Today, the Cambridge-based pharmaceutical company Biogen produced the highly-anticipated details of its alleged successful clinical trials for aducanumab, a drug to treat Alzheimer’s disease. If these two trials are considered successful by the US Food and Drug Administration when Biogen files for approval in early 2020, aducanumab could be the first new therapy for Alzheimer’s in nearly two decades. It would be the only one that actually treats the disease’s pathology, as opposed to its symptoms.
As tantalizing as the prospect of a success may be, there are reasons to be skeptical of these results, presented at the Clinical Trials on Alzheimer’s Disease meeting in San Diego. Biogen’s trials of aducanumab were originally cancelled in March after an early analysis failed to show significant effects. The work presented today was the result of a re-analysis of those same trials, with data that was collected while the original analysis was ongoing. Resurrecting a clinical trial is essentially unheard of; the fact that its success wasn’t obvious from the start has caused many experts to view the drug with raised eyebrows.
Even if approved, the compound won’t be a cure-all for everyone with Alzheimer’s disease. But one thing is certain: The Biogen saga could dramatically change the way scientists conduct clinical trials for these drugs. In about a third of all participants, the team tracked not only participants’ cognitive decline—the primary metric of success for the drug—but the biological changes in their brains. These analyses showed that the drug was working at the biological level it was supposed to, establishing a more reliable measure of their progress.
“It’s modernizing the way we do clinical trials,” says Howard Fillit, a neuroscientist at the Icahn School of Medicine in New York and founder of the non-profit Alzheimer’s Drug Discovery Foundation.
Tracking multiple measures
One of the biggest challenges in Alzheimer’s disease is catching it early, before a person’s cognition has deteriorated substantially. Alzheimer’s is characterized by globs of deformed proteins called amyloid and tau in the brain, and for decades, doctors could only diagnose the disease in an autopsy that identified these proteins. It was only around the time the last Alzheimer’s drug was approved, in 2003, that scientists started using spinal taps and PET scans to measure the toxic protein buildups in living people’s brains.
Inability to detect the disease early has made it one of the most expensive conditions to live with globally. But it’s also thwarted hundreds of clinical trials: Without being able to identify people in the early stages of the disease—those who could theoretically be helped by a drug—researchers couldn’t know if they were even testing the right groups. “Thirty-five years ago…I had no real idea who the patients were,” says Fillit, who has conducted several clinical trials for Alzheimer’s drugs.
Researchers relied on cognitive assessments to measure their mental faculties, but those tests are decades old—and they can’t distinguish between different forms of dementia. Though Alzheimer’s is the most common form of dementia, it’s one of several forms, each caused by different pathologies in the brain. Many people actually have multiple kinds of dementia—which means that a drug intended to treat only Alzheimer’s wouldn’t stand a chance at treating the rest of someone’s neurological issues.
But the Biogen trials were different: The two essentially identical studies, ENGAGE and EMERGE, recruited a majority of participants who either had mild cognitive impairment related to Alzheimer’s disease or mild Alzheimer’s disease. Their scores on several scales doctors use to measure the progression of dementia were recorded at the start of the trials. For about a third of all participants who received either placebo, low doses, and high doses of the drug, they took baseline and follow-up PET scans to measure amyloid in their brains. This way, they could track whether aducanumab, a synthetic antibody that binds to and clears buildups of amyloid, actually did its job.
Amyloid’s role in Alzheimer’s is still poorly understood; it’s not clear if it’s a result of the disease, or something that causes it. What scientists do understand is that amyloid can lead to Alzheimer’s disease, and when it does, it’s usually followed by buildups of tau proteins. There’s some debate about whether removing amyloid from the brain will actually improve the symptoms of Alzheimer’s; some think that by the time amyloid has carved out space in the brain, its damage to neurons has already been done. But it’s possible that removing amyloid could prevent the disease from progressing.
The results presented today suggest that aducanumab worked at the highest dose—infusions of 10 mg per kilogram of a person’s weight. Over time, participants who received this dose had the greatest reductions in cognitive decline: around fourth tenths of a point on a four-point scale. In the subset of participants who had amyloid measured, those who received the highest dose also had the least amount of amyloid present at the end of the trial.
Biogen has already stated that it’s been invited by the FDA to submit aducanumab for approval; it practically taunted regulators not to approve it, based on the trials’ positive results. But there are still reasons to be skeptical about its overall efficacy.
Most glaringly, only the EMERGE trial produced results that suggested a slowdown of participants’ cognitive decline. Perplexingly, in ENGAGE, participants who got the highest dose actually saw their cognitive decline speed up a little, and both the low and high dose of the drug seemed to reduce amyloid by the same amount. If aducanumab worked at the highest dose only, declines should have been much greater for those participants.
There also wasn’t much diversity among participants. Even though these trials included thousands of participants across hundreds of study sites across several counties, they were predominantly white, with Asian participants listed as the only minority. It’s not clear that aducanumab could work for everyone who develops Alzheimer’s.
Ideally, future trials will track more biological markers, in addition to cognitive decline, over time. Even if these trials fail, they can still give researchers clues about how the disease develops and progresses over time—which is still highly debated. “For Alzheimer’s disease, we’re gonna need not just one biomarker, but many,” says Fillit.
There are many factors that play a role in the disease, like genetics, inflammation in the brain, and the amount of tau a person has—each potentially its own trackable biomarker. Like with cancer and diabetes, it will require multiple drugs to tackle each of these aspects of the disease, with the potential to be used in unique combinations for individualized care.
Although it’s exciting that aducanumab may be the first drug that appears to target one aspect of Alzheimer’s disease directly, it’s highly unlikely that the drug will be a universal treatment for Alzheimer’s disease. But if drug companies continue to study new therapies using both cognitive assessments and biomarkers, they’ll improve their likelihood of success.