Vamikibart, an investigational interleukin (IL)-6 inhibitor that could be the first intravitreal nonsteroidal treatment for uveitic macular edema, demonstrated rapid improvement in vision that lasted for up to a year after starting treatment with a low incidence of retreatment, but twin phase 3 trials showed inconsistent results between dose levels.

“We saw rapid improvements in best-corrected visual acuity [BCVA] in all vamikibart arms as early as day 8 and that continued through week 16 in both trials and also continued through week 52 in both trials,” Nisha Acharya, MD, MS, director of the Uveitis and Ocular Inflammatory Disease service at the University of California, San Francisco, reported at the American Academy of Ophthalmology (AAO) 2026 Annual Meeting in New Orleans.

Acharya reported 52-week study results from the MEERKAT (n = 245) and SANDCAT (n = 256) trials that randomly assigned patients to receive either a 1-mg or 0.25-mg dose of vamikibart or sham treatment. Glenn D. Jaffe, MD, chief of the Vitreoretinal Disease Division at Duke Eye Center in Durham, North Carolina, reported the primary 16-week outcomes from the trials. The study participants received four monthly injections of vamikibart or sham, followed by as-needed doses from week 20 to week 48.

Vamikibart is an anti-IL-6 monoclonal antibody. In an interview, Jaffe noted the phase 1 DOVETAIL trial showed that IL-6 levels were elevated in eyes with uveitic macular edema. “Vamikibart caused a rapid and pronounced reduction of aqueous IL-6 levels,” he said.

16-Week Primary Endpoint Results

In MEERKAT, at the 16-week endpoint, 26.1% of patients who received 0.25 mg of the drug, 43.2% of those who received the 1-mg dose, and 6.3% of those in the sham group experienced an improvement in BCVA of ≥ 15 letters on the Early Treatment Diabetic Retinopathy Study (ETDRS) scale (P = .0008 and P < .0001, respectively). In SANDCAT, the proportions were 34% and 21.8% of patients in the treatment groups vs 13.3% of those in the sham group (P = .0019 and .1487, respectively).

The proportions of patients with an improvement of ≥ 10 letters at week 16 were 46% and 61% in the 0.25- and 1-mg vamikibart groups, respectively, compared with 16% in the sham group (MEERKAT) and 54% and 45% in the 0.25- and 1-mg vamikibart groups, respectively, vs 24% in the sham group (SANDCAT), Jaffe reported.

Among patients who had worse than 20/40 vision when they enrolled in the trial, the vision in about 50% improved to 20/40 or better by week 16 in both treatment groups in the MEERKAT and SANDCAT trials — 52% and 49% in the 0.25-mg group and 52% and 44% in the 1-mg group vs 22% and 25% in the sham group, respectively, Jaffe said.

The average changes in BCVA after 16 weeks were gains of nine and 12 letters in the 0.25-mg and 1-mg treatment groups vs three letters for sham in MEERKAT and 11 and nine letters vs five for sham in SANDCAT, he reported.

52-Week Results

At the 52-week mark of the MEERKAT trial, the proportions of patients who had ≥ 15-letter improvement were 28.3% and 21.7% in the 0.25- and 1-mg groups vs 6.3% (P = .0003 and .0054, respectively). In SANDCAT, 23.4% of patients in the 0.25-mg group and 17.3% of those in the 1-mg arm experienced a ≥ 15-letter improvement vs 12% of the sham patients (P = .0579 and .3436, respectively).

The differences in MEERKAT were “statistically significantly better” at the 16-week endpoint, but at week 52, they were only “nominally significant,” Acharya said. In SANDCAT, the 16-week endpoint for the 1-mg dose was not statistically significant compared with sham, but that for the 0.25-mg dose was considered “nominally significant,” she said. The 52-week differences in SANDCAT did not meet statistical significance, the researchers reported.

The rapid vision improvements in the treatment groups that appeared as early as 8 days after starting treatment continued through to week 52 compared with the sham group, Acharya said. However, in the SANDCAT trial, the improvements between the 1-mg and sham groups narrowed to a gain of seven and six letters on the ETDRS scale, respectively, representing a statistically insignificant difference (P = .6311). In all other treatment groups in both trials, the improvement was statistically significant.

