This transcript was automatically generated from the video recording and may contain inaccuracies, including errors or typographical mistakes.

More Than a Memory: Advancing Early Diagnosis and Amyloid-Targeted Therapies for Alzheimer's Disease

Introduction

Dr. David Geldmacher ( The University of Alabama at Birmingham): Good evening, everyone. I'm David Geldmacher. I am a professor of neurology at the University of Alabama at Birmingham, cognitive and behavioral neurologist, and tonight we'll be talking about More Than a Memory: Advancing Early Diagnosis and Amyloid Targeted Therapies for Alzheimer's Disease Management. Our presentation tonight will include some polling and audience participation, and we'll have an opportunity for questions at the end of the program.

Disclosures

Here are my disclosures. I work for several of the pharmaceutical manufacturers in clinical trials work as well as consulting.

Learning Objectives

We have a number of objectives for our learners tonight, including, as you see on the screen, integrating evidence and recommendations to support diagnosis of Alzheimer's disease, assessing the safety and efficacy of medications. Looking at the clinical role for novel and emerging therapies. Translating recent clinical trial findings into practice. Identifying benefits and considerations of different therapeutic means of administration for lecanemab and implement adherence and support strategies for patients receiving those therapies, including subcutaneous administration.

Defining Alzheimer's Disease 1906

So let's talk about what is Alzheimer's disease. Obviously, the point of our talk tonight is treating the illness. And in 1906, Alzheimer's disease was defined by Alois Alzheimer based on amyloid plaques and neurofibrillary tangles. And this remained the primary means of understanding the presentation of the pathological presentation of the illness and underlying the clinical presentation up through many years. But as of 2024, the definitions really became fundamentally changed, using a biomarker as the basis of the definition of Alzheimer's disease.

In particular, amyloid beta protein and its biomarker correlates.

2011: Alzheimer’s Reenvisioned

The pathway to that really became accelerated in 2011, when new criteria for Alzheimer's disease were published that recognized that amyloid plaques were accumulating in the brain of people for years, if not decades before the first symptoms, and they were reaching actually relatively maximal levels before individuals even reached into the stage of mild cognitive impairment. And at that point, the idea of preclinical Alzheimer's disease or asymptomatic - asymptomatic Alzheimer's disease entered into the lexicon. And that's become a key part for our understanding of the illness today, as well as opening up opportunities for treatment earlier in the course of the pathological process.

Case Study: Mrs. Z

So let's talk about one of my patients, Mrs. Z.

Case Presentation: Mrs. Z

She came to see me at age 65, accompanied by her husband. Chief complaint at the time was dementia. According to the patient, she was having problems with word finding, particularly proper names. And this was exacerbated by stress. There had been a major storm that had come through, and she felt that that really triggered a lot of her stress and their damage to their house and the like. Her husband noted that she had become repetitive. She had word finding difficulty and was more frequently misplacing items. Activities of daily living, including instrumental activities of daily living, like driving and managing the shopping were fully intact, according to her husband. She had a typical past medical history of hypothyroidism hypercholesterolemia. She had cervical spine surgery a number of years earlier, as well as carpal tunnel release. Both of those were no longer significant for her day to day life.

Case Presentation: Mrs. Z (cont'd)

As noted, she was married. She was recently retired as a nurse executive. She had a master's in business administration. Did not use either alcohol or tobacco. Family history was positive. Her father had had Alzheimer's disease. Her brother had a recent diagnosis of multiple sclerosis. On review of systems, they acknowledged that she was occasionally irritable, particularly when stressed. She was easily fatigued, but slept well. She had arthritic pain in both her hand and her neck, and she had seasonal allergies, which here in Alabama are a major problem for a lot of folks. An MRI was done and identified mild diffuse cortical atrophy and minimal white matter change.

Exam at Presentation

Because of her background with a master's degree and her ability to recount her history with good verbal skills, we chose to do the Montreal Cognitive Assessment at intake. She scored 20 out of 30 points on this, notably with zero of five recall and poor response to cueing. But she did recognize four of the four remaining words from lists. She struggled with the trail making test of executive function. She had minor errors on the cube and clock. And she had reduced fluency with a phonemic cue of how many words can you say in a minute, beginning with the letter F. 11 would be necessary to get the point on the MoCA. Her semantic fluency that is in a list of animals generated in one minute was also low for a person of her age and educational background. We would expect at least 14 and preferably more than that. And she generated only 10 animals in a minute. Her general neurologic exam revealed no focal abnormalities.

Workup To Confirm

So in our program, we use blood based biomarkers as a triaging test to gain access toward the PET scan because we have relatively limited PET resources and much better access to blood based biomarkers. As part of this process, we assessed apolipoprotein E status and she was an E3/E4. Her A beta 42:40 ratio in blood was low, and her p-tau217 was elevated, and that combination of the low A beta ratio and elevated p-tau was sufficient to then move toward florbetapir PET. And those images are shown here. And there is a loss of grey white differentiation indicating a positive scan with cortical amyloid plaques.

