This summer, Adverum Biotechnologies Inc. drew Wall Street's eyeballs to a potential gene therapy for wet age-related macular degeneration (AMD) when the company made known that its IND had become active for ADVM-022 (AAV.7m8-aflibercept).

The Menlo Park, Calif.-based outfit plans a multicenter, open-label phase I study called Optic in the fourth quarter of this year with the candidate, which is the only intravitreal gene therapy prospect entering the clinic for wet AMD, a proven lucrative indication for existing players.

ADVM-022 makes use of the AAV variant 7m8, shown to mediate efficient gene delivery to retinal layers in mice and nonhuman primates, and – though it's yet to be borne out fully in the clinic – the intravitreal route could improve site-to-site comparability, Cowen analyst Phil Nadeau said in an August report. The preclinical findings show that ADVM-022 shots can achieve robust levels of aflibercept (better known to some as strong-selling Eylea, the inhibitor of VEGFA and PIGF from Tarrytown, N.Y.-based Regeneron Pharmaceuticals Inc.) for up to 52 weeks of follow-up.

AMD is hot. About 1.2 million people are afflicted in the U.S. and some 3 million worldwide. The other drug cleared by the FDA for the disease is the anti-VEGFA agent Lucentis (ranibizumab) from the Genentech arm of Roche Holding AG, of Basel, Switzerland. Roche's Avastin (bevacizumab), another VEGFA-blocker, is deployed off-label. "Even modest penetration [by Adverum's gene therapy] could result in meaningful revenue," noted Nadeau.

For now, onlookers can only guess at the possibilities, and they tend to lavish more attention on a program further along in Adverum's pipeline: ADVM-043 (AAVrh.10ha1AT), designed to use an AAVrh10 vector to deliver the A1AT gene to patients with A1AT deficiency. The compound is undergoing a phase I/II dose-escalation study that in late August had dosed two patients each in the first two ascending-dose cohorts: ADVM-043 intravenously at a concentration of 1E12 vg/kg (around 8E13 total vg for an 80-kg patient) and 5E12 vg/kg (about 4E14 total vg for an 80-kg patient). The data monitoring committee recommended upping to the next dose level, and the first patient in the third dosing cohort, testing 1.5E13 vg/kg (roughly 1.2E15 total vg for an 80-kg patient) of drug, was dosed in late July. Higher doses could be added. Results from six or so patients from the first three cohorts are expected by the end of the year, with primary endpoints having to do with safety and tolerability. Secondary measures will provide changes in total and M-specific A1AT levels in the plasma.

"We believe that results demonstrating acceptable safety and early signs of protein expression would be encouraging and a major catalyst for shares," Nadeau said.

About 100,000 patients in the U.S. with A1AT deficiency bear the zz and z null genotypes and increased risk of emphysema, with only about 8,000 getting the current standard of care: protein augmentation. It's tricky to dose correctly, however, and can only be given once emphysema has developed, so a potentially curative gene therapy could mean a significant edge for Adverum.

The company was born via the May 2016 merger of Avalanche Biotechnologies Inc., of Menlo Park, Calif., and Paris-based Annapurna Therapeutics SAS. Terms called for Avalanche to acquire Annapurna's outstanding shares in exchange for about 17.6 million newly issued Avalanche shares, giving existing Avalanche shareholders 62.5 percent ownership of the combined company, calculated on a treasury method basis, and Annapurna shareholders the rest. Analysts at first weren't much excited by the deal and Avalanche's shares took a hit, but things could be looking up. Along with ADVM-043 and ADVM-022, Adverum has ADVM-053, a gene therapy for hereditary angioedema, slated for an IND in the fourth quarter of this year. The firm has early stage efforts in collaboration with Regeneron and Editas Medicine Inc., of Cambridge, Mass. (See BioWorld Today, Feb. 2, 2016.)

How magic might the AMD gene therapy approach turn out to be? Pundits find clues in Rockville, Md.-based Regenxbio Inc.'s progress with subretinally delivered AMD gene therapy RGX-314, which encodes for the active ingredient in Lucentis. The efforts by Adverum and Regenxbio are similar but not identical, given ADVM-022's intravitreal route, the fact that the ADVM-022 program screens patients for pre-existing neutralizing antibodies, and its higher doses. A proprietary enhancer and promoter element is said to drive production of aflibercept to higher levels, too.

Kodiak bears down with longer-life candidate

Raymond James analyst Reni Benjamin pointed out in an early September report on Regenxbio that the phase I/II trial update in wet AMD showed that RGX-314 was well-tolerated. Cohorts 1-3 were measured, with cohort 3 demonstrating better results on all fronts, including dose-dependent protein expression, reductions in anti-VEGF injections, maintenance of central retinal thickness and of improved vision as measured by Best Corrected Visual Acuity (BCVA). "These interim results not only appear to suggest that RGX-314 is therapeutically effective, but also that the one-time injection could ameliorate the use of anti-VEGF injections in this patient population," in his view. Results from all completed cohorts are expected to roll out during the American Association of Ophthalmology meeting in mid-October in Chicago.

Just as investors ought not to confuse Adverum's program with that of Regenxbio, they should differentiate ADVM-022 from the program earlier advanced by Avalanche. The latter's wet AMD candidate, AVA-101, yielded only marginal BCVA improvements (an added 2.2 letters from baseline). AVA-101-treated patients exhibited retinal thickening while control patients decreased in thickness. Control arm patients only required two additional injections of Lucentis as compared to AVA-101-treated patients, which wasn't viewed as clinically meaningful. AVA-101 used an AAV2 vector; Adverum's management said the AAV2.7m8 capsid variant yields superior transduction efficiency as compared to AAV2, which may be not optimal for AVA-101's subretinal injection. Adverum's promoter/enhancer stands out here as well, since it's more potent at driving aflibercept production as compared to what Raymond James' Benjamin called "the ubiquitous chicken [beta]-actin promotor utilized in AVA-101." Here again, the higher dose with ADVM-022 was cited.

Benjamin maintained an "outperform" rating on Adverum's shares.

More players are piling into the wet AMD market. During the first week of September, Palo Alto, Calif.-based Kodiak Sciences Inc. filed for an IPO, saying the number of shares to be offered and the price range had not yet been determined, though the targeted haul is $100 million. The company has phase I-stage KSI-301, described as an anti-VEGF biologic with an extended ocular half-life that could let patients stay on the mechanism for a longer interval between intravitreal injections. KSI-301 is also in the works for diabetic retinopathy. Morgan Stanley and BofA Merrill Lynch are acting as joint book-running managers for the proposed IPO. Barclays is also serving in that capacity, and Chardan has been chosen as lead manager.