but the GLP-1s are comparable to older classes of glucose-lowering drugs used to treat diabetes, according to research presented at the ASRS meeting.

“From my perspective, the fact that the GLP-1s did not cause an increased risk of [retinal] complications compared with the older medications is hugely reassuring,” said Andrew Barkmeier, M.D., an associate professor of ophthalmology at the Mayo Clinic, who presented the findings.

The two classes of older medications included in the study were the dipeptidyl peptidase-4 (DPP-4) inhibitors, a class that includes Januvia (sitagliptin) and Tradjenta (linagliptin), and the sulfonylureas, which include gliclazide, glipizide and glimepiride.

Barkmeier said it will take more research to sort out the difference between the SGLT2 inhibitors, a class that includes Invokana (canagliflozin), Farxiga (dapagliflozin) and Jardiance (empagliflozin), and the GLP-1s, which include Ozempic (semaglutide) and Victoza (liraglutide). For weight loss, semaglutide is sold under the brand name Wegovy and liraglutide as Saxenda.

In the findings he presented at the meeting in Stockholm, the SGLT2 inhibitors were associated with a 27% lower risk of treatment for sight-threatening retinopathy than the GLP-1s. The study encompassed four years.

When asked why the SGLT2 inhibitors might be associated with less retinopathy, Barkmeier answered, “That’s the question.” He said the finding about the advantage of the SGLT2 inhibitors was “hypothesis generating” rather than definitive.

Andrew Barkmeier, M.D.

Barkmeier noted that preclinical findings for the classes of blood glucose-lowering drugs show that they may have different direct effects on retinal tissue apart from their effects on blood sugar levels. Differing rates at which they lower blood sugar levels may also be a factor.

The superiority of the SGLT2 inhibitors in this area runs counter to the stream of recent research findings that has elevated the GLP-1s to almost wonder drug status. Researchers have found suggestive associations between the GLP-1s and a lower risk of everything from kidney disease to many cancers to Alzheimer disease.

There have, though, been some notable exceptions to the GLP-1 winning streak. Researchers have found have an elevated risk of diabetic retinopathy during early treatment with GLP-1s, a finding that fits with a body of research linking sudden drops in blood sugar levels to diabetic retinopathy in the short run even though lower blood sugar levels are demonstrably beneficial for the eyes, especially the retinas, in the long run.

In addition, findings reported in JAMA Ophthalmology in early July 2024 showed a possible connection between semaglutide and a potentially sight-threatening optic nerve condition called nonarteritic anterior ischemic optic neuropathy (NAION).

Barkmeier said his study had been designed to detect diabetic retinopathy differences among the four classes of drugs included in the study but not NAION.

Barkmeier tapped into an Optum database to conduct his study that included people insured by commercial insurers, Medicare Advantage plans or traditional Medicare. The people included in the study had type 2 diabetes with a moderate risk of cardiovascular disease but no history of advanced retinal complications from diabetes. Barkmeier used various statistical techniques to make his analysis resemble that of a clinical trial, comparing the patients treated with the four different classes of medication. Altogether, the study population included 371,698 patients, of whom 42,265 were treated with GLP-1s; 53,476 with SGLT2 inhibitors; 78,444 with DPP-4 inhibitors; and 197,513 with sulfonylurea agents.

Barkmeier found that in addition to being associated with a lower risk of retinopathy compared with the GLP-1s, SGLT2 inhibitors were 21% less likely to be associated with retinopathy compared with the DPP-4s and 29% less likely than the sulfonylureas agents.

The GLP-1s were associated with a similar risk of retinopathy as the DPP-4s and sulfonylureas, he reported.

Barkmeier said this study was motivated by concern that the studies that led to the approval of the new medications did not examine eye complications in depth.

“Our main question, with broad strokes, was, ‘Do these new medications that are powerful and have clear systemic benefits for heart and kidneys and obesity, are they causing retinal complications or problems?’” said Barkmeier. “Because their use is ramping up so quickly. And the studies that got them approved and got them ramped up for the systemic indications really do not look at retinopathy.”

For people with diabetes, social factors are a risk factor for vision loss

Social factors can have negative impacts on whether people with diabetes seek care for their eyes, according to a presentation at the ASRS meeting.

“Social determinants of health are the conditions that shape daily living,” Cindy X. Cai, M.D., a professor of ophthalmology at Johns Hopkins Medicine in Baltimore, Maryland, said during her presentation. “We have an increasing wealth of evidence suggesting that social determinants of health play a critical role in health outcomes, including in ophthalmology.”

Cai and her colleagues wanted to evaluate the association of social determinants of health with eye care use among people with diabetes.

They used the 2013-2017 National Health Interview Survey, a survey used by the U.S. Department of Health and Human Services to monitor trends in illnesses and disability. Researchers conducted a retrospective analysis of patients 18 years of age and older with diabetes who said they had used eye care services in the previous 12 months.

Cindy X. Cai, M.D.

Cai and her colleagues used the KFF framework for social determinants of health and considered economic stability, neighborhood and physical environment, community and social context, food environment, education and healthcare system. Addressing these factors can help improve health as well as disparities in health and healthcare, according to KFF.

Researchers at Johns Hopkins calculated a score and divided it into quartiles, with quartile 4 representing those with the highest burden related to social determinants of health. The outcome assessed was eye care use in the preceding 12 months. They found that of 20,807 adults with diabetes, 43% had not used eye care in the preceding 12 months and that greater social determinants of health burdens were associated with decreased odds of eye care utilization. Participants in quartile 4 (the highest quartile of adverse social determinants of health burden) had an estimated 58% lower odds of eye care utilization than those in quartile 1. The negative impact of adverse social determinants on eye care use was similar across race and ethnic groups and didn't differ between those with type 1 and type 2 diabetes.“This work emphasizes the interconnectedness of SDOH [social determinants of health] and their impact on health outcomes. Screening and addressing the impacts of adverse SDOH should consider the multidimensionality of these determinants,” the researchers wrote in an earlier version of the study results published in Ophthalmology last year. “Ultimately, addressing adverse SDOH and connecting patients to needed services could be a means by which to improve eye care use among people with diabetes and prevent vision loss.”