Editor’s Note (7/7/23): Scientific American published this interview with neurologist Samuel Gandy of the Icahn School of Medicine at Mount Sinai in January, when the Food and Drug Administration gave its accelerated approval for the new drug lecanemab. The article has been updated to include information about the FDA’s traditional approval of the drug, which was granted on Thursday.
No drug with consistent statistical evidence from clinical trials had ever been found to slow the course of Alzheimer’s disease before the Food and Drug Administration conferred its initial “accelerated approval” nod in January for lecanemab, which clears the brain of the toxic amyloid protein that has been a primary target for drug developers. After reviewing the results of an additional clinical trial, the FDA took the next and consequential step of granting traditional approval to lecanemab for early Alzheimer’s on July 6. In turn, Medicare will now cover much of the $26,500 cost of the drug.
The new approval also draws attention to an infrequent but severe side effect for some people: lecanemab will go to patients with a required black box warning about the possibility of “serious and life-threatening brain edema that can be associated with seizures and other severe neurological symptoms,” the FDA said in its announcement.
The drug, marketed as Leqembi by the companies Eisai and Biogen, is by no means a cure for Alzheimer’s. Its sometimes serious side effects, modest benefits and high price tag are enough to give some pause to patients and doctors alike. But the FDA’s traditional approval marks a milestone for a field that has witnessed numerous failures over the years.
Patients with early-stage Alzheimer’s or mild cognitive impairment will quickly find that lecanemab is not a neurological medication that sedates, stimulates or dampens pain. It will not make a person receiving an every-other-week infusion feel smarter. Their memory will not suddenly improve. The drug will just slightly slow the unremitting progression of Alzheimer’s.
Patient expectations will also need to be tempered in another way. Lecanemab works by removing a certain form of amyloid, but removing amyloid may not, by itself, suffice to hold Alzheimer’s at bay enough to maintain intact cognition. In addition to Iecanemab, patients may also need to take drugs that are still in development that tamp down inflammation or remove other aberrant proteins.
Samuel Gandy, a leading Alzheimer’s physician and researcher at the Icahn School of Medicine at Mount Sinai, has an understanding of lecanemab’s unique molecular targets: he works with his colleague Michelle Ehrlich, a physician and researcher at the Icahn School of Medicine, on making transgenic mice to learn more about Alzheimer’s-related proteins, basic research that may tell us something about how the drug works.
In January Scientific American spoke to Gandy about what a patient taking Iecanemab might expect in a doctor’s office and what future research is needed to make Alzheimer’s a manageable disease.
[An edited transcript of the interview follows.]
As both a researcher and a clinician, how would you describe the results of the lecanemab drug trial to one of your patients?
The trial shows that, over an 18-month period, patients receiving the drug declined at a slower rate than those receiving a placebo, which, in this case, was a dummy infusion rather than a sugar pill. The slowing of decline was statistically significant on five different prespecified end points (trial goals). This does not guarantee that the change is going to be noticeable to patients in their everyday lives. By way of managing expectations, I remind patients that the benefit is a slowing of decline and that patients and families should not expect to see improvement. I go on to say that, because of this, neither the patient nor the family is likely to notice any effect of the drug.
Were there any side effects?
There was some brain swelling in about 15 percent of patients in one of the clinical studies. And there are patients who have an Alzheimer's genetic risk factor called APOE4 that tend to be less responsive to the drug who are also predisposed to a greater chance of side effects.
The brain swelling can usually, but not always, be detected on brain scans before there are any symptoms. In some patients, this swelling can also include a small amount of bleeding that usually causes no problem. There is concern about three patients who were receiving lecanemab who had major brain bleeding and died. At least two of these patients were taking some form of blood thinner, and blood thinners can cause bleeding even when there is no lecanemab. When I evaluate patients for lecanemab, I mention these side effects, but I make it clear that we cannot be sure what role lecanemab played in the bleeding. This is ambiguous information, to be sure, but this is where the situation stands today.
Is the brain swelling and bleeding going to limit the patient population that will be able to receive the drug?
Yes, my estimate is that about 20 percent of the population of patients with MCI (mild cognitive impairment) may qualify for this drug. Patients who have a history of stroke or are on blood thinners will be warned that their chance of having side effects is greater than those who haven’t had a stroke and are not on blood thinners.
What does the drug do?
The unusual thing about this drug is that it targets not only the amyloid plaques that are a hallmark of Alzheimer’s but also clumps of amyloid, called oligomers, that float around inside and between brain cells, along with protofibrils that share properties of both oligomers and plaques. We believe that some oligomers may form in or around plaques. Because oligomers are invisible on amyloid brain scans, we cannot yet be sure when we have gotten rid of them. This is important because oligomers are even more poisonous to brain cells than plaques are.
For a long time, there has been report after report of Alzheimer's drug failures. What is the importance of this drug approval?
