麻豆传媒

SAN FRANCISCO -- A neuroprotective approach to geographic atrophy (GA) had a time-dependent effect on lesion growth and a significant impact on visual acuity, an early study showed.

Overall, the complement inhibitor ANX007 did not significantly slow the growth of GA lesions as compared with sham injections. An analysis of effect over time showed that therapy did not affect lesion growth during the first 6 months but substantially reduced the rate of growth thereafter to month 12.

ANX007 afforded significant protection from vision loss throughout the study, as more than twice as many patients in the sham group had vision loss ≥15 letters, reported David Lally, MD, of New England Retina Consultants in Springfield, Massachusetts, at the American Academy of Ophthalmology meeting.

"ANX007 demonstrated a novel neuroprotective mechanism for GA that consistently translated into clinically meaningful protection of vision that was statistically significant," said Lally. "[The effect on vision] was time and dose dependent and it was robust across multiple measures and patient subgroups. Importantly, it was not dependent on lesion growth."

"ANX007 offers the opportunity to transform the treatment paradigm for GA, offering new hope for protecting vision for our GA patients. Phase III preparations are underway."

The data on the time-dependent effects of the drug on GA were intriguing, said Mark Humayun, MD, PhD, of the University of Southern California in Los Angeles, during a panel discussion that followed the presentation.

"The first 6 months looks like little was happening and the second 6 months, a lot is happening," said Humayun. "Do you have any thoughts about why there's that 6-month initial period that is not quite as responsive as the second 6 months?"

The currently available drugs for GA block complement C3 and C5, which are involved in the removal of dead cells, noted Lally. ANX007 blocks C1q, which does not regulate removal of dead cells.

"What we may be seeing in those first 6 months is a delay in terms of the cells dying, where during the first 6 months, some of the RPE [retinal pigment epithelium] cells are too sick, and they're on the way out the door with C3 and C5 alternative lectin pathways progressing," he said. "As you keep the photoreceptors healthier over time, you may see those RPE cells survive later on."

Some experts have expressed concern about over-regulation of complement, and "complement is how we don't die, don't develop cancer, and so forth," said panelist Suber S. Huang, MD, of the Retina Center of Ohio in Cleveland. "What are you thinking about unexpected side effects of upstream inhibition of complement?"

Complement C1q is classical complement and is involved in activation of macrophages, said Lally.

"In preclinical studies, I'm not aware of any side effects," he said. "That's different from blocking C3 in mice, where you can see neurotoxicity. We're going to look at the clinical data, but what we have seen in this study is really encouraging in terms of side effects. This is the first blocking of complement that we did not see increased rates of CNV [choroidal neovascularization] at 12 months."

Complement C1q is a key driver of neurodegeneration and plays a major role in classical pathway activation on photoreceptor cells to induce inflammation and cell loss. ANX007 inhibits C1q and all damaging components of the classical pathway.

"Inhibiting C1q will block all downstream components of the classical cascade to protect photoreceptor synapses," said Lally. "The synapses appear to be more negatively impacted than the photoreceptor cell bodies."

The randomized phase II trial evaluated ANX007 in patients with GA, including foveal and nonfoveal lesions. The primary endpoint was change in GA lesion area as determined by fundus autofluorescence at 12 months. Key secondary endpoints included best-corrected visual acuity (BCVA) and low luminance visual acuity and deficit.

The study involved 270 patients randomized to sham injection or to monthly or every-other-month injections of ANX007. Baseline characteristics included a mean BCVA of 58 letters, foveal lesions in about half of the patients, mean GA lesion size of 7-7.5 mm², fellow-eye CNV in about 20%, and multifocality in about 70%.

Overall, the change in GA area size did not differ in the monthly (2.02 mm², -6.2%) or bimonthly (2.12 mm², -1.3%) ANX007 groups versus the sham control (2.15 mm²). An analysis of lesion growth over time showed that monthly and bimonthly ANX007 groups had a 1% decrease and 4.1% increase, in lesion growth, respectively, during the first 6 months of follow-up. Thereafter, monthly treatment was associated with a 10.5% decrease in lesion growth and bimonthly treatment with a 5.6% decrease.

Both ANX007 groups had fewer patients with ≥15-letter loss in BCVA versus the control group resulting in a pooled difference of 8.3% versus 21.3% in the control group (P=0.0024). Vision loss was reduced by 72% with monthly ANX007 (P=0.006) and by 48% in patients who received bimonthly injections (P=0.064). The finding is the first reported demonstration of significant vision preservation in GA, said Lally.

The BCVA benefit in favor of ANX007 was consistent across the 12 months of follow-up. A subgroup analysis also showed a consistent benefit.

The rate of ≥15-letter vision loss accelerated after discontinuation of treatment with ANX007, said Lally.

ANX007 was generally well tolerated, he continued. CNV occurred in four patients in each ANX007 group as compared with three in the control group. A total of three patients treated with ANX007 developed endophthalmitis, and one patient had retinal vascular occlusion. Three patients had intraocular inflammation with active treatment, but no cases of retinal vasculitis or ischemic optic neuropathy were reported.

Comment