This article was originally published in a sponsored newsletter.
For this month's Retina minute, I have the pleasure of interviewing Dr. Alex Melamud, who is a partner at Retina Group of Washington.
Dr. Melamud, recently at Retina Subspecialty Day, you gave an interesting lecture on the use of an oral drug for the treatment of geographic atrophy (GA). Can you tell us about the lecture and what was involved in the study and the results?
Certainly. It is a pleasure to be here and I am looking forward to spending a few minutes discussing this. SAGA is a phase 2/3 clinical trial that evaluated the safety and efficacy of the investigational agent, gildeuretinol. Gildeuretinol is a new chemical entity where 3 hydrogen atoms on the C-20 position of retinol are replaced with deuterium to make the molecule heavier. The retina is a high metabolic user of retinol which makes gildeuretinol useful.
The science behind this drug is that clumps of vitamin A accumulate as metabolic waste products. This happens not only with age, but also in certain disease states such as Stargardt's disease with a malfunctioning receptor. These receptors are ATP-activated and as ATP decreases with age, you get accumulation of vitamin A within cells. When these clumps or “dimers” form, vitamin A cannot cycle through the cell membrane.
Vitamin A dimers–also sometimes called toxic dimers–have been implicated in a number of disease states, both experimentally and clinically. And so the scientific premise behind gildeuretinol is that if you can reduce the dimerization–the clumping of these molecules–you can slow disease progression.
Tell us about the study.
SAGA was a 2-year study. Patients were seen at 6-month intervals, and the final follow-up was at month 24. The primary efficacy endpoint was the mean rate of GA lesion growth which was measured by fundus autofluorescence and OCT. Secondary endpoints were low-luminance visual acuity and best-corrected visual acuity. Patients were included in the study if they had a well-demarcated GA lesion. Eyes with both subfoveal and non-subfoveal involvement were included. Patients who had choroidal neovascularization in the study eye were excluded, but patients could have choroidal neovascularization in the fellow eye and still be included in the study.
Patient characteristics were generally balanced between the treatment and placebo groups. The mean age of patients was 78 years old, and predominantly female. Most lesions were subfoveal and multifocal. These characteristics fall in line with other studies in this space. As far as patient follow-up, approximately the same number of patients completed the study in both the treatment group and the placebo group. Specifically, 70% of patients in both groups completed the study. The most common reason for discontinuation in both arms was withdrawal by subject.
The primary efficacy endpoint did not achieve statistical significance at month 24, but there was a clinically meaningful separation of effect noted starting at month 6. Lesion growth was 13.4% slower in the gildeuretinol group than in placebo, but this rate did not reach statistical significance. However, a pre-specified sensitivity analysis was performed. The rationale here was that in pharmacokinetic studies, it has been shown that oral gildeuretinol takes several months to replace vitamin A in the body, so their pre-specified analysis looked at month 6 (first follow-up) to month 24 (endpoint). There was a statistically significant 15.3% slower lesion growth rate in patients who were on oral gildeuretinol, which was statistically significant. Equally as interesting was the secondary analysis on low-luminance visual acuity. Many of our patients with dry AMD complain of difficulty with vision and low-light, dark adaptation. This experience has been shown experimentally and actually linked with dimerization. Here, in the secondary analysis on low-luminance visual acuity, patients in the gildeuretinol group lost 4.4 fewer letters of low-luminance visual acuity than patients who took placebo, which was statistically significant.
Regarding best-corrected visual acuity, there was a trend at final follow-up toward fewer letters lost in the gildeuretinol group, but this finding was not statistically significant.
Another important feature here is gildeuretinol’s safety profile. We found that the treatment-emergent adverse events in both the gildeuretinol and placebo groups were balanced. There were no deaths attributed to the study drug; no reports of side effects due to hyper- or hypovitaminosis A, vasculitis, intraocular inflammation or endophthalmitis, and no difficulty with dark adaptation or chromatopsia. Most of the treatment-emergent adverse events in both groups were mild or moderate in severity.
Interestingly as well, there was a lower rate of choroidal neovascularization-related adverse events in at-risk patients in the treatment group. In the placebo group, more patients developed choroidal neovascularization in the study eye. However, no patients developed choroidal neovascularization in the gildeuretinol group; even those patients who had choroidal neovascularization in the contralateral eye.
These data obviously need more analysis and a deeper dive, but the safety profile as we understand it now appears impactful and exciting. I am hopeful that we will eventually have FDA approval and gildeuretinol will have a place in the toolbox for retina specialists.
These data are encouraging. They sound similar to the original pegcetacoplan (Syfovre, Apellis) data. Is there an extension study planned on this patient population?
Yes, it does sound encouraging, but gildeuretinol is quite different from Syfovre. Syfovre is an intravitreal injection and the mechanism of action is complement inhibition at C3 and C3b. Gildeuretinol is a new oral chemical entity and its mechanism of action is to slow dimer formation in the visual cycle. SAGA was only a phase 2/3 study and it still warrants a larger phase 3 program. However, the results of SAGA are encouraging, especially with a favorable safety profile. It will be exciting to see Alkeus’s next steps.
Let’s return to the difference in lesion growth at 24 months because it would be different than the straight growth over time and has been shown in both the Astellas and the Apellis trials where we looked at the rate of growth over time, first at a 6-month interval, then a 12-month interval, and finally at 18 and 24 months. Was that analysis performed?
That analysis has not been done by Alkeus at this time, but is something they may consider in the future. Currently the only analyses that were performed were evaluating the difference in GA lesion growth vs placebo from baseline to 24 months and 6 to 24 months. I want to reiterate the importance of the 6- to 24-month cut because it was statistically significant. Pharmacokinetic data has shown that it may take several months to have an 80% replacement to gildeuretinol systemically, so a treatment effect may take time to be seen. The primary endpoint was at 6 to 24 months for this reason in the gildeuretinol program for Stargardt disease.
If gildeuretinol is approved, how would you incorporate it into your arsenal when you see patients with GA?
Based on the current data and the favorable safety profile, I am most impressed by the protective effect on low-luminance visual acuity because low-luminance visual acuity is an early signal of disease state. We do not have another oral agent in this space yet that demonstrates a protective effect on visual acuity. It also does not have the risks of intravitreal injections. So to me, this would be a drug we could start early and potentially use with advanced disease states in combination with other therapies to achieve the maximum benefit.
This is fascinating. What is especially interesting is that, with its positive visual signal, gildeuretinol could be used in conjunction with other complement inhibitors on the market and may be able to prevent the development of choroidal neovascularization. As we start getting more and more medications in our armamentarium to treat this disease, I agree that this may have a place. I am looking forward to additional trials to see how patients do.
Yes, my feeling is that there is a place for multiple treatments and combination therapy. Given a good safety profile, I have a lot of hope that this will be something that retina specialists will be able to use in the future.