Treatment With Lecanemab

So we made the decision based on her diagnosis with mild cognitive impairment. And E3, E4 status to treat with lecanemab, we use an instrument called the General Activities of Daily Living scale to measure daily function, and that's reported by the spouse or other knowledgeable informant. And we use the Mini-Mental State Exam to track performance over time. And this was in part a requirement of her insurer. Beginning in March of '23, she had a 26 on the Mini-Mental and the full score of 26 and 26 on GADL. And she maintained that GADL scoring of 26, which means no significant daily functional impairments out through the next two and a half years. There was a period of stability and fluctuation with her Mini-Mental. And by two and a half years into treatment, her score was down to 21. So 32 months or 2.7 years, and she had a five point decline on the mini mental. And this is within the range of what we expect overall for people with MCI or early Alzheimer's disease. 1.8 points per year for individuals with MCI and two to three points per year in patients with mild AD. So she's - she's in that general range even on treatment.

Eligibility: Factors Influencing Patient Selection

So let's think about how we came to be treating her and what - what the factors are for eligibility for other patients.

Biomarker “Confirmation”

We used biomarker confirmation and we put confirmation in quotes here because that's really not what the biomarkers are doing. The biomarkers are pointing us in the direction of confirmation. And according to current practice guidelines, the blood based biomarkers like p-tau217, and A beta ratio are better at excluding Alzheimer's disease, the diagnosis than they are including it or proving it. They can support suspicions of Alzheimer's disease, but single standalone tests remain non-definitive in clinical populations. The A beta ratio is sensitive to age related amyloid accumulation. P-tau181 is increased in sleep apnea. P-tau 217 is increased in patients with kidney failure, so none of them are absolutely definitive. They assess for amyloid. They do not address tangles. I think this is important. Many people think because the name of the test is p-tau, p-tau217 or p-tau181 that they are assessing the burden of tangles, but it's actually an amyloid. It's a marker of amyloid severity rather than tangle severity, I hope. And the science is certainly moving toward the ability to assess tangles with assays like the MTBR-243, which should become available soon. It's available in research settings now. But tangle burden, tau burden is what really predicts the efficacy of the amyloid targeting monoclonal antibodies when there's more tangles. Those agents, the anti-amyloid agents, can be less clinically effective. So for now, in clinical populations, amyloid pet is definitive for plaque. CSF biomarkers are equivalent to amyloid PET. We do not use them in our site to the same extent.

Unfortunately for amyloid PET visual reads have proved less than consistent, and reader experience in reading amyloid PETs is critical for accurate determinations. More recently calculated centiloid values are highly desirable in mapping out treatment targets and response to treatment over time, and they provide a quantified or at least semi-quantified measure of amyloid severity that allows us to follow the response to therapy over time.

New Guidance on AD Biomarkers: 2025

So I mentioned the biomarker guidance. This came out from the Alzheimer's Association last summer. And the recommendation one is in people who have objective cognitive impairment presenting for specialized memory care. So those are two important qualifiers. The panel suggested use of a high sensitivity blood based biomarker as a triaging test in the diagnostic workup of Alzheimer's disease. And that's exactly what we do at our site. Recommendation two is in the same situation objective impairment, specialized memory care. If the test is high sensitivity and high specificity, that blood based biomarker could be used as a confirmatory test in the diagnostic workup, but no current blood test meets those thresholds for both sensitivity and specificity.

So good practices would say that a blood based biomarker test should not be obtained before comprehensive clinical evaluation by a health care provider, and the test results should always be interpreted within the clinical context. And in that clinical context, the provider should consider the pretest probability of AD pathology for each patient, remembering that there is age related increases in amyloid burden in the brain. And I'll talk about that right here.

Clinical Diagnosis vs Amyloid Positivity

First of all, clinical diagnosis are traditional clinical criteria for Alzheimer's disease are not as sensitive and specific as we would hope in the very first of the anti-amyloid agent trials with bapineuzumab, the subgroup who underwent Pet scan had a significant proportion. One of six participants in the PET substudy was amyloid negative, and all of them had fulfilled clinical criteria for AD dementia. Among APOE4 non-carriers over a third were amyloid Pet negative, so the clinical criteria were not as good at identifying amyloid positivity as we had hoped. The flip side is that amyloid positivity can increase without cognitive impairment as people age. And you can see here, the age ranges in the 50s, it's 2.5% increasing up to over 40% in patients in their 80s or non-impaired individuals in their 80s. These are folks with amyloid positivity without cognitive impairment. So both sides of this challenge are met that the clinical criteria may not be enough. And the amyloid positivity by itself may not be enough to make the diagnosis.

