Angiogenesis 2023: Efficacy and safety of high-dose aflibercept for treatment of DME

David S. Boyer, MD, discusses the PHOTON study examining the safety and efficacy of high-dose aflibercept for the treatment of diabetic macular edema (DME) from his presentation at the Angiogenesis, Exudation, and Degeneration 2023 conference with David Hutton, Executive Editor, Ophthalmology Times®.

Video transcript

Editor’s note: This transcript has been edited for clarity.

David Hutton:

I'm David Hutton of Ophthalmology Times. Dr. David Boyer made a presentation on the safety and efficacy of high dose aflibercept for the treatment of DME at the Angiogenesis 2023 conference. Thanks for joining us today. Tell us a little bit about your presentation.

David S. Boyer, MD:

Thank you very much. It was my pleasure to present the results of the PHOTON study, treating DME. As we all know, the current state of care of clinically significant macular edema is anti-VEGF therapy. And the thought behind this study was that if we increase the Mueller dose of intravitreal, anti-VEGF therapy, that perhaps we can reduce the treatment burden to our patients, and on a hypothetical basis that was found, and indeed it even in animals, it was found that it was helpful in reducing the time interval between the need for retreatment in rabbits that were treated ongoing.

So there has been clinical evaluation now, that shows that an 8 milligram dose, 4 times the normal dose, injected in 70 microliters of fluid into the eye will give you a longer time period before you need additional treatment.

So the study design was very easy. They compare the 8 milligram dose at 12 weeks and 8 milligram dose at 16 weeks, to the 2 milligram currently available dose at 8 weeks. The primary endpoint was at week 48. And what they did is they treated according to label in the 2 milligram dose, which required 5 initial injections, and then every 8 weeks after that, the 12 week, dose of 8 milligrams in the 16 week, dose of eight milligrams had 3 initial monthly injections, reducing the treatment burden. And then the patients were randomized either to 12 weeks or 16 weeks. And this was a worldwide study.

The inclusion criteria were fairly stable for all as compared to other studies that have been done, and included type 1 or type 2 diabetes with macular involvement of center retinal thickness and greater than 300 with the exclusion of active PDR, or status post-PRP or any anti-VEGF therapy within 12 weeks of screening.

So basically, patients received either 5 initial doses if they were in the 2 milligram dose, and then every 8 weeks. The 12-week dose of 8 milligrams was 3 times in the beginning and then went out to week 20, then 1 week out to 32 weeks, 16, again, 3 initial doses, that 8 milligrams at 16 weeks, then initial of dose of 24.

Now the patients could go backwards, if they didn't meet certain criteria as compared to week 12, when they were seen. It was a 10 letter loss and best corrected vision due to persistent fluid and a 50 micron increase in the central retinal thickness.

At week 16, the patients on the Q 12 and the Q 16 weeks can be decreased at month 24. They could also be decreased at week 32 and week 44, again, a decrease to a lower level. So in other words, if you were at 16 weeks, and you met the criteria, you would go to 12. Nobody went lower than 8. So if you were at 12, and you met the criteria, you would go to 8.

So the patient disposition was interesting. Almost 92% of patients completed the trial, which is really amazing. That was a very high percentage. And the baseline demographics were fairly equal in all groups, patients, hemoglobin, A1C was approximately 8 across. Almost 80% of the patients had hypertension, and the remainder of the age race all appeared to be fairly well balanced. Baseline characteristics of the study eye, almost over 40% of patients had previous treatment, at some point for their DME.

The number of injections was definitely reduced. In the Q 8, almost 8 injections were given over the 48 week period, 6 injections in the 12 and 5 injections in a 16 week.

The end result, the primary endpoint, was met with both 8 milligram doses, there was almost a 1.3 letter difference between the Q 8 and the Q 16. With almost 9 over 9 letters, 9.2 letters in the queue 8 and 7.9 letter gain the Q 16. With 8.8 and the Q 12. All were statistically significant.

If you look at the mean change of vision, by baseline best-corrected vision, you'll see that the patients that had poor vision gain more. The patients who had better vision gained a little bit less, which is what we would have expected as they have a ceiling effect if you have very good vision to begin with.

Again, changes in the CRT again showed if you were less than 400 microns, you had improvement of vision in all subgroups. That Q 8, 2 milligram dose the 8 milligram dose Q 12 and Q 16. And if it was greater than 400 microns, you had a bigger improvement of vision, again, equal across all groups. The 2-step improvement was about 25% at week 12, and that continued to be 48.

But if you look at the patients who had the 47 or worse, the 2 step improvement, rivaled what we saw in PANORAMA. Over 70% of patients did get an improvement, the highest being 77%. In the Q 8 with 63% and the Q 16.

The mean change in subretinal thickness was fairly stable. There were fluctuations, there was some saw toothing that was noted. The saw toothing appeared to reduce as we got further out in the study, it was more apparent earlier in the study, and then began to lose its effect.

In the Q 12, 91% of patients met remained in the Q 12, only 9% of patients went to Q 8. In the Q 16 about 7% of patients had to go down to Q 12. And we maintain there with only 4% going to Q 8. And if you look at the overall 93% of all patients went Q 12 or greater at the end of the trial. So this was maintained Fluorescein leakages also decreased fairly equally in all groups.

The most frequent ocular issues were things that were related to basically the injection technique, there was no real difference between any of the groups, interactive inflammation was minimal. And again, there was no difference in any of the different groups, including the intraocular pressure, which we would have expected possibly to go up is there have been reports of chronic anti-VEGF treatment, giving rise in pressure and also giving a little bit of a larger dose than we would normally give a 0.05ml.

There was really no difference. The ocular safety, obviously, extremely important, you're giving much higher dose of aflibercept. We were worried about APTC changes, and there was really no change compared to the 8 milligram dose. And overall, it was what is seen in other clinical trials, no problems.

So safety was definitely comparable to the 2 milligram dose. There were no cases of endophthalmitis and certainly no cases of inclusive retinal vasculitis, which you’ve seen with other products recently. And the APTC rates and death were similar between all groups.

So, in conclusion, it appears that the 8 milligrams Q 12 and 8 milligrams Q 16 were non-inferior to our present 2 milligrams Q 8, reducing the treatment burden considerably.

At Q 12 weeks, the 8 milligram had a non-inferiority margin of 50% and the proportion of patients with 2-step improvement in DRSS are very comparable. At the end at 8 milligrams randomized at different treatment intervals. 91% of patients maintained the Q 12 89% maintain Q 16. And overall 93% of patients went 12 weeks or greater.