The FDA has just given its first traditional approval for a disease-modifying drug to treat Alzheimer’s disease (AD). This is a significant milestone with serious implications for AD treatment. But as always some clinical context is helpful.
In January of 2023 the FDA gave accelerated approval to the drug, Leqembi (lecanemab-irmb). The accelerated approval pathway was created, as the FDA says:
“…to approve drugs for serious conditions where there is an unmet medical need, based on clinical data demonstrating the drug’s effect on a surrogate endpoint, that is reasonably likely to predict a clinical benefit to patients.”
This was a departure for the FDA, who previously required evidence of clinical benefit in excess of risk. The accelerated pathway has also been controversial. Proponents argue that it gives practitioners the ability to prescribe drugs that have a good chance of working years earlier than they otherwise would. For diseases that are progressive and irreversible, this time can be extremely valuable. It also allows academic clinicians to conduct research more easily, since they can simply prescribe the drug and get it paid for.
Further, many clinicians already do this with off-label use. If a drug is approved and on the market, and may be of benefit in a different disease, you can just prescribe it off-label. The belief that a drug may work for a new indication is often based on its mechanism of action. The accelerated pathway would essentially allow the exact same practice but with drugs not already approved for a separate indication.
Critics, however, argue that using biomarkers to predict clinical efficacy is tricky business, with a fairly low hit-rate. This has been particularly true in neurodegenerative disorders, which are complex diseases involving many biological markers. Research into such diseases, like AD, ALS, and Parkinson’s disease, have been characterized by a string of disappointments, as one biomarker after another does not translate into a disease-modifying treatment. These are the very diseases the FDA is focusing on with their accelerated approval pathway.
The challenge is that when cells are dying, a lot of things are happening biologically. It seems that no matter what aspect of cell function we look at, it’s different in disease cells compared to healthy cells. Almost anything, therefore, can be a biomarker of the disease state. But that does not mean that every biomarker is actually driving the disease. It could just be a consequence of the disease. Treating a biomarker, therefore, may be like getting cosmetic surgery to look younger. It may alter some of the effects of aging, but It doesn’t change one’s biological age.
This is why the Leqembi approval is so important. It is the first time researchers have definitively tied a biomarker of AD with a demonstrable clinical outcome. Perhaps even more important than the availability of a new treatment for AD is the clinical confirmation that the Aβ proteins and the soluble protofibrils they create are pathogenic in AD, and not just a downstream marker.
Two years ago I wrote about another AD drug, similar to Leqembi, approved through the accelerated approval pathway – Aduhelm. There were other issues with this drug, namely that the data regarding beta amyloid reduction was not that clear, with critics arguing the data was not even statistically significant. The company was given three years to provide clinical confirmation of benefit, and those studies are still under way. Meanwhile Leqembi beat Aduhelm to the punch.
What is the data that Leqembi actually works? The study involved 1795 participant with early dementia and confirmation of amyloid plaques on either PET scanning or by cerebrospinal fluid analysis. Leqembi is a humanized monoclonal antibody that binds to beta amyloid and reduces plaque formation in the brain. The multicenter study was a double-blind placebo controlled trial with a 1:1 ratio of treatment to placebo. Subjects were followed for 18 months with a standardized measure of dementia, and several secondary measures of function. The study showed:
The adjusted least-squares mean change from baseline at 18 months was 1.21 with lecanemab and 1.66 with placebo (difference, −0.45; 95% confidence interval [CI], −0.67 to −0.23; P<0.001). In a substudy involving 698 participants, there were greater reductions in brain amyloid burden with lecanemab than with placebo (difference, −59.1 centiloids; 95% CI, −62.6 to −55.6).
Leqembi reduced brain amyloid burden and moderately reduce progression of dementia compared to placebo. All secondary measures were also improved in the treatment group. This is well above the threshold for full regular approval by the FDA, so it’s not surprising that an FDA panel approved the drug unanimously. These fairly clear clinical results are and important confirmation of the amyloid hypothesis of AD, and also is likely to be taken as validation of the FDA accelerated approval pathway. But what will it mean for AD patients?
Given that AD is a progressive, irreversible, and otherwise untreatable disease, any treatment that slows the progression of the disease is potentially hugely useful. On average people live about 10-11 years after diagnosis with AD (depending on severity at time of diagnosis), but there is a long tail with some people living up to 20 years. Even if we take an average survival, a 20% or so reduction means 2-3 years likely greater survival, and some of this time can have meaningful function. If treatment is started early in the disease, this could mean several years more of independent living and meaningful interactions with loved-ones.
And of course the hope is that the longer survival will give more time for research to develop further treatments which also slow down the disease. We may ultimately develop a cocktail of several treatments addressing different aspects of the disease. The ultimate goal is that the disease is slowed so much that people generally die with AD rather than from AD.
But there are significant downsides to this treatment as well. Because it is a monoclonal antibody it has to be given in an IV infusion, which is given every other week. The total cost is estimated at $26,500 per year, which would mean a typical Medicare copay of $5,000 per year. And there are side effects:
Lecanemab resulted in infusion-related reactions in 26.4% of the participants and amyloid-related imaging abnormalities with edema or effusions in 12.6%.
The amyloid-related imaging abnormalities (ARIA) result from swelling and bleeding caused by the inflammatory reaction of the antibodies attaching to the amyloid. Often this is minor and asymptomatic, but ARIA can be symptomatic in some patients causing headaches and dizziness. In severe cases there can be actual brain bleeds, even to the point of being life-threatening. For this reason patient on blood thinners should probably not take Leqembi.
There are about 6.7 million patients with AD in the US at this time. Let’s say about 1 million of them will be in the early stages and qualify for treatment. If all of them were treated, that would cost 26.5 billion dollars a year. Also, with the availability of treatment, the diagnosis of early AD is likely to increase. That cost does not include the treatment of side effects. In terms of cost effectiveness, that would have to be weighed against the cost of traditional AD care, including nursing home care. Would treatment only delay entry into nursing home care, or would it potentially reduce it (as some patients may die before getting to that point)?
The availability of expensive drugs providing moderate benefit for horrible diseases is something we need to confront as a society. It is technology like this that has been mostly driving up the costs of healthcare. The monoclonal antibody revolution has accelerated this effect.
From the perspective of the patient and their loved-ones, this is not much of a dilemma. Facing the reality of AD or a similar neurodegenerative disease, any effective treatment is welcome. Further I would think of treatments like Leqembi as a likely (hopefully) transitional phase. The most significant thing, arguably, about Leqembi is that we finally have clinical confirmation that our basic science understanding of AD is on to something. We are not just looking at shadows on the wall – we actually understand something about how this awful disease works.
The hope is that this is the first baby step on what will become a steady journey of improvement, even if incremental. We may be getting the first peak of a future in which neurodegenerative diseases are no longer destiny, but are simply managed chronically.