Is it Really a Neurologic (Structural) Disorder?

So as we approach these individuals, if - if neither the criteria nor the blood tests alone are sufficient, how do we assess who's going to have Alzheimer's disease and get the right people to the right treatment? So the first question to ask, is it really a neurologic or a structural disorder? There are software syndromes that is non-structural syndromes that can lead to cognitive complaints, particularly the complaint of short term memory loss: meds, mood, pain, sleep. All of those can contribute to memory complaints. But when assessing the actual performance of the individual, either through neuropsychological testing or even careful interpretation of bedside screeners like the MoCA or the Mini-Mental, those memory complaints are related to observed deficits in attention concentration and executive function.

The background between behind these software issues is really the classic arousal performance curve. When the person is having reduced arousal, moving towards sedation is extreme. Things like depression or medication effects, we'll see less than optimal performance. On the other hand, if the person is driven by high levels of arousal, such as with pain or anxiety and agitation, their distractibility and executive function also becomes more prominent and they drop off optimal performance. So these are things to consider. And old data suggests that at least 50% of people presenting for memory clinic evaluations of memory loss have mood components, either depression or anxiety. So these are fairly frequent.

The Straight A’s of AD Dementia: Recognizing Core Symptoms

So in trying to figure out who got Alzheimer's disease as the cause of the cognitive impairments, I think going back to the - the old recognition of what the core symptoms of AD dementia look like, what I call the straight A's of Alzheimer's disease. First is amnesia, and that amnesia is for recent events and activities. All of us will have occasional word finding lapses, misplace our keys, forget why we walked in the kitchen. But in Alzheimer's disease, the typical pattern is forgetting events and activities. So going to a birthday party and not remembering it is a lot different than not remembering when someone's birthday is coming up.

These are supported by the cortical features that match with the tangle distribution in Alzheimer's disease, things like word finding, difficulty, apraxia, agnosia, and so forth. But these are masked, often by low awareness of the symptoms. This is sometimes mentioned as denial of illness, but very frequently it has an organic basis. And the individual literally does not know that they're having these impairments. They forget that. They forget - they forget what - they forget and so become unreliable reporters of their own impairment. That's particularly true as we move out of mild cognitive impairment and into mild dementia due to Alzheimer's disease.

The other is because of low concern or low awareness. There's also low action or apathy. So people tend not to be seeking help or care for these cognitive deficits, either because they're unmotivated or unaware. And then to make the diagnosis of dementia, someone has to have reduced abilities in their activities of daily living. That is, reduced abilities in home and work roles, or what the DSM is called, social and occupational function. In mild cognitive impairment, higher level functions may take longer or require more effort, things like balancing the chequebook or paying the bills, but they still get done. And mild dementia assistance is required with those more complex activities.

Using Cognitive Test Data, Not Scores

So one thing I recommend, and of course, I am a behavioral neurologist, so I'm likely to recommend this more than some others perhaps is use the cognitive test data and not simply the scores to make our definitions of illness. So the Mini-Mental in particular, the score is not sensitive to the changes in early Alzheimer's disease, nor is it sensitive to the kinds of changes that are seen with the software issues of executive function and attentional changes. So errors on memory tests are non-specific. They can be attributed to losses in other cognitive domains.

However, when memory errors are present, orientation to time is very suggestive of AD patterns, so that if a person is missing the day, date or month, those are things that are red flags for significant or meaningful cognitive decline. The clock drawing test can also be very informative. This is something that my group studied a long time ago, but when comparing mild stage clinical AD and unimpaired controls, errors in the clock, particularly in the hands of the clock, had a high predictive value for Alzheimer's disease. So based on these data, I recommend screening with the mini-cog instrument and then adding orientation to day, month, and year.

Practical Approach to Cognitive Assessment: Mini-Cog + Orientation to Time

And this is the algorithm for considering the mini-cog. The mini-cog involves providing three words for the patient to repeat, then having them draw a clock and then recall the three words. And if a person is zero of three on the recall, they need further evaluation. If they have three item recall, they're likely not to be demented. And if they recall one or two of the three words and then the clock is abnormal, they probably need further evaluation and the clock is normal. This is someone who might be most appropriate for further observation. You can see there the mini cog sensitivity and specificity for detection of dementia. And of course when we're talking about early Alzheimer's disease that also includes mild cognitive impairment, not just dementia. So we can add further information to our - our clinical decision making by looking at that orientation. This orientation to year, month, and date.

Clinical Trials Outcomes

So let's look at some of the clinical trials outcomes for the anti-amyloid antibodies. And this is in the IV infusion forms that were studied in the phase 2 and 3 trials. Lecanemab phase 3 clinical outcomes. The primary outcome that was used in this study was the CDR sum of boxes. These were people who were amyloid positive with MCI and mild dementia. And over the course of 18 months, there was a 27% reduction in decline in that primary outcome measure. Among the secondary outcomes, you can see that the biomarker outcome of amyloid clearance on PET scan was remarkably positive, and the others cognition, daily function and composite measures like the CDR showed similar slowing of progression, reduced decline over the interval of treatment.