This is the first drug targeting amyloid that has a statistically significant benefit for patients. This provides strong evidence that amyloid is part of the story, but it is also clear that amyloid is not the whole story. There are other factors that contribute to decline in Alzheimer’s patients involving inflammation and tangles aggregates of another protein, called tau, that accumulate inside the cell. And many experts believe that a cocktail of drugs will be required in order to have a more robust effect. By “more robust,” I mean “halting decline altogether.” This is worth noting because we are now able to detect the presence of Alzheimer’s pathology before there are any symptoms at all. If we could halt progression of pathology in that group of individuals, we might prevent decline from ever starting. But we are not there yet.
Given what you said about the drug’s limited immediate impact and the side effects, do you think it will make much of a difference in the treatment of your patients?
There’s some evidence from the study results that, as time goes on, the difference between the progression of the disease in someone taking a placebo and someone taking lecanemab becomes larger. The benefits may become larger the longer a person takes the drug. This is not necessarily because the drug works better but may be because many untreated patients continue to decline so that the difference between drug and placebo is larger.
Increasing skepticism has arisen in recent years because of the many drug trial letdowns related to amyloid as a key contributor to Alzheimer’s. What then are the implications of this trial?
The lecanemab trial does implicate the amyloid molecule in the progression or initiation of the disease. I think that the controversy in the field is understandable because there are so many molecules and cells involved in Alzheimer’s. Even researchers who work only on amyloid recognize that amyloid is not the whole story. The size of the effect on cognitive decline produced by lecanemab is fairly small, but I think that combining antiamyloid therapies with other approaches may give more robust effects. I mentioned before that we might want to begin some interventions before there are any symptoms at all. If that approach proves effective, that would be very “robust” indeed.
Because of the recent failures of drugs targeting plaques, might lecanemab be a test of whether future drugs should target oligomers instead?
What this trial shows is that it may be best to target both oligomers and plaques. But it will also be worthwhile to continue trying to develop a blood test, spinal fluid test or brain scan that will measure oligomers, which are currently undetectable in living patients.
You mentioned before the need for a drug cocktail. What might go into it?
The first two ingredients, I would think, would be an antiamyloid drug like lecanemab and some anti-inflammatory molecule that targets microglia, the inflammatory cells of the brain. As amyloid begins to accumulate, that buildup triggers the activation of these microglia, the garbage cleanup cells in the brain. This is both good and bad. Microglia try to ingest amyloid and get rid of it, but once they are activated, dozens of microglial molecules cause inflammation that injures nerve cells and synapses and really compounds the problem. Many people know inflammation as the cause of the pain of arthritis, but inflammation contributes to the brain damage in Alzheimer’s as well.
Would tau and the tangles it creates be another target?
I think that once we see what the antiamyloid–anti-inflammatory combination looks like, we’ll know whether we need to target the tangles that also affect the functioning of neurons. The antitangle antibodies that have been tested so far have not looked very promising.
How long do you think it will take before patients will see these cocktails?
The very first clinical trials for the inflammation modulators in microglia are only just beginning. None of them have given any results yet, so it’s really impossible to predict. We don’t know which target in the inflammatory process we want to aim at. We don't know whether we want to turn it up or down or what phase of the disease we want to treat. There are around 100 genes that have been associated with Alzheimer’s, and many of them are present at their highest levels inside microglia.
Do you think lecanemab by itself can be further optimized?
There’s evidence that people who have no Alzheimer’s symptoms have detectable chemical changes in their spinal fluid or blood that predict that Alzheimer's pathology has begun. It might be possible to treat those people with drugs like lecanemab or others and have them never develop any symptoms at all. As I said before, this is what we are aiming for; this is not where we are now.
Do you think an effective drug for people in later stages of Alzheimer's would be possible?
The problem is that late-stage dementia patients have lost so many neurons and synapses, and we have no way of beginning to reconstruct them. There are medicines that stimulate the birth of new neurons. But the new neurons have to be born in the right place in the brain and talk to specific other neurons to be effective. There are experimental medicines, one of which our lab is working on, that stimulate neurogenesis in the hippocampus, the brain’s memory hub. We can prevent the memory decline of mice that have only oligomers, mice that have plaques or mice that have tangles. The complexity of the pathology makes the neurogenesis approach especially attractive. About half of patients with dementia have multiple types of pathologies, and it’s hard to imagine that targeting one protein will be ideal for them. That’s not to say that we shouldn’t try; it’s just that I am skeptical that targeting a single molecule will be sufficient.
What would you say is the overall significance of lecanemab for Alzheimer’s research, given the field’s labored progress?
I think it means that at least part of our understanding of the disease is correct. And now we need to focus on other targets and optimizing [the drug’s] benefits. The science behind the development of lecanemab suggests that we should target both plaques and oligomers. The situation in Alzheimer’s now is similar to where oncology was in the 1960s and 1970s, with drugs that produced small effects. More research made it possible to improve on their benefit through multiple generations of optimization. Some cancer drugs have now achieved major successes.