Disease Management on the Horizon in Alzheimer Disease

Marwan Sabbagh, MD

With multiple advancements for Alzheimer disease making their way through early stages of clinical development, the excitement around treatment has never been greater.

For Marwan Sabbagh, MD, director, Cleveland Clinic Lou Ruvo Center for Brain Health in Las Vegas, Nevada, this excitement is different than previous iterations. In a field that has seen hundreds of failed trials and therapies, cautious optimism is the norm, but Sabbagh believes what many in the dementia space are seemingly beginning to share: This disease could soon be manageable.

With that potential on the horizon, Sabbagh told NeurologyLive that clarification about the cause of dementia is becoming more critical. Physicians need to know not just that there is dementia, but what state the dementia is in and its etiology. Sabbagh provided insight into this, among other topics, in a recent interview.

NeurologyLive: How important is it to identify and treat neuropsychiatric symptoms?

Marwan Sabbagh, MD: It’s a big deal. When we really talk about the dementia state now, we draw down on staging it as mild, moderate, and severe, and we start to see the emergence of the neuropsychiatric features in the late mild, early moderate stage. Oftentimes, they can eclipse the entire disease. After a while, patients and their families don’t care about the forgetfulness, they care about the wandering, the paranoia, the delusion, the agitation, the trouble sleeping, and that really starts to take over the disease. It becomes a huge amount of the burden of the disease.

While we know that we use off-label drugs, there are no clear consensus guidelines, and we’re winging it when it comes to managing neuropsychiatric features. There are several drugs being developed for behavioral symptoms, and those are actually very, very close to being approved. I suspect we’ll see more drugs approved to treat behavioral symptoms within a year or two. That is a huge area of unmet need, and I suspect we’ll see some drugs to fill the gap.

How important is it to detect the disease early?

Very. There are the new ATN criteria—amyloid, tau, neurodegeneration criteria—which is redefining, and this came out this year, in 2018. But the ATN criteria recasts the disease from a clinical construct to a pathological construct. This has not been applied clinically, but it will kind of evolve. The reason this is important is we know that pathological changes precede the onset of symptoms by up to 20 years, and, in fact, amyloid can be detected as early as 20 years before the onset of symptoms, according to the DIAN [the Dominantly Inherited Alzheimer Network observational] study. DIAN has shown clearly that they can detect changes in the brain decades before the onset of symptoms. The point is there’s now a push to detect the disease with or without symptoms.

The reason I say this is that people mix terms together, and precision is becoming paramount in how we discuss things. We no longer say Alzheimer disease. We would say dementia due to Alzheimer or due to Parkinson, Lewy bodies, etc. So, the state would be dementia, meaning clinical cognitive decline associated with functional impairment—that’s dementia—and we would talk about that followed by the etiology. Then, with mild cognitive impairment, again, it’s the state followed by the etiology. Impairment due to Alzheimer, etc. Then we would talk about a preclinical state of Alzheimer disease. All of it is Alzheimer, whether you have symptoms or you’re fully demented. All of it is Alzheimer, but we’re starting to talk about Alzheimer in the different states as 1 disease in different states, and that’s why the ATN data [are] going to be so important moving forward.

Can you speak to the misconceptions you’ve identified in the space?

The first common misconception was the definition that I just talked about, confusing disease with the disease state. People use the terms interchangeably. It’s important that we define the state—meaning normal, mild cognitive impairment, dementia— and then etiology. That’s misconception No. 1.

Misconception number 2 is that you can only diagnose Alzheimer disease with an autopsy. That was what was put forth in 1984 using NINCDS—ADRDA [National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer’s Disease and Related Disorders Association] criteria. I am here to tell you that our clinical and specialty centers like mine and other centers around the country, we have precision rates north of 90% accuracy with the use of PET [positron emission tomography], CSF [cerebrospinal fluid], genetic markers, other biomarkers. We are very much farther along in using advanced diagnostics to diagnose Alzheimer with great confidence.

Then, misconception number 3 is the field is still mired in this large veil of futility, that “there’s no point, so why bother?” We are so much closer than people think to changing the trajectory of this disease. We are so much closer to making this from a terminal disease to a chronic disease. This is the diabetes of our time. This is the HIV of our time. We will do it, not a generation from now but fairly soon. Not cure it. Don’t get me wrong, that’s not what I want to say—but manage it. You’ll come in, we’ll do some diagnostic tests, we’ll diagnose the Alzheimer, we’ll put you on a treatment plan, which will probably be a combination of 3 to 5 drugs and then you’ll stabilize, and that’s what’s going to happen, fairly soon I suspect.

Can you discuss the current therapeutic landscape?

It’s changing very rapidly, and we, in the last, say, 2 years had a big push for the monoclonal solanezumab, which was a monoclonal targeting monomeric Aβ [amyloid beta], with the idea that by clearing out amyloids, you would slow down the progression. It did not meet its prespecified endpoint, and for the most part, for symptomatic dementia, solanezumab has not moved forward. The reason for that is because the prevention trial that uses solanezumab, the A4 (Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease) trial, is moving forward and continues to move forward. In the monoclonal antibody space, there are a few compounds that are still moving forward, and again, the idea is they target amyloid and clear amyloid. One is aducanumab, that’s the Biogen drug. [Editor’s Note: In March 2019, Biogen and Eisai announced the discontinuation of the aducanumab clinical trials due to futility.] New to the scene is BAN2401, which just announced their data about 2 months ago [at] AAIC [Alzheimer’s Association International Conference], showing a commensurate change in direction, with amyloid clearance and clinical stabilization going in the same predicted direction.