Donanemab Phase III Clinical Outcomes

Donanemab had very similar outcomes. Their primary outcome was something called the iADRS 22% decline. They're looking at the CDR sum of boxes, a 29% decline in the overall treatment population, again over 18 months of therapy. These were large scale trials 650 to 700 800 people in each of the treatment groups.

A Closer Look: CDR-SB Side by Side

If we look at the same outcome measure from the two trials and recognizing that statistically this is not a comparison that can be made with statistical analysis, but just looking at the raw data and the - the reported percentage reductions in decline very, very similar between the donanemab and lecanemab populations. I will show that there are differences between the populations in terms of their amyloid burden and the like, but just raw clinical outcomes look pretty similar. And if we map this out, what does a one year outcome look like? On this CDR sum of boxes, it's about a one point a little bit - a little bit less than one point change very similar between the two agents.

Comparison: Efficacy by the Numbers

So here are some of those differences between the populations. Lecanemab population had a lower amyloid burden and had a somewhat less percentage of amyloid clearance out of that lower starting point. However, as I saw, as we've shown, the primary clinical outcomes were very, very similar.

Comparison: Donepezil and Lecanemab 1-Yr Data

There's also a misperception among many of our patients that these agents are going to cure their Alzheimer's disease or they're going to improve cognition. But I just wanted to show this comparison again, different clinical trials now statistically compared but looking at outcomes. So our standard long term treatment for Alzheimer's disease, dementia donepezil first approved 30 years ago. At the end of one year in this placebo controlled trial, there was no statistically evident decline from baseline. If we look at one year from initiation of lecanemab therapy, however, there is a significant decline from baseline. So these two agents, these two classes of agents can be used together and they have different outcomes. When I describe this to our patients, I talk about donepezil helps the living brain cells work better, and the lecanemab helps them last longer. And the two of them in combination offer real value to our patients.

Risks

Like any medication, the anti-amyloid therapies come with risks.

Amyloid-Related Imaging Abnormalities (ARIA)

And the primary one that we - we have to be concerned about because of its potential severity, is the risk called amyloid related imaging abnormalities. This first emerged - first was recognized in the earliest treatment with anti-amyloid antibodies with bapineuzumab back in the 2000s. The two big flavors of ARIA are ARIA-E, the E stands for edema or effusion, and ARIA-H, which stands for hemorrhage, and this includes micro and macro hemorrhages as well as superficial siderosis. The ARIA appears to be related to accumulations of vascular amyloid and possibly inflammation associated with that.

There are predictors of who is more likely to develop Aria on an amyloid targeting antibody. People with APOE4 genetics are at more risk, and it is related to gene dose. And people who have had prior hemorrhages are at increased risk for new hemorrhages on the amyloid targeting antibodies. In the clinical trials, ARIA was most frequently asymptomatic, so it requires prospective awareness that is, knowledge that the risk exists, and then these are managed or identified with scheduled MRIs based on typical periods of -of risk when the imaging abnormalities might emerge. That's usually in the first six months of treatment.

Comparison: Safety by the Numbers

So if we look at some of the dosing that's available and the risks for ARIA, it's important to know that the package insert for donanemab, the original package insert, used an old dosing regimen, and that's been recommended to be superseded by a new slower titration, which just - which does change the risk numbers. But we can see that in people who are E4 homozygotes, we're seeing a 25% to 35% risk for ARIA-E. We're seeing a 29% to 39% risk for ARIA-H. And these are the raw measures of that. Not - not adjusted for placebo.

ARIA Rates in Lecanemab Phase III

So significant numbers, but when we break that down to symptomatic that is people who had complaints associated with detected ARIA or serious, that is requiring intervention or hospitalization. Those numbers are not as threatening as they might seem, just based on the raw numbers. So only about 10% of APOE homozygotes experience symptoms from ARIA-E on their treatment. And symptomatic hemorrhages did not exist in that group and didn't - didn't exceed 1% in any of the groups. You see, the serious adverse events which required intervention were somewhat more frequent, but nowhere near that double digits that we see with the just overall count of ARIA-E and ARIA-H identified by routine screening, scheduled MRI scans.

Qualification Process: Pragmatics

So pragmatically, when we think about initiating anti-amyloid therapies in our clinic, we need an MRI at least 1.5 T within one year, and it needs to have susceptibility weighted imaging or some equivalent to that. And we trust our radiologists to do the reads on that, but it's important to maintain consistency of that MRI scanner and the software. That is what sequences are used over time. If you switch back and forth between different modalities like GRE and SWI. The different sensitivity to microhemorrhages and siderosis may mislead you in safety monitoring.

