Some antiamyloids remove target plaques from the brain, yet don’t seem to halt cognitive decline. Researchers are working to understand why.
On November 6, 2020, researchers who have been laboring to find a drug that will treat Alzheimer’s disease (AD) dialed in to a public meeting of the U.S. Food and Drug Administration’s (FDA) Peripheral and Central Nervous System Drugs Advisory Committee. The committee would review drug trials of Biogen’s aducanumab, and conclude with a vote on the drug’s safety and efficacy in treating AD. The independent advisors’ decision wouldn’t be the official one for aducanumab, but their vote usually mirrors the final FDA decision.
Aducanumab, a human monoclonal antibody, is a top contender for near-term approval to treat AD by clearing amyloid plaques from the brain. Amyloid is a protein that accumulates in the brains of AD patients as the disease progresses, and is believed to be the first stage in a process that leads to cognitive impairment. It is one of the two main targets of AD research today, along with tau tangles (tau is another protein, which in altered form creates tangles within nerve cells, damaging them). Aducanumab’s believers think it will be the first “breakthrough” drug that could change the course of the disease.
But the drug’s ascent has been turbulent. Back in 2015, aducanumab had gained interest when it looked as though it removed brain amyloid and slowed cognitive decline in a Phase 1 trial. Then, in March 2019, the drug’s developer, Biogen, put the brakes on, ending Phase 3 trials after reviewing interim patient data in what’s called a “futility analysis.” The analysis projected that the trials would miss their primary endpoints. It appeared that aducanumab had become the latest casualty in a line of failed antiamyloid candidates. Six months later, however, the drug developer announced that aducanumab did work. There were, Biogen said, promising findings from a larger dataset that included additional participants: those who had received a higher aducanumab dose in one study (study 302) had 22% less cognitive decline on a standard dementia assessment than those in a placebo group . Biogen said that based on feedback from the FDA they would be pursuing regulatory approval once again. Aducanumab had risen from the dead, observers wrote. A year later, at the FDA advisory meeting, its performance in these two large Phase 3 clinical trials was now under scrutiny.
Mixed signals from the data
Frustratingly, the trial data gave mixed signals as one study showed evidence of aducanumab’s effectiveness while the other did not. The research team explained that when mid-trial dosages were increased, more patients received them in one study than in the other. One study also contained more patients with rapid cognitive decline. An FDA director argued that the results of one trial should not “detract from the persuasiveness” of the other, and that there could still be a path to approval despite the different outcomes. But this suggestion to, in effect, “unsee” the one negative study didn’t sit well with the committee. Ultimately, “the data left the panelists and listeners puzzled about which patients might benefit from aducanumab,” Madolyn Bowman Rogers writes in a summary of the outcomes on Alzforum.org. “Not only did the EMERGE [trial] benefit seem to be confined to apolipoprotein E4 gene (APOE4) carriers [who are genetically at an increased risk for AD], but there were also unexplained interactions with age, sex, disease stage, and even nationality” .
Commenting on a forest plot of subgroup effects in the EMERGE data, Rogers notes, “By country, [aducanumab] may have worked in Spain but not Italy, in the Netherlands but not Germany, the U.S. but not Canada.” Another problem with the studies included a side-effect of the drug that potentially biased the results by making evident to those participants who experienced it that they were not on the placebo. The FDA’s statistical reviewer Tristan Massie concluded that “there is no compelling substantial evidence of … disease slowing” .
Despite a favorable case made by some internal FDA reviewers, the advisory committee voted “no” on aducanumab. The broader AD research community agreed: there needed to be another trial to establish the drug’s efficacy. And as much as AD patients, caregivers, and physicians desperately want a therapy to reverse the disease “yesterday,” many say they would rather wait for a drug that is proven, with no caveats, to work. Biogen says it will continue its effort to bring the drug to market, and the FDA has until March 2021 to decide about aducanumab.
Battered but still marching onward: The amyloid hypothesis
The stops and starts of aducanumab’s trials are not unusual in this field and the failure of other antiamyloids has seeded doubt in the amyloid hypothesis. Subsequent studies are trying to learn from other drugs’ failures, and adjust their design accordingly. One design factor under focus is determining when the drug will be given to participants and how far along they may be in experiencing AD, with the goal of intervening increasingly earlier, before symptoms develop. Another element AD drug trials grapple with is the subjectivity of some of the tests used to measure participants’ cognition and functioning, such as ratings on the clinical dementia rating-sum of boxes (CDR-SB) and activities of daily living (ADL) scales, which are based on information provided by the patient and their caregiver/s. Another factor is dosage. Higher doses of some antiamyloids come with potentially dangerous side-effects, such as increased risk of vasogenic edema, especially in APOE4 carriers.