Monoclonal antibodies for Alzheimer are still moving forward, monoclonal antibodies that are targeting tau are starting to pick up steam, and there are 2 that are now in phase 2 trials, AbbVie and Biogen.

Base inhibitors, which are oral anti-amyloid production oral medications, have not shown as much success as we would like. The Merck drug verubecestat did not meet its prespecified endpoints, and I think, for the most part, [the trial] has ended. The others in the class have had significant emphatic toxicity, and therefore I don’t see the base inhibitors making a lot of progress moving forward. The early study, which was a version of the A4 study, which was a prevention trial using a base inhibitor, was stopped because of liver toxicity, and so again, the class of base inhibitors [is] not moving forward.

In the last 18 months, we were also hot about a class of drugs called 5-HT6 antagonists, there were 2 lead compounds: idalopirdine and intepirdine. One was the Lundbeck compound, and one was the Axovant compound, but both failed to meet the prespecified endpoint of efficacy in their phase 3 programs, and for the most part, that class of drug has been abandoned.

Another drug that has been looked at is the RAGE [receptor for advanced glycation end products] compound; this is Azeliragon. That drug didn’t meet some of the prespecified endpoints in their phase 3 study, but there is some efficacy of signal still, so I think the company is moving forward with more studies. And so that class hasn’t concluded its direction.

All these data I’m talking about are in the last 18 months. Everything I’m discussing was not 10 years ago; this is very current. The point here is that there are many, many drugs and devices, for that matter, actually in the phase 3 space, and there’s still a cause for excitement and hope.

What are the value of biomarkers in this space, and which are specifically needed?

That’s a very important question, very germane to the timing. Ten years ago, we would’ve said detection of plasma amyloid is unreliable and we would have said you are dreaming if you are even trying to work on detection of plasma tau.

In the last year, there’s been huge traction around the concept of reliable detection of plasma amyloid that the ratio of Aβ40 and Aβ42 turned out to be a very good predictor of the disease process. There are now several companies working on reliable assays for plasma tau, and the reason there is a real hard push to peripheral diagnostics is that CSF testing is expensive, invasive, and hasn’t picked up as much traction as we liked. Athena has owned that space for a long time, Roche and Fujirebio are developing CSF detection assay platforms, and probably within the year, we’ll see alternatives to the Athena platform for [the] CSF detection assay. Ideas showed that there was at least, in the preliminary data, some evidence of clinical utility for amyloid imaging in the clinic; however, it is still not reimbursed. Amyloid imaging has not become a common practice in clinical medicine.

The point is that there’s a huge area of unmet need on the diagnostic side, and peripheral diagnostics are moving forward because they can be far cheaper and far more efficient. So it’s getting a real hard look right now. At the AAIC, there were many discussions about development of peripheral biomarkers.

Physicians need to know this shouldn’t be long term. I suspect we could see peripheral blood type tests within 5 years, certainly. Even maybe sooner than that. Additionally, physicians need to know what they’re all going to look like from a medical standpoint. They would kind of be like a prostate-specific antigen for a man or hemoglobin A1C—not inherently diagnostic, but detection of [a] signal that would warrant further investigation.

What are you thoughts on the amyloid hypothesis?

I like speaking to my peers, and I’ve done a lot of web work, WebExs, video seminars and conference calls, and the reason I say this is because my field is changing so rapidly. But to most physicians, they think everything is staying the same and are just giving Aricept [donepezil] and assuming nothing’s changing. But that is absolutely not true.

Many people believe amyloid [buildup begins] early in the disease process, it is an early seminal event. Overproduction or undercleansing of amyloid leads to a cascade of events, [which] leads to tau, neurofibrillary tangles, inflammation, excitotoxicity—so that the antecedent event to all the events that occurred later on is the overproduction or undercleansing of amyloid. There are species of amyloid that are toxic and [with which] accumulation of amyloid tend to be negative or deleterious.

People say the amyloid theory is dead. I think the amyloid theory is more nuanced. We can’t be reductionists and say we’re amyloid, we’re baptists and tauists; it’s not that simple. We know that in symptomatic dementia, amyloid does not correlate well to clinical progression; tangles, tau, and synapse loss correlate better to symptomatic dementia.

We’ve been using drugs to target amyloid in symptomatic dementia, and a lot of them have failed. The question is, is it because it’s the wrong target or is it because we’re treating the disease too late? Meaning the brain’s full of amyloid [and] there’s no point in trying to cut it out once your brain’s full of amyloid. I think a lot of focus is on using drugs that target amyloid in a more dynamic state earlier on.

We were really down on production or clearance—[with] production meaning the base inhibitors that I talked about [and] clearance meaning monoclonal antibodies that I talked about— as not working. The early generational drugs of bapineuzumab or solanezumab showed a discordance between signal efficacy and target engagement. Bapineuzumab showed very reliably you could reduce the amount of amyloid in your brain on PET scan but didn’t show a lot of clinical signal, probably because we kept reducing the dose. Solanezumab showed a clinical signal in the early studies but showed no evidence of cutting amyloid on PET.

The new generations of the drugs, particularly aducanumab and BAN2401, show directional concordance of reduction of amyloid correlating well with clinical stabilization. Amyloid is still important, still germane to the conversation, still a reliable target, still important to the whole overall conversation, and we still could see drugs approved through that pathway.