Therapy is contraindicated with more than four microhemorrhages or significant siderosis. APOE genotype is important for patient education and risk determination. And that's now readily available through commercial labs. And there's good evidence that communicating APOE status to symptomatic individuals does not cause them psychological harm in general. We use diagnostic prediction versus our - via our blood based biomarkers using the combination of the A beta ratio and the p-tau. And then using amyloid pet to confirm that and quantify that, to give us a quantifiable measure of amyloid burden and success of amyloid clearance with the agents.

Practical Treatment Considerations

At a practical level, we're now considering IV versus subcutaneous treatment for maintenance of therapy. Lecanemab is now indicated for maintenance therapy beyond the 18 months of initiation. And here are some of the differences and similarities between the drugs for initial administration. The first 18 months IV is recommended. Dose frequency of Q2 weeks for lecanemab, Q4 four weeks for donanemab. 18 months is the recommended initial regimen and for - for lecanemab - but for donanemab, assessment of therapeutic success at 12 months is appropriate. MRI monitoring is scheduled. And you can see that for each the - each of the elements of the MRI monitoring that there has - there are four scheduled MRIs at similar intervals.

A big difference between the two agents is that for lecanemab, based on its proposed mechanism of action against protofibrils maintenance therapy as continuation of therapy after the 18 months is recommended, but for donanemab consideration of treatment discontinuation when amyloid is cleared is appropriate. For that maintenance therapy, lecanemab is available both for IV and subcutaneous use, and subcutaneous use is weekly home administered self-injections and for the donanemab treatment until amyloid clearance is obtained is recommended, but after that is not required. So how do we choose between IV therapies and subcutaneous therapies for these individuals? Well being able to manage treatment at home is an important value for many families versus repeated visits to the infusion centers or clinic.

For some of our patients getting into the infusion center and the social aspects of that is actually very valuable. They enjoy that and they've chosen to maintain IV maintenance therapy rather than subcutaneous because of the ability to maintain the relationships that were built over 18 months of ongoing initiation therapy. Being able to get the medication and get it to your home is an important consideration versus having a clinic - having to get it. Administration time, so the self-administration takes minutes and the administration by IV is hours or more.

So for many individuals the self-administration is much preferable on multiple levels. However, we have people who have injection site reactions, people with needle anxiety, people who value the ongoing contact with the nursing support in the infusion center is valuable. Although there is nursing support available for both. So very much an individualized conversation with the patient and family as to what would work best for them for their maintenance therapy IV versus subcutaneous.

On the Horizon

Now, there is obviously active research going on in the field, and there is active research continuing with the amyloid targeting therapies. The FDA evaluation of subcutaneous weekly dosing of Lecanemab is for treatment initiation is currently underway, and the FDA decision is expected sometime during May as to whether treatment initiation with subcutaneous form will be appropriate or not. Our patients here at our memory clinic are voicing a lot of excitement about the possibility of getting to self-administration earlier than 18 months in and for all the reasons that I mentioned in the last slide,

Roche - Genentech, Roche is testing a brain shuttle technology to optimize delivery of the anti-amyloid antibody into the central nervous system. That's been one of the big challenges, is getting sufficient dosing into the brain without triggering the vascular issues associated with ARIA. And so a brain shuttle technology should be able to deliver higher parenchymal dosing with lower ARIA risk. And this is in clinical trials right now.

And both lecanemab and donanemab are in studies of preclinical Alzheimer's disease. If you recall back toward the beginning of the presentation, I mentioned that amyloid plaque is building up for a considerable period before the first cognitive symptoms. Those are the individuals who are in testing right now, and we expect those studies to read out sometime in the next several years, 2028, 2029. Not all the research has been positive. So in terms of recent failures, semaglutide in an oral form did not provide efficacy in Alzheimer's disease. And an anti-tau antibody posdinemab also was negative. And these were both reported within the last year. The field remains in very rapid evolution, particularly in the biomarker side of things, and I expect that before long, we will be able to give assessments of tau burden with tests such as the MTBR243, as well as the existing measures of amyloid burden. And that may reduce the need for the more expensive diagnostic testing like the PET scans, and be able to push more of the diagnosis into earlier stages with non-specialists. So those are all opportunities that are on the horizon.

Summary and Conclusions

So let me take a few moments to summarize. Beginning monoclonal antibody treatment - antibody treatment reduces progression of Alzheimer's disease. The amyloid removal is readily documented by PET scan. The clinical effects are difficult to discern at an individual level. There's so much variability in the progression of these illnesses or the illness of Alzheimer's disease based on things like age, concomitant illnesses, and so forth. So it's very hard in any one individual to say that the treatment is working. And as a result, public health benefits remain under study. Is there enough cumulative benefit to really know that the medication is providing benefit for the - for the burden the patients bear and as well as the health care system?

