Article In Brief
An advisory panel of the US Food and Drug Administration voted to reject approval of aducanumab, an injectable, monoclonal antibody that targets amyloid-beta, in Alzheimer's disease. Experts said the drug failed to demonstrate efficacy and more research is needed.
After a long day of presentations on November 6, the vast majority of experts on an advisory panel of the US Food and Drug Administration (FDA) voted that study data were not consistent enough for the FDA to approve aducanumab, an amyloid-clearing monoclonal antibody, for the treatment of Alzheimer's disease (AD).
If the FDA chooses to approve the drug, despite the panel's recommendation, it would be the first amyloid-reducing therapy for AD. The medication, developed and tested by Biogen, targets one of the hallmark pathological features of the disease—amyloid-beta plaques.
“Everyone wants a treatment that works for their patients,” said Michael Greicius, MD, associate professor of neurology at Stanford University, who co-authored a perspective on the findings in the November issue of Alzheimer's & Dementia. “But it doesn't mean you skimp on the evidence. This is pushing us in the wrong direction. If it is approved, it will slow down finding something that does work.”
Dr. Greicius and his colleagues call for more research through a third phase 3 study. The two phase 3 studies on aducanumab were stopped in March 2019 when an interim futility analysis of the data showed that the studies had not met their endpoints. After further analysis, Biogen, makers of the drug, said that one of the studies (302) was positive, and the other (301) was not. The studies were identical in design, which triggered concern among the FDA's advisory team members.
Biogen scientists described the results of the phase 3 studies and explained why they thought that two identical studies could have led to discordant results. Several factors came into play: a change in dosing during the study and a high number of rapid progressors in the 301 group that could have skewed the results.
From the FDA's point of view, the question was whether one phase 3 study, 302, could be used to push the company's application forward to its approval, as well as results from a phase 1b PRIME study that showed a strong signal that the drug was reaching its target and clearing out toxic amyloid beta. Usually the FDA requires positive results from two phase 3 clinical trials, but the documents sent out prior to the meeting suggested that the FDA had a positive attitude towards the drug, although an FDA statistician laid out substantial weaknesses in the data.
At the end of the day, questions were put to a vote. Does the positive study, “viewed independently and without regard for the negative study, provide strong evidence that supports the effectiveness of aducanumab for the treatment of Alzheimer's?”
Eight of the reviewers on the advisory panel said ‘no,’ that it does not. One said ‘yes,’ and two abstained. Aaron S. Kesselheim, MD, JD, MPH, associate professor of medicine at Harvard Medical School, an advisory team member, said: “There was a suggestion of an effect but enough red flags.”
Another advisory panel member, Joel S. Perlmutter, MD, FAAN, professor of neurology and radiology at Washington University School of Medicine, added that “302 was a positive study but I still have concerns whether it is strong evidence. There is a huge danger in approving something that is not effective.”
The next question was stated this way: “In light of the understanding provided by the exploratory analyses of Study 301 [the negative study] and Study 302 [the presumed positive study], along with the results of Study 103 [the preliminary study that even the investigators who published this result in the journal Nature said it did not provide evidence of efficacy] and evidence of a pharmacodynamic effect on Alzheimer's disease pathophysiology, is it reasonable to consider Study 302 as primary evidence of effectiveness of aducanumab for the treatment of Alzheimer's disease?”
Ten of the 11 panel members said ‘no,’ that the study could not be used to support that the drug was effective for the treatment of AD.
When asked whether there was strong evidence for a pathophysiological effect, five agreed that biomarkers showed lowered amyloid-beta, and the others said that they were just not sure what that meant for the progression of the disease.
The advisory panel members agreed that the drug does find its target and clears amyloid-beta from the brain but many said that they were not convinced that the benefit observed on the clinical rating scale was necessarily linked back to the reduction of amyloid beta.
“There is some question of whether the imaging biomarker reflects or predicts clinical benefit,” said Dr. Perlmutter. In response to this question all but one said “no” and one voted “uncertain.”
The data “does not support a recommendation for approval of this drug,” Dr. Perlmutter said after the hearing. “The initial study as planned failed. Only by trying post-hoc subgroup analysis—done by breaking apart the two identical studies—could it potentially have demonstrated some benefit in one of the studies. This type of retrospective approach has lots of statistical issues and is commonly not validated when a follow-up prospective study is done. This reflects the bias of trying to do post-hoc analyses of these types of studies.”
