Failure of once-promising Alzheimer’s drug reinforces doubts about amyloid beta

For nearly a decade, aducanumab played a starring role in Biogen’s late-stage pipeline.

It had made its way into two Phase III trials thanks to Phase Ib data demonstrating its efficacy at reducing plaques of the peptide amyloid beta — a defining feature of Alzheimer’s disease — with observable clinical benefit to boot.

But last month, Biogen and its partner, Japan’s Eisai, shocked the world by halting the Phase III studies, ENGAGE and EMERGE, after a futility analysis by the program’s independent data monitoring committee showed the trials were likely to fail. The news sent the Cambridge, Massachusetts-based biotech’s stock plummeting. But perhaps more importantly, it reinforced growing skepticism about a hypothesis that has dominated Alzheimer’s research for decades: the idea that by reducing or eliminating amyloid beta, also known as Abeta, it’s possible to stop the disease progression that leads to the cognitive decline and dementia associated with Alzheimer’s.

A string of failures of Abeta-targeting drugs had already lowered investors’ enthusiasm for the target, and the discontinuation of the aducanumab studies led some analysts to expect that other anti-amyloid beta drugs in Biogen’s pipeline, like BAN2401, would also be shelved.

But the very next day, Eisai announced it had initiated Clarity AD, a Phase III study of BAN2401 in patients with early Alzheimer’s, for which Biogen is listed as a collaborator. The trial is expected to enroll 1,566 patients and so far as one site up and running, in the Miami suburb of Lake Worth, Florida, though more sites will inevitably come online soon. In other words, many physicians specializing in Alzheimer’s disease soldier on, hoping to find that one biomarker or mechanism of action that will finally make targeting of Abeta work, as some companies continue undaunted.

Eisai spokeswoman Libby Holman emphasized in an email that BAN2401 and aducanumab are different antibodies, with the former having higher selectivity and specific binding to protofibrils, which are toxic Abeta aggregates. Moreover, she wrote, the initiation of Clarity AD was based on an 856-patient Phase II study that showed clinically meaningful effects on cognition, function and biomarker correlation. Biogen declined to comment due to a quiet period ahead of a company earnings statement.

To be sure, amid a graveyard of amyloid beta drugs that includes Pfizer and Johnson & Johnson’s bapineuzumab and Eli Lilly & Co.’s solanezumab, analysts didn’t necessarily have high expectations for aducanumab to begin with. SVB Leerink, for example, had given the Phase III program a 35 percent probability of success “despite” concerns about Abeta as a target, according to a March 21 report.

“I just felt that the amyloid hypothesis had pretty much been tested by those drugs, and if there is a class effect, then I’m not surprised by what happened to Biogen,” said George Zavoico, an analyst with investment bank B. Riley FBR, in a phone interview. “Each of these antibodies differed in their mechanism, but it was too late in the disease progression to get the amyloid down and demonstrate a meaningful effect, at least the way those trials were designed.”

Even with the already low expectations, aducanumab’s failure was still a huge blow, particularly for the amyloid beta hypothesis as a whole. After all, in Phase Ib testing, the drug was shown definitively to reduce Abeta plaques and slow decline in clinical measures among patients with prodromal or mild disease. So if even that powerful a drug has ultimately turned out to be ineffective in more than 3,200 patients in two randomized Phase III trials, it seems unlikely to many observers that any other Abeta-targeting drug would work either.

“In the past we’ve seen other big amyloid beta-targeted drugs fail, but the excuse was that maybe they weren’t so potent at reducing amyloid beta in the brain,” Cowen analyst Phil Nadeau said in a phone interview. “Here, the Phase 1b data is clear that it was effective at reducing amyloid beta, so you have to question whether targeting amyloid beta can ever be effective.”

Not surprisingly, physicians felt the sting of losing such a key battle acutely.

“It’s devastating news – this is the one drug that looked very promising,” said Indiana University School of Medicine professor of Alzheimer’s disease research Dr. Liana Apostolova, in a phone interview. “It’s devastating to us as clinicians — and to patients — as we don’t have much to offer apart from nonspecific therapies that help a bit, but don’t slow disease progression.”

There have been other high-profile failures lately as well. In January, Roche’s Genentech subsidiary said it would discontinue its Phase III trials of the Abeta-targeting crenezumab in early Alzheimer’s. But it noted that it was continuing a program, the Alzheimer’s Prevention Initiative, to find if crenezumab could stave off Alzheimer’s in individuals genetically destined to develop the disease. The trial compares the drug against placebo in 252 participants who carry the PSEN1 E280A autosomal-dominant mutation, but are in the “preclinical” phase of Alzheimer’s, meaning mild cognitive impairment and dementia have not yet begun to manifest.

