Top-line findings for Biogen Inc.'s anti-LINGO-1 monoclonal antibody candidate, BIIB033, that were released earlier this year gained a bit more credibility when additional data were disclosed today in advance of the company's presentations next week at the American Academy of Neurology's (AAN) 67th Annual Meeting in Washington.

The data, from five abstracts that Biogen researchers will present at AAN, showed additional evidence from studies in acute optic neuritis (AON) that the investigational therapy may repair myelin in humans. The AON data are designed to advance the scientific thesis on anti-LINGO-1, which the company also is testing in multiple sclerosis (MS).

The randomized, double-blind, placebo-controlled phase II RENEW trial evaluated anti-LINGO-1's ability to enable repair of an optic nerve lesion through axonal remyelination after the onset of a first episode of AON. The study enrolled 82 patients following their first incident of AON – a disease that typically affects one eye and is characterized by inflammation, damage to the nerve fibers and loss of myelin within the optic nerve – across 33 sites in Europe, Canada and Australia. Patients received six intravenous infusions of 100 mg/kg anti-LINGO-1 or placebo every four weeks.

AON was selected as an initial indication for the drug because an estimated 50 percent of those affected with the eye condition later develop MS.

RENEW examined the effects on remyelination by measuring the latency of nerve conduction between the retina and the visual cortex in the brain using full field visual evoked potential, or FF-VEP. In January, Biogen reported that patients treated with BIIB033 showed improvement in recovery of optic nerve latency, or the time for a signal to travel from the retina to the visual cortex, over 24 weeks, providing evidence of biological repair to the visual system. However, the results – a 34 percent improvement (p = 0.0504) in the recovery of optic nerve latency compared to placebo – were not statistically significant.

Top-line data showed no effect on change in thickness of the retinal layers – optic nerve neurons and axons – and visual function, as measured by spectral domain optical coherence tomography and low contrast letter acuity, respectively. (See BioWorld Today, Jan. 9, 2015.)

The full dataset, which captured results through final patient visits at 32 weeks, showed that people treated with BIIB033 who did not miss more than one dose continued to improve over time, compared to placebo. The six-month improvement of 34 percent – on average, an improvement of 7.55 milliseconds in recovery of optic nerve latency – progressed to 41 percent (p = 0.01) at eight months, or an average improvement of 9.13 milliseconds, according to Diego Cadavid, Biogen's senior medical director.

"This is an important finding, as this is a small study with limited power," Cadavid told BioWorld Today. "To see this type of change is very encouraging."

In addition, the percentage of subjects whose VEP latency in the affected eye recovered to normal or nearly normal (within 10 percent of the normal eye) following treatment with BIIB033 was more than double that of patients on placebo – 53 percent vs. 26 percent, respectively – at 32 weeks.

Although top-line data represented the average of the population, "we did this pre-specified, very interesting analysis of the percentage of patients who actually made a very good recovery on the primary endpoint," Cadavid explained. "That finding was also very encouraging."

A substudy using the exploratory multifocal visual evoked potentials, or mfVEP, showed similar treatment effects, according to Biogen, which plans to provide details at AAN. Nearly half (48 percent) of subjects in RENEW were enrolled in the substudy using mfVEP, in which visual stimuli are provided sequentially to distinct regions of the visual field to provide a wider and more precise analysis, conducted mostly by computer.

"Since no one had done a full field VEP in the context of a clinical trial before, we wanted to not only look at full field VEP but also to have another way to corroborate the findings that could actually be more robust and potentially more sensitive and reproducible," Cadavid explained. The mfVEP technique fit that description.

"We found it very encouraging that the results are actually very consistent with the full field VEP, both in the magnitude of the treatment effect and the timing," said Cadavid, who oversees the anti-LINGO-1 program at Cambridge, Mass.-based Biogen and is a fellow with the AAN. "The results give us further confidence that the changes that we are seeing are real and not some odd artifact."

Biogen said the anti-LINGO-1 treatment was generally well tolerated. The overall incidence and severity of adverse events (AEs) was comparable across treatment arms. The most common AEs occurring at higher rates in the anti-LINGO-1 arm than the placebo arm were fatigue, nausea and paresthesia.