“We all recognized the urgent, unmet need for treatment of a devastating disease,” Dr. Perlmutter said. “However, approving a treatment that ultimately does not work can be harmful to those people we are trying to help. If this drug were approved and it was not effective, it would still likely substantially slow recruitment into ongoing and future studies to identify a useful treatment. It could slow the enthusiasm and support for testing other potentially more effective treatments. Also we may be required to test any new treatment not against placebo but against this drug, which could confound subsequent studies. All of this could combine to slow our attempt to bring forward a drug that could really slow AD progression.”
There is a lot at stake, and there are scientists and clinicians standing resolute on both sides of the decision about this monoclonal antibody. Dozens of potential AD drugs, and many that target amyloid-beta, have been tested and failed, and this would be a major breakthrough for patients. On the other hand, some scientists say that if the drug is approved and doesn't deliver the clinical benefit that the field is hoping for it could set back the development of other Alzheimer's drugs. Also, clinical trials moving forward would have to test the drug against the standard of care, which in this case would be aducanumab.
If approved, the drug would be targeted at people with mild cognitive impairment and those with early AD. The therapy is administered intravenously once a month, and one of the troubling side effects observed in the study was brain inflammation seen on the MRI scans in a third of patients on the highest dose. The inflammation was reduced when lowering the dosage but the patients also had to undergo routine brain scans to manage this potentially serious side effect.
Paul Aisen, MD, the founding director of the USC Alzheimer's Therapeutic Research Institute Professor of Neurology at Keck School of Medicine of USC, said that he has studied the data and believes that aducanumab “appears to be an effective treatment.” At the same time, he added, “There are serious statistical questions. One study was positive and one was negative. I see both perspectives. There is a great need for a disease-modifying treatment and a delay in approval has an enormous cost. The path forward is difficult.”
Dr. Aisen did not serve on the advisory panel. “The usual standard would hold that all data needs to provide robust evidence of clinical effectiveness and have a favorable benefit to risk ratio. But I think all the questions can be explained by looking at what happens as the trial progresses.”
The study dosing was changed in the middle of the study and this could have affected the futility analysis, he added.
“Whatever happens, it is a step in the right direction in developing and testing drugs that target the underlying causes of Alzheimer's. Ultimately, it will take a combination of therapies,” said Ronald C. Petersen, MD, PhD, FAAN, director of the Mayo Clinic Alzheimer's Disease Research Center and the Mayo Clinic Study of Aging.
“I commend Biogen and FDA for doing all they can to consider the important issues at hand and considering what is best for patients and families,” added Eric Reiman, MD, executive director of Banner Alzheimer's Institute and a leader of the Alzheimer's Prevention Initiative.
“They have been considering a very promising drug in which the trials discontinued before they were completed, and when the findings may have been more definitive. One trial met its primary and secondary endpoints, and one study did not. At the same time, patients and families have an urgent unmet need, and they cannot afford to wait another four years before a new trial is completed.
“To me, aducanumab is reasonably likely but not certain to have a clinical benefit and has extremely promising biological effects. Patients and families that would value the chance to take a chance on the drug, but there is also a compelling need to confirm the drug's efficacy before or after it received accelerated approval. If the drug did receive accelerated approval, one would also a need to consider the impact on the further evaluation of this and other treatments that are being evaluated in clinical trials.”
In response to the FDA panel's vote, the Alzheimer's Association's chief executive officer, Harry Johns, published a statement saying, “I am disappointed to report that a Food and Drug Administration (FDA) Advisory Committee recommended today that a new drug not be approved by the FDA to treat Alzheimer's. This is not the end of the road. It is one step in the FDA review process. The Alzheimer's Association will continue to advocate for the FDA to approve aducanumab.”
The organization's chief strategy officer, Joanne Pike, PhD, provided testimony to the committee: “Though the treatment would potentially delay decline, not stop onset or progression, given the devastating toll of Alzheimer's, we believe the publicly-released data on aducanumab justify FDA approval, accompanied by a Phase 4 post marketing surveillance study.”
The federal agency could dismiss the concerns of the advisory team or recommend that Biogen conduct another phase 3 study.