That's one of the reasons that both agents are required to - people on the agents are required to participate in registries to gather cumulative - cumulative data on clinical effectiveness, risk, and so forth. What we do know is that patient selection is critical. First of all, it's important to know that Alzheimer's disease is causing cognitive syndrome, and that's where the biomarker testing and the PET scans come in. We have to be able to effectively evaluate and communicate the risk of the treatments. So how does APOE status influence the risk for ARIA? What about anticoagulation? What about pre-existing cerebral amyloid angiopathy? All of those are things that significantly influence the risk of complications. And in that individual, we may shift the risk and benefit ratio. And then predicting benefit, there's growing evidence that people who have milder stage amyloid accumulation that is lower centiloid counts may have greater value - A greater benefit from the treatment in terms of maintaining independence or slowing progression across disease stages.

My experience has been that patient burden is significant. The frequency of infusions every two or every four weeks interferes with things like travel, family get-togethers and so forth. This is being addressed by alternate dose forms such as the subcutaneous administration that we'll learn more about next month. And then the risk for infrequent but life altering drug induced complications require - that require frequent MRIs. That adds to the burden of the patient having to get to scheduled MRI scans. And although the raw numbers of individuals who have serious or life threatening ARIA events is low, those are very serious for the people who do experience them. And we can't discount that.

Another important factor provider burden is substantial and in general is uncompensated maintaining individuals in the registries, making sure that they're getting their safety scans, making sure their safety scans have been reviewed by the appropriate individuals in the appropriate time frame is a lot of work. And for the provider who doesn't have the kinds of institutional resources that I'm able to - to have here at a university memory clinic, it can be quite a bit. And many of our community neurologists at this point have chosen not to administer the agents or to prescribe the agents, instead referring. This, I think, will get better over time as we learn more about the agents and become more comfortable with the risk and severity of the adverse events.

Another burden on us is managing ARIA when it emerges. So there are very few guidelines except for stopping the agent in terms of what to do when a person experiences ARIA, should they receive anti-inflammatory agents like steroids for how long? How do we know when to stop those added on treatments and so forth? So you need more guidance on that than we've gotten so far and that literature is growing. On the other hand, although provider burden is substantial and generally uncompensated, organizational benefits can be highly attractive. Specialty pharmacy revenue, infusion center revenue, scan revenue are all things that organizations look for, and that my organization has looked for certainly as we have developed our amyloid treatment therapy programs.

So those are, I think, important things for consideration. There are substantial pros for our patients who are able to tolerate the agent and can experience slowed disease progression, but there is a lot of work that goes into that from the provider base and from the caregiver base.

Let's Vote!

Okay. So we'll go through some now reassessment of knowledge now that you've heard the data.

Posttest 1

So of cognitively unimpaired people age 60 to 69, about what percentage demonstrate amyloid pathology on PET? The CNS is the unimpaired people age 60 to 69. Go ahead and vote. Is it

A. 5%;

B. 10%;

C. 20%;

D. 25%; or

E. 45%.

Okay. We can pull up those results. There we go. Yep. That's exactly what we talked about. So about half of the audience said 20%. And that is appropriate for that 60 to 69 age group. We'll move on to the next question. Now there's the rationale. Sorry about that. About 20% 18.6 in the Mayo Clinic study of ageing has been going on for a long time. They have a very large and cumulative data set for this.

Posttest 2

Okay. Now, second question. What rate of slowing in an 18 month study was seen as a primary outcome in the placebo controlled phase 3 trials of lecanemab and donanemab in patients diagnosed with early Alzheimer's disease? Was it

A. 5% to 10%;

B. 10% to 20%;

C. 20% to 30%;

D. 35% to 40%; or

E. 50% to 60%.

Go ahead and vote, please. Okay. We can plot the answers to this one. All right. So 20% to 30%. Yes. The data pretty consistently were in that range for the clinical outcomes. And in clarity AD that was the lecanemab phase 3 trial. Looking at their primary end point of the CDR. Some of the worst thing was reduced by 27%, and in TRAILBLAZER, the donanemab trial primary endpoint of the eye was reduced by 22%, both falling within that 20% to 30% range.

Posttest 3

Okay, question number three, what's the major potential benefit of brain shuttle technologies over the currently approved anti-amyloid therapies? Is it their

A. Efficacy in later stage AD;

B. Efficacy in patients with known lower amyloid burden;

C. Lower risk of ARIA; or

D. Lower risk of infusion related reactions.

Please pick one and vote. And we can close that out and look at the results. All right. Lower risk of ARIA. 61% of the audience chose that. And that is exactly what I presented and that across all three of these questions, the audience vote - the audience preferences have moved in the direction of the data I presented. So I think that those learning objectives are on their way to being met. So here was the brain shuttle. Antibodies improve the blood brain barrier penetrance, improve safety, such as ARIA risk compared with those agents that don't have that additional opportunity to get an augmented concentration in the brain.

