DUBLIN – Genentech's decision to move crenezumab into a second phase III trial in Alzheimer's disease represents yet another test of one of the drug industry's most heavily tested hypotheses.
At times, it may seem as if the biopharma industry is bashing its head against a brick wall in its efforts to treat Alzheimer's by tackling amyloid beta (Abeta) deposition, long considered the key pathological hallmark of the condition. After almost two decades of clinical development, clearcut clinical success remains elusive, but Andrea Pfeifer, CEO of AC Immune SA, the Lausanne, Switzerland-based company that discovered crenezumab, takes heart from the faint efficacy signals that have been detected in a small handful of trials to date. Moreover, most of the other trials involving antibodies that target Abeta have been found wanting – either because they were inadequately dosed, because they recruited patients whose disease was too far advanced or because of safety issues.
Last November, Indianapolis-based Eli Lilly and Co. finally called it a day with solanezumab, after an extensive phase III program that had yielded some modest signs of efficacy, but nothing on a sufficient scale to warrant further investigation – or investment. "It had a harder mechanism of action to succeed – if the dose had been higher it would have had a chance," Pfeifer told BioWorld Today.
Solanezumab, which selectively binds monomeric forms of Abeta, is postulated to act as a peripheral sink, by mobilizing Abeta from the brain to periphery. Its dose may not have been sufficient for that task, however. The drug was administered intravenously in a 400-mg dose every four weeks – equivalent to about 5.7 mg/kg of bodyweight, Pfeifer said. That may not have been enough to deliver an effect, particularly as the antibody did not bind directly to the neurotoxic, oligomeric forms of Abeta.
The progress of aducanumab, which Cambridge, Mass.-based Biogen Inc. in-licensed from another Swiss firm, Zurich-based Neurimmune Holding AG, has engendered the highest levels of optimism in recent years. Phase Ib data, published in the Sept. 1, 2016, issue of Nature, suggested it had dose-dependent effects on Abeta clearance and on slowing cognitive decline. The data are necessarily tentative, however, given the under-powered nature of the study.
From a safety standpoint, the highest dose, of 10 mg/kg, gave rise to transient amyloid-related imaging abnormalities (ARIA) in 55 percent of patients, which are indicative of cerebral edema; that arm also had a dropout rate of 38 percent vs. 25 percent for the placebo arm. "There is a limited therapeutic window," Pfeifer said. Biogen has attempted to reduce the ARIA problem by gradually titrating the dose upward to 10 mg/kg, with some degree of success. "I think it's a challenge to deal with the side effects, which, of course, we don't have," she said.
Pfeifer's optimism around crenezumab stems from its safety profile – it can be safely dosed up to 60 mg/kg. That's four times the dose that was employed in two phase II trials, which failed to demonstrate statistically significant efficacy in the overall study population, but which showed a clear separation between drug and placebo in a subgroup of patients with mild disease. (See BioWorld Today, July 17, 2014.)
Jefferies analyst Peter Welford supports AC Immune's rationale. "Unlike other prominent Abeta antibodies such as aducanumab and 'sola' that have an IgG1 backbone, crenezumab contains an IgG4 backbone," he wrote in a research note this week. "Importantly, this IgG4 backbone results in clearance of Abeta without inducing an inflammatory response, enabling use of higher doses and a better side-effect profile. Higher doses should result in increased brain exposure and therefore potentially greater efficacy."
"This is not an assumption or a hypothesis. We have shown it," Pfeifer said. The ongoing phase III Cread trial is employing that high dose. So too will the second phase III trial, which like the first, will also recruit 750 patients with early stage disease.
Before either reads out, another glimpse of crenezumab's potential could come later this year, when Genentech, a subsidiary of Basel, Switzerland-based Roche Holding AG, is expected to report eagerly awaited interim data from a phase II prevention study in a familial cohort in Columbia with a genetic susceptibility to early onset Alzheimer's. The dose on that trial has not been disclosed. The imaging data will also be important – identifying the transition from a healthy to a diseased state is a complex undertaking. Often, patients may present with symptoms one or even two decades after beta-amyloid deposition starts. Studying the concentrations and patterns of misfolded proteins in the Columbian cohort could help to shed light on that, Pfeifer said.
Combo approaches up next
The best hope for the first wave of monotherapies is to slow the rate of cognitive decline. "To stop progression, we'll need combination therapies," Pfeifer said. The first such trials may not be too far off. "I think the FDA is quite open to combination therapies," she said. The individual components of any combination will need to demonstrate efficacy – it is unclear at this point, she noted, whether phase II or phase III data will be required.
Other mechanisms that could be paired with an Abeta inhibitor include tau inhibitors and alpha-synuclein inhibitors.
"The theory right now, which is quite well supported, is that Alzheimer's is an Abeta-driven tauopathy," she said. "Tau is a general indicator of brain injury." How early it appears in the disease process is unclear, although it is downstream from Abeta deposition. More sensitive and selective diagnostic tools are aiding longitudinal studies that help to answer that question. "I think that will lead then to the treatment scheme which we are all looking for," Pfeifer said.
AC Immune has several other therapeutic modalities in development, including an Abeta vaccine, ACI-24, in phase II and an anti-tau-antibody in phase I, both of which are also partnered with Genentech. A tau-targeting vaccine, ACI-35, which is partnered with Johnson & Johnson, of New Brunswick, N.J., is also in phase I. (See Table 1: Selected Alzheimer's Drug Development Programs, below.)