A way forward for Alzheimer’s disease drug development

Written by George Perry (University of Texas at San Antonio, TX, USA)

George Perry (University of Texas at San Antonio, TX, USA) is a Professor of Biology and Chemistry and is the Editor-in-Chief of the Journal of Alzheimer’s Disease. He has been working in the field of Alzheimer’s since the early 1980s.

In this interview, we had the pleasure of speaking with George to discuss his opinions on the recent trial results from ENGAGE, EMERGE and DIAN-TU. For example, what can we take from these findings? Are the underlying issues with anti-amyloid therapies related to dose? Could early in the disease course still not be early enough? Considering the trial results so far, what might be the next steps? Take a look at the full discussion below.


What are your thoughts around the recent trial results from ENGAGE, EMERGE and DIAN-TU?

I think all these trials are in a straight line of not supporting the amyloid cascade hypothesis as an explanation of Alzheimer’s disease. This is particularly true for DIAN-TU because that study focuses on exactly the genetically associated patients on which the amyloid cascade hypothesis is based. The amyloid cascade hypothesis derives from explaining Alzheimer’s disease in people who have amyloid pathway mutations, which the hypothesis defines as amyloid causation. The DIAN-TU trial failure leads all of us to re-think, what is genetic causality? It would be better to think of genetic causality – or genetic studies that link to certain diseases – as statistical associations and not the origin of the disease.

Clinical studies are the test of causality. For example, if you remove amyloid and it is the cause, then the patient should get better. Thus, absence of improvement in all clinical studies with amyloid-based therapy disproves that amyloid is the cause of the disease. It does not mean that amyloid is not important in the disease because that is what the genetics show. It just does not show what role amyloid has in Alzheimer’s disease.

Why would aducanumab ‘work’ when all other amyloid-targeting agents have failed?

“Continuing with the prior patients will again limit the statistical interpretation of the trial. Even if they find a similar slight slowing of disease progression, would aducanumab substantially benefit patients, families and society?”

Recent attention on the possible benefit of aducanumab (Biogen) must consider that the trial showed borderline statistical significance and limited slowing of patient progression. Further, the “positive” signal was after post-hoc analyses, which is suspect because you are selecting subpopulations to analyze. In a study that Eli Lilly (IN, USA) sponsored of an amyloid-based drug, where the subgroup analysis suggested that there might benefit, the signal disappeared when properly repeated.

My understanding is that aducanumab (Biogen) is approved by the US FDA for continuing the study with the same patient groups. Continuing with the prior patients will again limit the statistical interpretation of the trial. Even if they find a similar slight slowing of disease progression, would aducanumab substantially benefit patients, families and society?

The main argument given for amyloid trial failure is that treatment started too late, when the patients’ brain degeneration is irreversible and unable to recover from amyloid. A couple of the counter arguments include: first, Alzheimer’s disease heterogeneously damages brains with considerable overlap with ‘cognitively normal’ patients who in advanced age also show degeneration; second, everything that we know about the brain supports recovery from insult (e.g., stroke); finally, why has not one patient recovered function following amyloid removal?

Could the doses have been too low with other anti-amyloid therapies?

Every time an amyloid-based trial, fails there are excuses. For example, they were the wrong patients or there weren’t enough patients studied. The amyloid cascade promised to reverse the disease and put patients on the road to a cure. No amyloid trial demonstrated a reversal of disease or cure.

The gold standard for dose is, ‘Did it remove amyloid?’ In the Biogen studies, they showed compelling data of amyloid removal.

In your opinion, do you believe it’s time to move past developing anti-amyloid drugs for Alzheimer’s disease? Or are amyloid plaques too intimately involved in the disease to give up on this approach? 

We should give up on the amyloid cascade hypothesis. It is dead; the DIAN-TU study was the last hope of validation. Prior to the DIAN-TU study, I always opened the possibility that amyloid could initiate genetically caused Alzheimer’s disease. The DIAN-TU study now shows that the amyloid cascade does not explain Alzheimer’s disease even in hereditary cases.

We should intensify studies of amyloid because understanding what it does during the biology of the disease will shed light on the origin of the disease. Genetics surely shows amyloid is critical to development of Alzheimer’s disease. Our own studies support amyloid as an active response to the disease: a protective response much like inflammation.

A lot of focus has been driven toward the early treatment of Alzheimer’s. Given the recent results from the DIAN-TU trial, which focused on the early stages of autosomal dominant Alzheimer’s, could ‘early’ in the disease course still not be early enough?

If you go earlier than this, when are you going to treat people? In a debate at the American Academy of Neurology a few years ago, I said then you’re going to end up treating in utero.

If you start treating, for instance, 50-year-olds who don’t have mutations, then how many years do you have to wait until they have a clinical change? Decades. The study becomes not only expensive but also cumbersome to complete. It also delays the field. This is an excuse for failure.

Why are the topline results from the DIAN-TU trial so critical? 

“The DIAN-TU study is so critical because every patient has a mutation in amyloid processing consistent with a linear explanation of Alzheimer’s disease. Removing amyloid from these patients did not make any clinical difference.”

The DIAN-TU trial is so critical because the people developed the disease due to a mutation in amyloid processing. The way in which you would prove that amyloid is the driver would be animal studies. In rodents, amyloid deposition is not part of normal aging, but overexpression of mutant amyloid-related genes can lead to amyloid deposits and altered behavior. Removing amyloid doesn’t surprisingly restore the animals to normal, which is the basis for the amyloid trials.

In humans, amyloid develops during the aging process, probably related to metabolism, mitochondrial dynamics and iron/copper redox control, and plays a critical role in how we adjust to the aging process. Mutations may alter amyloid deployment to defend against age-related metabolic failure.

The DIAN-TU study is so critical because every patient has a mutation in amyloid processing consistent with a linear explanation of Alzheimer’s disease. Removing amyloid from these patients did not make any clinical difference. Clearly the link of amyloid accumulation to Alzheimer’s is not simple – it is part of the disease, but it is not causative to the disease.

Moving forward, what can we take from these findings? What might be the next steps in the search for a disease-modifying treatment for Alzheimer’s?

We should not put all our eggs into one basket with anything, whether that is amyloid or tau. I think it is important that we understand the biology of Alzheimer’s disease and other neurodegenerative diseases in a more rigorous fashion. Right now, much of our understanding is primitive.

A lot of work in Alzheimer’s disease was done with concepts from the time I entered the field (the 80s and early 90s), which is when genes and structures were analyzed. If you re-examine those analyses, you will see many are less than complete. Careful reassessment of our foundation is essential.

Is there anything else that you would like to add to the interview before we conclude?

I think the most important thing is that we are not having the same interview in 5–10 years’ time.


Disclaimer
The opinions expressed in this interview are those of the interviewee and do not necessarily reflect the views of Neuro Central or Future Science Group.