Thank You! Question and Answer Session

All right, so we can open this up for questions and answers at this point, and see if we have any that have come in so far.

Speaker (DCE): Okay, great. We do have a few questions from the audience and we have about 15 minutes. So audience members, please, if you - if you have any more, please feel free to send them in. But the first question is I have heard that anti-amyloid treatments can cause white matter loss, i.e., atrophy. Is this true? And if so, how do you consider this effect in deciding whether to treat.

Dr. Geldmacher: That's a great question. And yes, the data do tell us that there is reduced brain volumes at the end of treatment in many individuals. I'm not sure it's just white matter. I think it's overall brain volume. And one of the arguments, one of the reasons that's been put forward for this, none of it's been proved yet is that the reduced amyloid plaque reduces the size of the brain and therefore we get atrophy. To me, the relationship between brain atrophy and cognitive function is a fairly loose relationship. And we know from the clinical outcomes like the CDR sum of boxes that patients on therapy did better than those on placebo. So I am not overly concerned about the reports on brain atrophy.

Now, if it turns out that there's later longer term adverse effects related to that, I'm certainly willing to be updated by the data, but the current ones it's - the atrophy does not seem to be the driving factor in clinical outcomes.

Speaker: All right. Another question. Are we moving away from using memantine with or without donepezil?

Dr. Geldmacher: So that's really off topic for this. Memantine is certainly indicated for moderate to severe dementia due to Alzheimer's disease. Its efficacy seems to be optimized when used in conjunction with donepezil. But it is a very different space in our treatment than the anti-amyloid antibodies and has a different mechanism than the donanemab or sort of donepezil does. So we still use it regularly in our clinic for people who are in moderate to severe stages of Alzheimer's disease.

Poll 3

Okay. So we're going to get a couple of more poll questions going on here. Do you plan to make any changes in your clinical practice based on what you learned in today's program? This is a single choice. Go ahead and vote for that. And we're not going to review these results I don't think. All right. Other questions.

Speaker: And we can just keep going with questions while those last poll questions pop up. Here's another question about MRIs. In my part of the country, scheduling an MRI takes weeks or months. Do you have issues scheduling the frequent MRIs for monitoring protocols?

Dr. Geldmacher: Yes we do. And we've had to turn to our MRI group and let them know that we're going to be having these scheduled MRIs and scheduling them months in advance, actually, in order to reserve slots. We have had to shift some of our PET scans off of our MRI scanner so that now we're getting our - our confirmatory PET scans on the Pet CT rather than the PET MRI to be able to maintain MRI availability. And we are using community sites to get MRIs for many of our patients. I live in a very rural area. And people getting to Birmingham, maybe a two or three hour drive. So we are using peripheral sites with that. But trying to get our radiologists to provide the quantitative - the reads on quantitative assessment of microhemorrhage and things like that has been its own struggle. So yes, the safety monitoring is burdensome on patients and on systems.

Speaker: And sort of a question in relation to that, how do you manage monitoring for - for ARIA if your radiologist is not experienced in detecting them?

Dr. Geldmacher: So I haven't had that problem because our radiologists have participated with us in our clinical trials as we've done them. So we've been able to build that expertise in that - in our radiology community locally. And then they have their own training protocols. There are training protocols for ARIA detection that are out there and CME processes. But it often is something that requires some interaction between the more experienced clinician. I've been involved in the anti-amyloid therapy clinical trials dating all the way back to 2006 where we first learned about ARIA. So there's, there is a dynamic there, and yes, we do need to - to do some shared training with our radiologists on ARIA detection and quantification. We also refer them to the appropriate use criteria paper - papers that talk about the quantification and measurement and determination of severity of ARIA when it exists.

Speaker: And I should add on that note that at Decera, we just put up an ARIA training module, a radiology focused module with both a video and a text module available. So if you search for ARIA on our - on our website. Anybody who's interested in that CME please - please do. So let's see. Here's a question about older patients, which says since the disease can present clinically in the very old, such as above age 85 or 90, do you restrict usage of these agents in the very old? And they obviously will have more commonly medical comorbidities.

Dr. Geldmacher: Yeah. So we do make individualized clinical decisions and age is part of that determination. Amyloid burden is also part of that determination. The clinical trials generally excluded people at age 90 and above. And so we have very little data in the oldest old and the relative benefit toward maintained function or even life expectancy in these agents. So we don't use a blanket cut-off of age in our clinical consideration if I had an extremely robust and healthy 92 year old whose brain looked like it was a 65 year old's brain. I might go ahead and treat even though we don't have much data on that.

