A recent study in The New England Journal of Medicine presents data, for the first time, showing a treatment that slows down the progression of Alzheimer’s dementia. The study – “Lecanemab in Early Alzheimer’s Disease” – tests a monoclonal antibody that is given as an IV infusion every two weeks. While this should be viewed as a legitimate breakthrough, it also shows the continued challenge of understanding and treating this complex disease.
Alzheimer’s (AD) is the most common form of dementia, which is a chronic global cognitive impairment. Usually memory is affected early and most significantly, but all cognitive functions can be affected. AD is considered a neurodegenerative disease, in which cells of the nervous system progressively die of. This entire category of diseases has proven extremely challenging. One of the challenges is that research discovers that lots of things are happening when nervous system cells die, but it is hard to know which things are driving the progression of cell death and which are themselves just downstream effects or markers. Is that inflammation killing the cells or just cleaning up already damaged cells?
Another challenge is that many of the neurodegenerative diseases are likely multiple diseases. They are clinically defined by which cell populations are being affected, but there may be multiple processes that are killing those cells, and each one would require a different treatment. Even a single neurodegenerative disease may have multiple causes, with the cumulative effects leading to cell death.
For AD there are two main contenders for the ultimate cause of degeneration, both abnormal proteins. One protein is called amyloid-beta (Aβ), which is a misfolded protein that aggregates outside of neurons, creating what are called plaques. The other is tau, which is also a misfolded protein, but this one aggregates inside neurons and causes neurofibrillary tangles. Tau is normally a protein that is part of microtubules, which transport nutrients throughout the cell. Further, Aβ and tau seem to work together, described as the “trigger and bullet” that causes neuronal degeneration.
Further complicating attempts at treating AD is the fact that Aβ and tau build up may occur for years and even decades prior to the onset of clinical symptoms – so the disease is fairly advanced pathophysiologically before we know it exists. Treatments aimed at Aβ and tau may therefore be too late to stop significant damage. There are screening tests for these proteins, in the blood, CSF, or through PET scanning, but these suffer from relatively low sensitivity and specificity. This means that many people with AD test negative, while many people without AD test positive. For this reason their clinical use is limited (although they are very helpful in research), mainly to atypical cases, and only 1-2% of people with AD get such protein tests.
I am glossing over a lot of complexity, but that is the basic picture. It’s no wonder that the last few decades of clinical research into treating AD has essentially been a string of failures. We may need to give multiple treatments to the right patients early enough in the disease process to make a clinically significant difference. Misfolded proteins are also not the whole picture, as other processes, such as inflammation, are playing a role.
With that background, we can understand why a treatment aimed solely at Aβ in patients who already have clinical AD may have limited utility. But that is the new treatment presented in the NEJM – lecanemab. These antibodies bind to Aβ, directing immune cells to them and leading to the breakdown of the Aβ plaques. The study itself is very rigorous:
We conducted an 18-month, multicenter, double-blind, phase 3 trial involving persons 50 to 90 years of age with early Alzheimer’s disease (mild cognitive impairment or mild dementia due to Alzheimer’s disease) with evidence of amyloid on positron-emission tomography (PET) or by cerebrospinal fluid testing.
The primary outcome measure was change from baseline to 18 months in clinical dementia rating. Several other dementia scales were used as secondary outcomes, as well as the amount of amyloid seen on PET scanning in the brain. Here is the main result:
The mean CDR-SB score at baseline was approximately 3.2 in both groups. 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).
That is, by all accounts, a modest effect size, about a 25% difference in outcome over 18 months. I previously wrote about another drug, aduhelm, which is also a monoclonal antibody that targets amyloid plaques. The controversy with aduhelm was that it was approved by the FDA based on preliminary data of reduction in amyloid only, without proof of clinical benefit. That data for lecanemab is better, and the FDA approval process is underway. This study, however, does lend some support to the potential for aduhelm to also work, since it works by the same mechanism.
The “controversy” with lecanemab is primarily about the ultimate utility of this approach. Monoclonal antibodies are expensive products, and this one would have to be given every two weeks as an IV infusion, presumably over years. That could represent a huge health care burden, for a modest effect. I am not saying it’s not worth it, and I imagine anyone facing AD or a loved-one with AD would take the 25% slowing down of progression. But we do need to think of the potentially massive effects such a treatment could have on health care systems.
Further, lecanemab was not without side effects, which were significant in 26.4% of patients. There was a 17% incidence of brain bleeds, and a 13% incidence of brain swelling in the treatment group (with few in the placebo group). It’s possible these side effects may be more detrimental than the benefit of the drug.
What we need to know at this point is how well lecanemab will work in the clinical setting. Will every patient need to get a PET scan or antibody study to qualify for treatment? How many years, on average, will patients be getting a regular IV treatment with the drug? Will the side effects get worse over longer treatment periods? On the optimistic side, researchers report that the average patient with AD lives at home for 6 years before going into a nursing home. Over that 6 years treatment with lecanemab can extend their independence by 19 months.
Perhaps the most useful thing about this study is that it is a proof of concept – treating amyloid beta early in the disease can have clinical benefits. Lecanemab (or aduhelm) may not transform AD treatment, because of the modest effects and side effect profile. But we now know we can make a clinical difference with this treatment approach. Next generation treatments may be more effective and safer. Also, we may need to combine amyloid treatments with tau treatments. We may also need better screening tests to tell early on, even before symptoms begin, who can benefit from such treatments. Hopefully we are just seeing the beginning of effective treatments for AD.