Overall, the full RENEW dataset offered "very convincing evidence of biological repair of the brain – specifically, the optic nerve – in the group treated with anti-LINGO," he added. "We see RENEW as informing, with very sensitive and specific techniques, as to when demyelination – that axonal loss – happens, when remyelination occurs on its own and with anti-LINGO and how the axonal loss and the remyelination interact with each other. These are pieces that we didn't have, and RENEW is giving them to us. They will be very valuable to design the phase III program."

'LINGO EXPECTATIONS HAVE COME DOWN A LOT THIS YEAR'

Biogen isn't ready to advance the anti-LINGO-1 into that program quite yet, however. SYNERGY, the phase II study of BIIB033 in 419 patients with MS is ongoing, with data estimated to report in mid-2016, according to Cortellis Clinical Trials Intelligence. Biogen plans to wait for those findings before finalizing the design of a pivotal phase III program, according to Cadavid, who will present more details from the AON studies at AAN, including the impact of treatment on vision-related quality of life and baseline data from SYNERGY.

Shares of Biogen (NASDAQ:BIIB), which reported the BIIB033 data after the market's close, gained 21 cents Tuesday to close at $423.57 but were up nearly $10, or approximately 2.3 percent, in after-hours trading.

After mixed reviews in January, analysts have been waiting for more details on BIIB033 at AAN. Following its 2015 Global Healthcare Conference, Leerink Partners LLC analyst Joseph Schwartz reported that key MS opinion leaders were "hopeful but cautious about anti-LINGO's promise in MS," especially in light of a "limited efficacy signal" in the phase II AON study. "Specialists are not convinced that AON is the best model to evaluate BIIB033 since many of these patients recover well enough on standard of care, although the lack of concordance among functional and anatomical endpoints in RENEW was not particularly encouraging," he added.

RBC Capital Markets LLC analyst Michael Yee was more upbeat in a first glance Tuesday afternoon.

"As LINGO expectations have come down a lot this year and data [are] hard to interpret (Street cares on Alzheimer's more these days . . .) the company has suggested their confidence went up after the data because biology (remyelination) is working," Yee wrote. "Only question is will this actually improve clinical outcomes (disability, walk speed, etc.) in an MS patient."

Although the SYNERGY data are roughly a year away, "we think this is possible," despite low Street expectations, he added. The MS study is 18 months, compared to AON, which was only six months, giving the drug more time to work, Yee pointed out. In addition, "MS is a progressive, degenerative disease" in which neurons can still be protected, while in AON "the damage was long ago and too late to fix."

Release of AAN abstracts last month prompted Wells Fargo Securities analyst Brian Abrahams to posit another interesting theory. FP-1039, the fibroblast growth factor receptor 1 (FGFR1) inhibitor from Five Prime Therapeutics Inc., of South San Francisco, could offer a challenge to BIIB033, he suggested. Although the drug – partnered with London-based Glaxosmithkline plc – is being tested in cancer, the mechanism may also work as a remyelinating agent in central nervous system (CNS) diseases, potentially serving as an alternative way to inhibit LINGO-1.

"Previous data had suggested reducing expression of FGFR1 enabled maturity of oligodendrocytes, CNS cells that produce myelin, increased remyelination, and improved axonal integrity in mouse models," Abrahams wrote. "It is also known that LINGO-1 expressed in these cells modulates differentiation and myelination."

An emerging science abstract at AAN suggested that FGFR1 knockout mice with the mouse model of MS have reduced LINGO-1 expression, he explained, with the authors concluding that LINGO-1 may be the downstream mechanism by which FGFR1 functions in such cells. "As such, blocking FGFR1 (such as through FP-1039, an FGFR1 decoy) could enhance remyelination via LINGO-1," Abrahams wrote.

"Much would still need to be understood (including blood-brain barrier penetration) to know whether [Five Prime's] FP-1039 could have a potential future role as a remyelinating agent in broad-opportunity CNS diseases like MS," he quickly added. "Still, especially given how emerging science had guided FPA008's progression from an RA-focused agent to a promising immuno-oncology asset, we believe the science behind FGFR1's relationship with LINGO-1 will be interesting to watch, and could suggest a potential new role for FP-1039 down the line."