Higher Dose, Longer Treatment Interval With Aflibercept Shows Promise in Eye Disease

CHICAGO — Bumping up the dose of aflibercept (Eylea) led to prolonged treatment intervals for neovascular eye disease without a loss in efficacy or major increase in adverse events, two randomized trials showed.

Dosing the VEGF inhibitor at 8 mg every 12 or 16 weeks achieved non-inferiority to treatment with 2 mg every 8 weeks with respect to best-corrected visual acuity (BCVA) in patients with diabetic macular edema (DME). More than 90% of patients randomized to 8 mg maintained the 12-16 week dosing interval through 48 weeks, according to a presentation at the American Academy of Ophthalmology (AAO) meeting.

Also at AAO, investigators presented results from a separate randomized trial involving patients with neovascular age-related macular degeneration (AMD). It showed that 8 mg every 12 or 16 weeks achieved BCVA similar to 2 mg every 8 weeks at 48 weeks.

PHOTON

“The 48-week results show the non-inferiority of 8 mg Q12 or Q16, despite greatly increased dosing intervals,” said David Brown, MD, of Retina Consultants of Texas in Houston. “Remarkably, 91% of patients can be maintained on Q12 and 89% of Q16 and overall, 93% on Q12 or greater. We hope to get this benefit out to our patients in the coming future.”

Brown reported initial results from the phase II/III PHOTON randomized trial evaluating a novel 8-mg formulation of aflibercept. The four-times higher molar dose compared with standard aflibercept dosing is hypothesized to achieve a longer effective vitreal concentration and more sustained inhibition of VEGF signaling.

Investigators in the PHOTON trial randomized 658 patients with untreated DME 1:2:1 to intravitreal aflibercept 2 mg every 8 weeks (2q8), 8 mg every 12 weeks (8q12), or 8 mg every 16 weeks (8q16). Patient age was around 62, about 39% were female, and about 72% were white.

Patients in the 2q8 arm initially receive five monthly doses of aflibercept whereas patients in the two 8-mg arms received three monthly injections. Baseline characteristics did not differ significantly among treatment groups, including a mean hemoglobin A1c of 8% across the three groups.

The primary endpoint was change in BCVA from baseline at 48 weeks, and the trial was statistically powered for non-inferiority. The primary analysis showed improvement in BCVA of 8.7 letters with the 2q8 group, 8.1 letters in the 8q12 group (P<0.0001) and 7.2 letters in the 8q16 group (P=0.0031), meeting prespecified statistical criteria for non-inferiority.

Maintenance of dosing intervals in the experimental arms was a secondary endpoint. The results showed that 91% of patients randomized to 8q12 remained at that interval, as did 89% of patients randomized to 8q16. Overall, 93% of patients randomized to 8 mg maintained dosing intervals ≥12 weeks, said Brown.

Change in central retinal thickness also did not differ significantly among the three groups, ranging from -148 µm in the 8q16 arm to -172 µm in the 8q12 arm.

Rates of adverse events (AEs, all grades) averaged 31.0% across the three treatment arms and did not differ significantly. Rates of individual types of AEs also did not differ across the treatment groups. The incidence of intraocular inflammation averaged 0.8% and did not differ significantly among the treatment groups.

AAO panelist Shlomit Schaal, MD, PhD, of the University of Massachusetts in Worcester, pointed out that the higher dose of aflibercept produces the same number of anti-VEGF molecules as bevacizumab (Avastin) after 4 weeks.

“Wouldn’t you think that rather than increasing the dose, you should focus on other mechanisms of the disease in DME, not just increasing the same molecule?” she asked Brown.

Acknowledging the presence of “millions of cytokines produced by diabetic eyes” as potential treatment targets, Brown said anti-VEGF remains the “big gorilla.” Any strategies that can increase dosing intervals, even modestly, will make a “big difference in this working-class population that has to take off school or get a substitute teacher. Anything we can do to decrease treatment burden [is helpful].”

PULSAR

The same three dosing intervals were evaluated in the phase III PULSAR trial, which included 1,009 patients with untreated neovascular AMD, randomized 1:1:1 to the three dosing schedules. The primary endpoint was change from baseline to 48 weeks in BCVA, said Paolo Lanzetta, MD, of the University of Udine in Milan. Overall, 93.3% of patients (age around 74, more than half female, more than three-fourths white) completed the 48 weeks of treatment and follow-up.

At 48 weeks, all three groups achieved improvement in BCVA from baseline: 7.0 letters with 2q8, 6.1 letters with 8q12, and 5.9 letters with 8q16. The 8q12 (P=0.0009) and 8q16 (P=0.0011) treatment groups demonstrated non-inferiority to the 2q8 group. Overall, 83% of patients randomized to 8 mg injections of aflibercept maintained ≥12-week dosing intervals through 48 weeks.

Significantly more patients in both 8-mg groups had no retinal fluid in the center subfield at 16 weeks, a secondary endpoint. The percentage of patients without fluid in the center subfield was 63% in the two 8-mg groups combined versus 52% in the 2q8 group (P=0.0002).

Mean reduction in central retinal thickness from baseline was 147 µm in the 8q16 group, 142 µm in the 8q12 group, and 126 µm in the 2q8 group.

The incidence of AEs was similar across the treatment groups.