An investigational inflammasome inhibitor reduced geographic atrophy (GA) progression in a phase 2 trial, with additional analyses suggesting that the treatment effect may increase over time. The findings also showed a difference in visual acuity favoring treated eyes in a subgroup of patients with nonsubfoveal GA, although the study was not powered to evaluate functional outcomes.
Liangbo Shen, MD, presented the findings for kamuvudine K8 (Kamzinoflast; Inflammasome Therapeutics) at Retina Subspecialty Day during the American Academy of Ophthalmology (AAO) 2026 meeting in New Orleans. Dr. Shen, a consultant to Inflammasome Therapeutics, reported results from a 6-month, multicenter US study evaluating the investigational intravitreal implant.
Figure 1. In the phase 2 trial, kamuvudine K8 0.7 mg reduced geographic atrophy growth rates by 43% during months 0 to 3 and 70% during months 3 to 6 compared with untreated fellow eyes. The overall reduction was 56%.
K8 is a modified nucleoside reverse transcriptase inhibitor designed to retain inflammasome inhibition while eliminating antiviral activity and mitochondrial toxicity associated with the original drug class. Unlike the approved complement inhibitors pegcetacoplan (Syfovre; Apellis/Biogen) and avacincaptad pegol (Izervay; Astellas Pharma), K8 targets an alternative inflammatory pathway.
“There might be a role for a new mechanism,” Dr. Shen said, citing the limitations of current GA therapies, which he said include modest reductions in lesion growth compared to sham and the absence of significant best-corrected visual acuity (BCVA) benefits in pivotal trials.
The trial enrolled 30 patients with bilateral GA across 9 US centers. Each participant received K8 in the worse-seeing eye, with the untreated fellow eye serving as a control. The treated eyes were divided into 3 dose cohorts of 10 eyes each, receiving 0.3 mg, 0.7 mg, or 1.05 mg at baseline and again at month 3. Each cohort was compared with all 30 control eyes using a mixed model accounting for the paired-eye design. The primary endpoint was mean GA lesion growth rate, assessed by an independent reading center with masked readers.
In the primary analysis, K8 0.7 mg reduced GA growth rate by 54% compared with control eyes. Dr. Shen also reported that 25 of the 30 participants had slower GA progression in their treated eyes than in their fellow control eyes. The 0.7-mg and 1.05-mg cohorts demonstrated statistically significant reductions in progression, whereas the 0.3-mg cohort showed a numerical trend favoring treatment.
A piecewise analysis of the 0.7-mg cohort found that GA growth was reduced by 43% during the first 3 months and by 70% during months 3 through 6, compared with control eyes (Figure 1). The overall reduction in this analysis was 56% (P=.0084).
“We need a longer study and a larger study to further demonstrate the increasing effect,” said Dr. Shen.
The study also examined BCVA in a prespecified subgroup of patients with nonsubfoveal GA. Among 14 treated and 16 control eyes, the covariate-adjusted analysis found a 4-letter advantage favoring K8 (P=.004). “We still need a larger study to validate the functional benefits,” Dr. Shen observed (Figure 2).
No drug-related serious adverse events or dose-limiting toxicities were reported through month 6. Investigators also observed no cases of endophthalmitis, intraocular inflammation, neovascular AMD, retinal vasculitis, or optic neuropathy. The small study population, however, limits conclusions about uncommon adverse events, cautioned Dr. Shen.
K8 is delivered through a bioerodible implant using a preloaded 24-gauge injector, with sustained drug delivery over approximately 3 months. The formulation does not require refrigeration. Dr. Shen said that Inflammasome Therapeutics is planning a global phase 3 pivotal trial to further evaluate K8 in GA. RP
Figure 2. In a prespecified analysis of eyes with nonsubfoveal geographic atrophy, kamuvudine K8 was associated with a covariate-adjusted 4-letter advantage in best-corrected visual acuity compared with untreated control eyes. The study was not powered to assess visual acuity.