Yet another blow came last week with the publication in the New England Journal of Medicine of results from a Phase III trial of Merck & Co.’s verubecestat in prodromal Alzheimer’s. Like the aducanumab studies, the verubecestat trial was also halted for futility. Verubecestat, an orally administered small-molecule drug, is a BACE-1 inhibitor designed to reduce the production of Abeta and limit deposition of amyloid plaques.

But this time, the drug wasn’t just ineffective: Patients receiving it actually saw a worsening of symptoms compared with those in the placebo arm. In an editorial responding to the study, Mayo Clinic neurologist Dr. David Knopman wrote, “To be blunt, [amyloid beta] lowering seems to be an ineffective approach, and it is time to focus on other targets to move therapeutics for Alzheimer’s disease forward.”

The sentiment was echoed by a biotech CEO.

“What I see is, the industry needs to diversify away from amyloid,” said Adam Rosenberg, CEO of Rodin Therapeutics. The Cambridge, Massachusetts-based startup is developing approaches to neurological diseases like Alzheimer’s around enhancement of synaptic integrity. “Regardless of what your view is or was on the likelihood of success for amyloid, it’s clear that as an industry we’ve been overweight in our investment in this mechanism.”

But others aren’t ready to give up.

Dr. Lars Lannfelt, senior professor of geriatrics at Uppsala University in Sweden and leader of the lab that discovered BAN2401 in the late 1990s, pointed to the drug’s targeting of soluble protofibrils as a feature differentiating it from other anti-Abeta agents.

“I think amyloid beta is the right target, but you should target what is the toxic part of amyloid beta,” he said in a phone interview. He emphasized that Alzheimer’s represents a relatively young research field compared with, say, cholesterol and diabetes, with amyloid beta only detected in the 1980s and BAN2401 discovered in 1999. “With every failure we learn something important,” he said. “It’s a new field – people should not be pessimistic.”

Meanwhile, others are looking at genetics to provide clues to tackling the disease.

Apostolova pointed out that Indiana University’s specific interest is in tying imaging with genetics and using genomic sequencing to try and identify biomarkers. Since the first gene contributing to late-onset Alzheimer’s was discovered in the early 1990s, more than 30 others have been identified. At the same time, she said she supports the idea that tackling amyloid beta alone will likely lead to many additional failed drugs. That necessitates broader thinking and the acknowledgment that additional factors like inflammation and tau – a protein whose defectiveness is associated with neurodegenerative disorders like Alzheimer’s and Parkinson’s – as well as other processes should also be pursued. This could include a preventive, vaccine approach.

“I’m definitely in the camp that says we can’t just focus on amyloid anymore,” she said.

It’s one thing to try and prevent Alzheimer’s in patients who are genetically predisposed to it, as in Genentech’s aforementioned ongoing crenezumab study, but another to try and target amyloid beta among the broader population even in the prodromal phase, when mild cognitive impairment symptoms begin to appear. It would likely be too costly to treat the entire population, and in the prodromal phase it’s not even necessarily clear whether a person has Alzheimer’s given that there are no definitive biomarkers for the disease, Zavoico said. “I think the jury is still out on that,” he said, referring to the viability of targeting Abeta plaques in prodromal disease. “But my thoughts are that it wouldn’t work because that’s a phenomenon of the disease that’s likely the result of dysfunction in one or more upstream pathways, not the cause.”

Dr. Omid Omidvar, a neurologist at Southland Neurologic and MemorialCare Health System near Los Angeles, expressed a similar view. Alzheimer’s, in his opinion, is a protein-misfolding disease, similar to Parkinson’s, and the question is where that misfolding is occurring, but it’s likely something upstream of amyloid beta – potentially brought on by environmental factors or infectious agents – with Abeta as the end result. “I tell my residents, it’s trying to clean out the ashes in our yard when the fire is burning,” he said. “Try as much as you want, but as long as the fire is burning, you’re going to get more ashes.”

For now, there is potential for targeting of amyloid beta to be used as a way of slowing disease progression and delaying onset of early Alzheimer’s symptoms, but it does not seem to have the promise it once did.

“In terms of looking for an amyloid antibody that not just slows the decline, but stops progression, amyloid beta is dead,” Zavoico said.

Photo: Alaric DeArment, MedCity News