High-Dose Eylea Shows Unique Properties in nAMD and DME | AAO 2024

Some patients can extend the dosing interval with Eylea (aflibercept) from 8 weeks to as long as 20 weeks as a treatment for neovascular age-related macular degeneration (nAMD) or diabetic macular edema (DME), and a new statistical model and simulation may shed light on the underlying mechanisms that determine why this is the case.

The findings, which were presented at the poster theater at the 2024 annual meeting of the American Academy of Ophthalmology (AAO), also highlighted some potential differences between the 2-mg and 8-mg formulations, the latter of which is marketed as Eylea HD.

Michael W. Stewart, M.D., Mayo Clinic

Image credit: Mayo Clinic

"Aflibercept 8-mg is a new drug formulation that delivers four times the molar dose as the pre-existing aflibercept 2-mg," Michael W. Stewart, M.D., from the Mayo Clinic in Jacksonville, FL, said during the presentation. "The molecule is exactly the same but the difference between the 8-mg and the 2-mg is the formulation. We have different excipients, we have different pH, and as best we can guess we think it's part of the formulation that allows the drug to stay in the eye longer and decrease its clearance."

The modeling used Population Pharmacokinetics (PopPK) and two exposure-response models and showed that Eylea 8-mg was cleared 34.4% slower than the 2-mg dose, showing longer durability in the eye. Moreover, effective concentrations of the high-dose drug were still seen in the eye 6 to 8.9 weeks longer compared to the reference product. The modeling further showed a correlation between dosing interval, central retinal thickness (CRT) and ocular clearance levels.

"When it comes to deciding which drug to use, there is equal efficacy, extended durability, and equal safety factors. The biggest deciding factor, comes down to cost differentials," Stewart said. "More and more I think we'll see physicians using the 8-mg dose instead of the 2-mg for the durability advantages."

For the study presented at the AAO meeting, researchers created a semi-mechanistic model using 2,744 individuals from across 16 clinical trials exploring Eylea at difference doses and administration routes. The model was based off free and adjusted bound Eylea concentrations from the plasma. From these initial patients, a PopPK model created a virtual dataset of 10,000 simulated patients with nAMD or DME and predicted their free Eylea concentrations.

Two separate exposure-response models were also created to examine the patient characteristics of those treated with 8 mg Eylea. These models included longitudinal dosing interval data, looking at time to first dosing interval in 726 patients with nAMD from the CANDELA and PULSAR trials and 491 patients with DME from the PHOTON trial. Additionally, it looked at time to first dosing interval extension in 621 patients with nAMD from the PULSAR trial and 441 patients with DME from the PHOTON trial. It also included post-hoc collected Bayesian PopPK estimates.

The PopPK simulation estimated ocular clearance of Eylea 2-mg was 0.625 ml per day whereas the clearance was 0.410 ml/day with the 8-mg formulation of the agent. The model further projected various levels of concentration of the drug across time on treatment. With the 8-mg dose of Eylea specifically, the model predicted nearly half of patients would have effective in vitro concentration at 20 weeks.

The exposure-response model predicted area under the curve for nAMD and DME patients. These findings found the interval of Eylea treatment was typically decreased as ocular clearance increased and when baseline CRT increased. The dosing interval was increased when the clearance decreased, specifically for nAMD, and when baseline CRT decreased in both patients with nAMD or DME.

"There are, of course, limitations to this research. This PopPK model used plasma samples. It did not use samples obtained from intraocular specimens," said Stewart. "The second thing is that we assume there is no change without the clearance. And thirdly, we assume that the eye is a single compartment and that the vitreous and aqueous are one compartment. We could not parse out the vitreous and aqueous concentrations within this model."

Although limitations exist, Stewart noted that it shed light on the differences between the two formulations. He also added that outside of these factors the disease progression, comorbidities and variability in treatment response also play a role in the dosing interval modification.