For this month's Retina Minute, I have the honor of interviewing my dear friend Priya Vakharia, who is a retina specialist at Retina Vitreous Associates of Florida.
Recently, Dr. Vakharia, at the Retina World Congress meeting, you presented some interesting data on a new compound that is being studied for the treatment of geographic atrophy (GA). Can you tell us about the study and what you found?
This was a cool presentation, Dr. Singer. To recap, we talked about a particle called ANX007, which is produced by Annexon. In the phase 2 ARCHER trial, they studied GA patients who received ANX007 monthly or every other month, compared to sham monthly or sham every other month. Like other GA trials, they assessed the rate of change of GA lesion size.
The investigators followed these 4 groups of patients for over a year and then for an additional 6 months for safety. They did not meet their primary endpoint of change in GA lesion growth, but interestingly, in a prespecified secondary analysis of visual acuity outcomes, they found that patients who received ANX007, especially monthly but also every other month, had less 15-letter loss of vision compared to sham at 2 consecutive visits. The numbers were pretty striking.
Further, the number of patients who had confirmed 15-letter loss through month 12 actually ended up being higher in the sham group at 21.3% compared to 5.6% in the ANX007 monthly arm and 9.8% in the ANX007 every-other-month arm.
This secondary analysis showed a trend toward something that we have all been looking for: visual acuity results. And this is the first known clinically meaningful preservation of vision in GA. Given these results, the investigators changed their new primary endpoint to BCVA of 15-letter loss or more in the phase 3 trial that will follow ARCHER.
In my presentation at Retina World Congress, I also mentioned that 7 patients would have needed to be treated in the ARCHER trial to prevent 1 patient from developing clinically significant vision loss. For me, that puts into perspective what could be possible with this molecule.
How does that play out?
The number needed to treat is calculated by looking at 1 over the event rate in sham minus the event rate in the monthly group. What this means is the number of patients who need to be treated to prevent 1 additional adverse outcome. In this case, you need to treat 7 patients to prevent 1 from having clinically significant vision loss.
Let's put that in perspective. We often see patients who are on aspirin or statins to prevent cardiovascular issues. The number needed to treat to prevent stroke with statin therapy is 125, and the number needed to treat to prevent a nonfatal heart attack with aspirin therapy is 333. A number needed to treat of 7—in the case of ANX007—could be more favorable compared to medications that we see prescribed all the time.
How is this medication being delivered?
ANX007 is delivered via an intravitreal injection that is given like a standard intravitreal injection through the pars plana.
Tell us about how the phase 3 trial is designed and how it is going.
There are only 2 arms in the phase 3 trial: Patients either get ANX007 monthly or sham monthly. That trial is currently enrolling with a primary endpoint looking at 15-letter loss of vision and several key secondary endpoints. It will be interesting to see what the results show. I think at the end of the day, what most of our patients are looking for is not necessarily reduction in GA lesion size; they are looking for improvement in vision or preservation of vision.
How long is the phase 3 study?
The primary endpoint for the ARCHER II study is at month 12, looking at BCVA protection against that 15-letter loss, but they are continuing out to 2 years to assess safety. Secondary endpoints include low luminous visual acuity and the rate of change in GA lesion size. This will be the first phase 3 trial for GA we have had to date that is looking at an endpoint of vision.
Other studies were done using noncenter-involved lesions, and some were done with both center-involved and noncenter-involved lesions. What sort of lesions are being studied in the ARCHER II trial?
They are allowing both center-involved and noncenter-involved. However, the visual acuity needs to be at least a Snellen equivalent of 20/100 or better. They are not allowing patients who have already had significant vision loss, but there is no requirement for foveal or nonfoveal involvement.
How about history of choroidal neovascularization (CNV)?
Patients cannot have a history of CNV in their study eye, but they can in the fellow eye.
One of the other things that has been interesting is an increased focus on the ellipsoid zone, more so than looking at GA on fundus autofluorescence. In the phase 2 trial of ANX007, the investigators found that the compound provided approximately 60% protection of photoreceptors compared to sham within the central 1.5 mm of the fovea. That was measured by ellipsoid zone changes through the 12-month endpoint. Protection was approximately 29% for the full retinal field.
I think that will be another interesting thing that we look at going forward—specifically looking at the ellipsoid zone rather than looking at full-thickness GA.
This is fascinating information. Obviously, we are still trying to find a therapy for GA that potentially affects vision, so hopefully ANX007 will be acceptable to physicians, patients, and regulators, both domestically and internationally, to help us move the whole field forward.
Thank you. I agree that it is exciting, and I think it has a lot of potential in the phase 3 trial.
