To assess the efficacy of aducanumab, two large studies of the drug versus placebo were performed. The trials were stopped in March 2019 because the drug didn’t appear to be effective. But analysis of additional data showed that in the first, there was a small slowing of decline in a group of patients receiving a high dose of the treatment. Another group received a low dose of the antibody and continued to decline at a rate that was not statistically different from that in patients receiving a placebo. In the second study, patients receiving aducanumab, whether at a high or low dose, declined at the same rate as patients on placebo.
While there is precedent for the F.D.A. approving a treatment based on substantial evidence generated from a single trial, the limited evidence provided by the first of the two large aducanumab trials, the outright failure of the second and many other inconsistencies hardly meet this threshold. An F.D.A. advisory committee, on which one of us served, agreed and expressed with near unanimity serious concerns with the evidence to date.
Despite this, Biogen has done everything it can in news releases and investor reports, and at scientific conference presentations, to explain away the uncomfortable and disappointing fact that this product has not been proved to work.
Biogen says that the results from the failed trial would have been positive had investigators followed the patients longer and that analysis of a subset of patients with longer exposure to the high dose in the second trial supports the positive findings of the first trial.
But this post hoc justification simply cannot replace additional, well-designed, blinded, placebo-controlled randomized trials.
Given our lack of effective treatments, some may argue that aducanumab is better than nothing. We strongly disagree. In the aducanumab trials, three out of 10 patients exposed to a high dose had brain swelling as a complication, and although this was usually asymptomatic, in some patients it led to confusion, disorientation and falls. The swelling was detected with the use of rigorous safety screening, including routine M.R.I. scans. Such regular screening is unlikely to occur outside of the clinical trials, and because similar symptoms can be seen in progressive Alzheimer’s, distinguishing these adverse effects from disease progression would be especially difficult.
Approval of aducanumab will also, inevitably, slow progress in finding a new drug that is clearly safe and effective. A lot of continuing and forthcoming drug trials require regular infusions of the drugs being tested and safety M.R.I. scans. Conducting these trials will be more challenging in a setting where many patients are already on monthly infusions of an F.D.A.-approved drug that frequently causes brain swelling. Some patients and caregivers may be reluctant to enroll in a study if they are taking a newly approved drug they presume works. Others, while taking aducanumab, might be ineligible for new trials, since it would be hard to know whether adverse effects such as confusion or brain swelling were from aducanumab or any new investigational drug.
As millions of Americans know all too well, there is an urgent need to identify new treatments for Alzheimer’s. But there is no fundamental conflict between that challenge and maintaining the standards that have earned the F.D.A. the respect of regulatory agencies around the world. Our patients and their families deserve nothing less, and approving aducanumab without persuasive evidence that it actually works will only slow the discovery of what we need most — treatments that we can be confident actually work.
Dr. Michael Greicius is a professor of neurology at Stanford, where he directs the Stanford Center for Memory Disorders. He is also a co-founder of SBGneuro, a company that analyzes M.R.I. data in clinical trials. Dr. G. Caleb Alexander is an internist and professor of epidemiology at the Johns Hopkins Bloomberg School of Public Health. He served on an F.D.A. advisory committee evaluating aducanumab.
The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: email@example.com.