Lessons in Alzheimer Disease

Alireza Atri, MD, PhD: Regrettably, we’ve had a number of very high-profile experimental drug program failures in Alzheimer disease in the last year. Along with that, we’ve learned a fair bit. One of the things we’ve learned over the last several years is that it’s very important, if we are going to have an experimental drug that’s targeting amyloid or tau, that we make sure that the individual, even though they may have a clinical syndrome—let’s say it’s cognitive impairment or dementia—be clinically thought to have Alzheimer disease. Or make sure that it’s actually proven to be, in some ways, by biomarkers, due to Alzheimer disease.

Now we have ways, through spinal fluid, of looking at amyloid and tau profiles, or from PET [positron emission tomography] scans to make sure that individuals have a sufficient amount of amyloid in their brain. And now even experimentally, tau can be seen on PET scans. Again, that’s available experimentally. So that’s on the diagnostic side. That’s very important because you want to give a treatment that is working on a certain protein for an individual who actually has an abnormal amount of that. So that’s one lesson that’s been learned.

The other lesson that’s very important, and not all of the data are out on this, is there have been tremendous failures, unfortunately, of amyloid antibodies and even BACE [beta secretase] inhibitors in mild to moderate stages of Alzheimer disease/dementia, even reaching down to the MCI [mild cognitive impairment] stages of Alzheimer disease. Not all of the programs have failed. There are at least 2 drugs in phase 3 studies right now for antibodies that are still being tested. For BACE inhibitors, I think 5 out of 7 programs have either been halted because of a lack of efficacy or because, in some cases, there were some detrimental signals on cognition that were mild but very reproducible.

There are 2 drugs that are BACE inhibitors that may have a different profile. They may not have the same off-target effects. They may be affecting BACE2 instead of BACE1, and they may be modulating the production of amyloid in a certain way that it’s not too much. With some of the BACE inhibitors, we’re inhibiting production by lowering amounts by about 80% or 90%. Well, is that too much? Is there a sweet spot? Should it be 20%, for example? So those answers are still out.

I think while there’s less enthusiasm for amyloid approaches in the dementia stages of Alzheimer disease, even though there are a few drugs that are still out there being tested, the understanding that if you intervene earlier in the amyloid pathway is still something that has to be investigated. Particularly, for example, in very early onset or familial patterns where individuals have symptoms, let’s say in their 30s and 40s, that’s a very different disease. And potentially by kind of turning off the spigot of amyloid production, or taking away plaques for people who have developed them, that may still potentially put them on a pathway of less neurodegeneration and tau formation.

I think there’s a lot of promise in tau, but these are still relatively early days. We don’t know which epitopes of tau to target, or how much targeting engagement to pursue. Should it be intracellular? Should it be extracellular? There are a number of programs that are investigating that, but I think that’s a very promising approach. There are also approaches for mitochondrial function, for neuroprotection, for modulating inflammation, which is again like a fire in the brain, and also the components of vascular brain injury. And also, pathogens. If one has amyloid, for example, certain pathogens that we’re all exposed to may actually make the effect of amyloid more toxic, or the effects of APOE4 [apolipoprotien E4], which is a risk factor that we have to induce tau damage much more.

So where this is going in 10 or 15 years is that we’ll be able to say not everybody has the same flavor of Alzheimer disease or dementia, and we should be able to get a particular profile with individuals’ amounts of amyloid and tau, and neurodegeneration and inflammation, and be able to diagnose that in a personal way but also hopefully have more targeted treatments. I think we’ll get there, but I don’t think there’s going to be a magic bullet. It’s really a complex system, and we require studies that go on for many years.