Serious side-effects would seem to be a straight forward limit point at which to pause a drug treatment, but even in this regard, there are still unanswered questions. Consider the experience of Daniel Gibbs, a participant in Biogen’s aducanumab ENGAGE trial who developed a case of amyloid-related imaging abnormalities (ARIA) that required intensive care unit (ICU) hospitalization after his fourth dose of aducanumab. Gibbs is a retired neurologist and APOE4 carrier. Despite the severity of his side-effects, he remained open to the idea that the drug could help presymptomatic patients and that the ARIA might even have been instrumental in allowing the drug to impact the amyloid. He told Alzforum, “I’m not really worried about the side effects. I believe my case of ARIA is the most severe that has been reported, and I fully recovered and may even have benefited… I think that ARIA may actually be necessary to open the blood-brain barrier and allow the [drug] to engage with amyloid.”
After six months, Gibbs’s cognitive abilities returned to baseline and, he writes, “A subsequent amyloid positron emission tomography (PET) [scan] showed reduction of amyloid in the areas where the ARIA was worse, but a tau PET showed no improvement compared to a previous one three years before. In fact, the tau had progressed” . Gibbs’s experience demonstrates how mixed the outcomes can be, not just between individuals but for an individual. He is left, understandably, with mixed feelings about aducanumab. Of the Biogen trials, he says, “Some have argued that this is nothing but a sign that the amyloid hypothesis is incorrect. This may be true, but I believe that it is more likely that antiamyloid drugs will only be effective if used prior to the beginning of cognitive impairment, before neurons are lost.”
Many in the AD research community feel this way—that the antiamyloid hypothesis is still live—and that past failed therapies do not add up to abandoning this approach. Paul Aisen of the University of Southern California (USC) Alzheimer’s Therapeutic Research Institute writes what many echo elsewhere: “The failures have taught the field important lessons regarding target engagement, dosing, and study populations that greatly increase the likelihood of success.”
The DIAN-TU trials
Earlier in 2020, another high-profile antiamyloid study swung between promise and pitfall. Started back in 2012, the dominantly inherited Alzheimer’s network trials unit (DIAN-TU) set out to test whether either of two drugs could slow, prevent, or stop AD in people who are at greatest risk for the disease, family members who carry a mutation in a gene for autosomal dominant Alzheimer’s disease (ADAD), also referred to as early-onset familial Alzheimer’s disease. The study was the first global prevention trial for these at-risk families, and operated at 24 sites in seven countries and four languages. The DIAN-TU was a phase II/III randomized, double-blind, placebo-controlled study, testing Eli Lilly’s solanezumab, a monoclonal antibody targeting soluble amyloid beta, and Roche’s gantenerumab, an antibody binding fibrillar amyloid beta (Figure 1).
Often, trials test a single drug given to a patient population at a set dose. The DIAN-TU study was instead a trial platform, testing two drugs in parallel, at the same time, but in different people. Generally speaking, participants were younger than the typical Alzheimer’s patients and didn’t have other brain abnormalities.
For an average of five years, DIAN-TU followed participants, who received regular drug or placebo infusions or injections, blood tests, brain scans, and cognitive tests. “Every assessment, every procedure, every person—it’s all planned out ahead of time, and then you stick to the plan,” explains Randall Bateman, Charles F. and Joanne Knight Distinguished Professor of Neurology at Washington University School of Medicine and DIAN-TU director (Figure 2). “You keep going, even though you don’t know how it’s going, you just keep going. So there’s a lot of anticipation and excitement that builds up over those years of work.” Many eyes were on the release of the initial findings, due out in early 2020.
It’s built into the study design that no one looks at any of the data until the end. In February 2020, the research team gathered to get their first look at the outcome of years of testing. “We had no idea what the results were until the day the data was unlocked by the computer. We were all in a room together, and we all saw the data at the same time. That’s where the shock comes,” Bateman says.
In this case, the shock was one of disappointment. What Bateman and his team saw that day was that neither of the drugs had slowed cognitive decline in the volunteers; the primary outcome of the study had not been met. The study team was crushed, as were the participants. “We were so hopeful,” Marty Reiswig, one of the study participants told the New York Times. “The devastating part of it for me is that I was really hoping this would change the world.”
