When Samantha Budd Haeberlein, Biogen’s head of clinical development, took the stage in San Diego, California, before a room full of Alzheimer’s disease researchers and physicians this morning, she knew she had some explaining to do. In October, the pharmaceutical company, based in Cambridge, Massachusetts, unexpectedly revived an experimental Alzheimer’s drug that it had declared a failure 7 months earlier. Ever since, scientists and industry analysts have been hungry for more detail about two large clinical trials meant to prove that Biogen’s drug, an antibody called aducanumab, slows down cognitive decline in the early stages of disease.
At the Clinical Trials on Alzheimer’s Disease congress today, Budd Haeberlein tried to clarify what has emboldened the company to apply to the U.S. Food and Drug Administration (FDA) for market approval for aducanumab early next year. After analyzing more patient data than were available at the time of a discouraging preliminary analysis, she explained, the company found evidence that the higher of two tested doses led to 22% less cognitive decline after 78 weeks than a placebo in one trial. However, the other trial failed to show any benefit, leaving some researchers with a grim outlook on the drug.
“I surely don’t think that it should be given market approval on the basis of these data,” says Robert Howard, a psychiatrist at University College London who has run clinical trials of potential Alzheimer’s treatments. More positive results from a subset of patients that weren’t preselected at the trial’s launch are not convincing, he says. “[Biogen has] broken all the rules, really, about how you analyze data and report it.”
Some still welcome the glimmers of clinical benefits Biogen revealed. In a panel discussion at the meeting, Sharon Cohen, a behavioral neurologist at the Toronto Memory Program in Canada and a principal investigator in the Biogen-funded trial, pointed to ratings on a scale of daily activities that suggest aducanumab helped trial participants retain some independence. “Those of us who know this disease well know what it means to lose yourself, slice by slice, and anything you can hang on to and do well is a triumph.”
The stakes are high for aducanumab because it’s among the last potential drugs standing that targets beta amyloid, the protein that forms sticky plaques around neurons in the brains of people with Alzheimer’s. Genetic evidence had built a compelling case that amyloid deposition drives the neurodegenerative disorder. But the failure of several antiamyloid drugs in large clinical trials have suggested plaque buildup might be the wrong target to stop the disease’s progression once people show symptoms.
Still Biogen and its partner, Tokyo-based Eisai, had high hopes for aducanumab, which binds to and eliminates forms of amyloid thought to be particularly noxious. An early clinical trial confirmed the drug was a powerful plaque buster. In 2015, the companies launched two large, identically designed clinical trials, known as ENGAGE and EMERGE, each testing two different doses of aducanumab against a placebo in more than 1600 people with early-stage Alzheimer’s and confirmed amyloid buildup in the brain.
In March, Biogen and Eisai announced they were halting both trials after a planned “futility analysis” in a subset of participants suggested the drug wasn’t helping stave off dementia. Then came the October surprise, an announcement that the drug showed signs of benefit after all and that Biogen planned to seek FDA approval. In both trials, aducanumab had significantly reduced amyloid buildup, Biogen reported, and in the EMERGE study, the high-dose group showed significantly less cognitive decline than the placebo group after 78 weeks, based on a standard scale called the Clinical Dementia Rating-Sum of Boxes (CDR-SB). But the results from ENGAGE were disheartening. In that trial, the high-dose group actually had a slightly worse decline than the placebo group.
What little the company shared in October left many puzzled about Biogen’s sudden reversal. “A drug company doesn’t stop a trial lightly … so the data that they had must have been compelling to stop,” Howard says. After the futility analysis, patients who had enrolled later in the study provided crucial follow-up data that changed the drug’s outlook, the company explained. But, “It’s really hard to get your head around how a small number of additional people recruited to the study could have made such a difference,” Howard says.
In today’s presentation, Budd Haeberlein elaborated on Biogen’s explanation for the two trials’ conflicting results. A major factor, she said, was how the trials treated participants who had a genetic variant called APOE4 that put them at increased risk of brain swelling as a side effect of the drug. (Overall, about one-third of patients getting a high dose of aducanumab had this swelling, which sometimes caused symptoms such as headache, dizziness, and nausea, but typically resolved within 4 months, Budd Haeberlein reported.) The researchers started the high-risk APOE4 patients on a lower dose as a precaution, but in 2017, decided they could safely ramp up to the trial’s highest dose. ENGAGE started about 1 month before EMERGE, and had more participants enrolled at the time of the change. As a result, ENGAGE ended up with a smaller proportion of its participants getting the high dose for the maximum of 14 doses, which might account for the lack of benefit. When Biogen analyzed the subset of participants who got this high, consistent aducanumab dose after the change to the protocol—nearly 300 people for each trial—they saw about a 30% reduction in CDR-SB decline compared with placebo in both ENGAGE and EMERGE.
The need for high dose of the drug over a long period is “a reasonable explanation,” for the conflicting results, says Zaven Khachaturian, editor-in-chief ofAlzheimer’s & Dementia, the journal of the Alzheimer’s Association, who is based in Washington, D.C. “I’m not sure whether it’s a complete explanation,” he adds. “The final answer will come [when] studies with higher doses for longer periods replicate these findings.”
But Khachaturian wouldn’t insist on waiting for that replication to put aducanumab on the market. He sees an argument for conditionally approving the drug based on these data but requiring the company to replicate the finding before that approval is finalized. Aducanumab “may have all kinds of blemishes,” he says, but, “I would say, give it a chance with the hope that others will come in and improve on this compound, this strategy. Because now we have a way of affecting the course of the disease.”
Howard is more inclined to see the positive EMERGE results as potentially spurious, given the failures of the antiamyloid trials before it. After publishing a skeptical take on the preliminary trial results inNature Reviews Neurologylast week, Howard heard from two families of participants in the Biogen trials, who were conflicted and anxious after being taken off the drug when the trial stopped only to learn it may have had benefits after all. (Biogen has said that these patients will have the option of restarting treatment.) Even if you did get the drug and not the placebo, the benefits were likely so tiny that they might not be evident for an individual patient, Howard told them in his reply. “Try not to be too full of regret or worry about it.”