The 52-week results also looked at the proportion of patients who needed retreatment or rescue treatment with vamikibart as needed after week 4 onward. Worsening vision, macular edema, or uveitis were criteria for retreatment, Acharya said. Patients who did not need rescue treatment by week 20 could receive as-needed dosing through week 48.

Overall, about 20% of the vamikibart patients received rescue treatment vs about 50% in the sham group, she said. Among the vamikibart patients who did not meet the criteria for rescue treatment, about 30% required as-needed dosing after week 20 and 85% of them needed one dose, Acharya added.

Anatomic Changes

Acharya additionally reported the anatomic changes in the treatment groups at week 16 did reach statistical significance compared with sham in both trials. In MEERKAT, central subfield thickness, a key biomarker of retinal inflammation, decreased by 187.6 µm and 196.1 µm in the 0.25- and 1-mg vamikibart groups, respectively, compared with a reduction of 57.1 µm in the sham group (P < .0001). In SANDCAT, the reductions were 209.6 µm and 193.3 µm, respectively, vs 43.3 µm in the sham group (P < .0001).

Visual improvement in the 1-mg vamikibart group did not reach statistical significance in SANDCAT, but Acharya and Jaffe said that the overall results were noteworthy.

“The treatment effect was durable; the visual acuity remained relatively stable after the endpoint at 16 weeks up through the end of the study at 52 weeks,” Jaffe said. The as-needed injection rate averaged about one between weeks 16 and 52, he added, and rescue treatment for inadequately controlled inflammation was much less common in the vamikibart treatment groups than in the sham groups.

“Uveitis is a very heterogeneous group of conditions, so vision outcomes may be affected by the underlying uveitis or how long the uveitic macular edema was present,” Acharya told Medscape Medical News. “It’s important to look at the totality of data from both trials, as well as safety outcomes. We are in need of additional nonsteroid treatment options for this condition.”

Overall, safety was “quite good” in both trials, Acharya said. Up to week 16, 33.3% of patients in the 0.25-mg group and 32.9% in the 1-mg group experienced a safety issue across both studies compared with 25.9% of sham. From week 16 to 52, the proportions were 21.5%, 27.0%, and 32.9%, respectively. The rate of serious events was also similar across the study groups: 1.3% for 0.25 mg, 0.6% for 1 mg and 1.2% for sham treatment up to week 16; and 0%, 1.2% and 0%, respectively, in the latter part of the trial. No episode of occlusive retinal vasculitis was reported, Acharya said.

A Potential Alternative

A therapy that inhibits IL-6 can provide an alternative to corticosteroids for the treatment of uveitic macular edema, according to Cesar Michael A. Samson, MD, MBA, director of uveitis services at NYU Langone Health in New York. “This can be particularly valuable in patients vulnerable to complications from steroids, such as patients who develop steroid-induced glaucoma,” he said.

The low rate of as-needed treatment after 20 weeks reported in the trials was “encouraging,” Samson said, and suggests that “many patients can maintain their initial improvement without requiring additional injections through week 48 beyond the initial four monthly injections.”

One issue that raised questions about the trials was the inconsistent effectiveness of the 1-mg dose between the two studies, he said.

“In MEERKAT, 43.2% of patients receiving 1 mg achieved a 15-letter improvement, compared to 26.1% receiving 0.25 mg. However, in SANDCAT, the lower-dose arm did better, with 34% responding to 0.25 mg vs only 21.8% to 1 mg,” he said. “So there was not a consistent dose-response relationship for visual outcome, despite good anatomic efficacy at both doses.”

Samson also wondered why the anatomic response did not translate to an equivalent improvement in BCVA.

“Hypotheses to explain the observed discrepancy could involve factors such as the stage of disease of certain patients, biases introduced by the study design, or yet unknown mechanisms of the drug at higher doses,” he said. “No conclusions can be made without more details of the study.”

Roche funded the clinical trials. Acharya and Jaffe are consultants to Roche, which funded MEERKAT and SANDCAT. Samson reported no relevant financial conflicts of interest. Some of the researchers on the trial presentations are Roche employees.

Richard Mark Kirkner is a medical journalist based in the Philadelphia area.