Weeks later, the DIAN-TU team had more news to share, after further analysis of the biomarker data. Gantenerumab, they saw, had removed amyloid from subjects’ brains. Measuring amyloid plaque both through cerebrospinal fluid and PET scans, Bateman says, they saw that subjects’ amyloid levels had improved “by about 40%” for those taking gantenerumab. “They didn’t get to normal levels, but it removed a good amount depending on the amount they started with,” says Bateman. “This is the first time in this population we’ve actually been able to reverse amyloid plaques.”
There were additional positive downstream effects of gantenerumab: a decrease in the detected soluble tau forms found in the cerebrospinal fluid—these “improved about 30% of the way toward normal,” Bateman reports (Figure 3). Another marker of neurodegeneration, neurofilament light chain (NfL), had also improved for those on the drug. “It is intriguing,” writes Heather Snyder, vice president of the Alzheimer’s Association, referring to the fact that an amyloid-targeting drug also affects tau and NfL .
However, the DIAN-TU results have left some scratching their heads. What are we to make of a drug that, as Philip Scheltens of Vrije University Amsterdam puts it, “reaches its target and has significant downstream effects, but no clinical effect, in an otherwise well-designed and well-executed trial?” .
One factor in why cleared amyloid may not translate into clinical efficacy may be time. The DIAN-TU trials highlight these questions: how early (in relation to disease onset) a patient is treated; how long it takes for a certain dose to either slow, pause, or reverse the plaques; and how long (if at all) after brain amyloid is removed could we expect to see a cognitive or functional benefit?
In the case of the DIAN-TU trials, a planned exploratory open label extension (OLE) to the study offers that additional time, Bateman explains (Figure 4). The OLE is open to all participants of the DIAN-TU trials, and intends to answer the question of what happens if gantenerumab can completely remove amyloid plaques to the point that people reach normal levels of amyloid again. “In the duration we had participants on high dose in this trial,” Bateman says of the trials that just concluded, “we did not get the majority back to normal yet. But in the OLE, we aim to get there.” Getting there, of course, can only happen with the DIAN participants, who continue to show up for this work (Figure 5). For some of them, staying on through what could be either the eventual disappointment or the ultimate vindication of gantenerumab, means not just added years of clinic trips, needles, infusions, and scans. It also necessitates being told (as part of the OLE eligibility screen) what they may have deliberately avoided knowing—that they do in fact carry the genetic mutation for developing Alzheimer’s. Unless—and this is the binding hope of all involved—the research finally proves that there is a drug that will protect them.
- K. Servick, “Biogen’s Alzheimer’s drug candidate takes a beating from FDA advisers,” Science, Nov. 2020. Accessed: Dec. 18, 2020. [Online]. Available: https://www.sciencemag.org/news/2020/11/biogen-s-alzheimer-s-drug-candidate-takes-beating-fda-advisers
- “Aducanumab still needs to prove itself, researchers say,” Alzforum, Nov. 2020. Accessed: Dec. 18, 2020. [Online]. Available: https://www.alzforum.org/news/research-news/aducanumab-still-needs-prove-itself-researchers-say#comment-38576
- D. Beasley and M. Maddipatla, “Biogen Alzheimer’s drug closer to approval with U.S. FDA staff backing, shares jump 40%,” Nov. 2020. Accessed: Dec. 18, 2020. [Online]. Available: https://www.reuters.com/article/biogen-aducanumab-fda/biogen-alzheimers-drug-closer-to-approval-with-u-s-fda-staff-backing-shares-jump-40-idUSKBN27K24L
- [Online]. Available: https://www.alzforum.org/papers/symptomatic-amyloid-related-imaging-abnormalities-apoe-e4e4-patient-treated-aducanumab
- “Confused about the DIAN-TU trial data? Experts discuss,” Alzforum, Apr. 2020. Accessed: Dec. 18, 2020. [Online]. Available: https://www.alzforum.org/news/conference-coverage/confused-about-dian-tu-trial-data-experts-discuss
- “In DIAN-TU, gantenerumab brings down tau. By a lot. Open extension planned,” Alzforum, Apr. 2020. Accessed: Dec. 18, 2020. [Online]. Available: https://www.alzforum.org/news/conference-coverage/dian-tu-gantenerumab-brings-down-tau-lot-open-extension-planned