I would also share with them that we're making this decision based on assumptions rather than data. But the older patient who's got more medical comorbidities, particularly things like kidney failure are people that we are more cautious about and those who have relatively limited caregiver support. So for instance, if that 90 year old was living by himself and didn't have someone to report ARIA's symptoms to us, we might be more cautious about that in our group. But at this point, except for that explanation that we don't have clinically determined data on people over age 90, we don't have any hard cut-offs.

Speaker: Okay. All right. I have a question from a neurologist with their own practice who says designing a workflow for anti-amyloid treatments has been quite intimidating. Do you know of any resources out there to help small neurology practices be able to get more comfortable incorporating these therapies.

Dr. Geldmacher: Yeah, it's a question that I get from our community neurologists here in Alabama. And I am not aware of those resources. I think some of your best bets might be to speak with the representatives from the companies, particularly the medical scientific liaison. I know that for a while the manufacturers lecanemab have had a visiting nurse kind of program available to help monitor patients. I don't know the current state of those. I know that organizations like the Academy of Neurology and the Alzheimer's Association do have educational programs on this, but that - that pragmatic aspect of how do I maintain safety monitoring, get people scheduled and get well-qualified reads on both my PET scans and my MRI scans are - are major concerns for our community neurologists here in Alabama. And I don't have a good answer to how to fix that.

Speaker: All right. Great. Here - here is a question. Are you aware of and do you have an opinion about a recent Cochrane review of anti-amyloid therapy.

Dr. Geldmacher: I am more than aware of it. I think every patient I've seen on therapies in the last week has brought it up. The Cochrane review has a major flaw and that is that it is incorporating all anti-amyloid therapies into the review. It's including multiple trials of which only two had clinically positive outcomes as determined by their statistical models. So all anti-amyloid therapies are not the same. Some target monomers of amyloid. Some target only plaques. Some target soluble fibrils or protofibrils. And so this would be like taking a bunch of different antihypertensives and saying you know, or different antihypertensive trials and lumping them all together regardless of mechanism.

So the fact that amyloid is the target, sure. But what - how is amyloid recognized? What's the epitope? That varies considerably across the agents. The other is that the earlier anti-amyloid trials limited dosing to limit the frequency of ARIA. And it's become clear that there is a dose response relationship between the amyloid clearance ability and the ARIA. So I don't think that the inclusion criteria in that Cochrane review were appropriately selected. And if you looked at the two positive trials, you would your - your - your meta-analysis would be positive. If you looked at the seven negative trials in that paper, your meta-analysis would be negative. But there were a lot more negative trials than there were positive ones. So it's hard for the meta-analysis to be considered valid or to - to translate those results to the two drugs that have shown positive results in their phase 3 trials.

Speaker: All right. And I think - and I think with time we'll just do one more question. In your practice, do you repeat amyloid PET to monitor response to treatment? And then what are your guidelines for continuing or discontinuing treatment based on that?

Dr. Geldmacher: Yeah, that's something we didn't predict when we launched into amyloid therapy programs here, was that we would start to use PET scans to measure response to treatment, and we now routinely do that. So we use centiloids as our overall measure, and we look for centiloids counts below 25 to say that - that was an effective course of therapy. And if at the end of 18 months of therapy, the centiloid count is still above 25, we have recommended to our patients to continue with the ongoing for lecanemab every two weeks treatment and to continue the donanemab treatment at its usual Q4 weeks interval until we get to those doses or those amyloid levels below a centiloid of 25.

And that mean. That means bringing them out to 18, 24 months of therapy. I think our longest duration of treatment for a patient now is out close to two and a half years and still having sort of very resistant amyloid in their brain. We're well-beyond what the data tell us at that point. We're going to just try and go with the understanding that more - more plaque is bad and less plaque is - less plaque is better.

Downloadable, print-ready Clinical Practice Resource

Speaker: All right. Well, with that, I want to make sure that we also mention you can see on the screen right now the clinical practice resource that Dr. Geldmacher, you also developed for us. So this is a - basically a summary of - of what's been covered in this talk on the diagnosis and treatment of Alzheimer's disease. It's downloadable PDF which you can find on our website, or take you directly by scanning the QR code that you see on the screen right now. And Dr. Geldmacher, I'd just like to thank you for your time again for being here with us this evening. Thank you all the learners from everywhere that you are from. Dr. Geldmacher, do you have any final thoughts, anything that has come to your mind that you didn't get a chance to mention before we close?

Dr. Geldmacher: Yeah, absolutely. This is a rapidly evolving field. This is a different talk than I would have given a year ago, and it is a different talk than I will give a year from now. So certainly this is something where neurologists treating Alzheimer's disease and other - other providers treating Alzheimer's disease should stay attuned to the biomarker and the therapeutic developments because they're changing very quickly over time.

Speaker: All right. That's very true. Very true. Well, thank you and good night. I look forward to speaking to you again in the future.

Dr. Geldmacher: All right. Good night everyone. Thanks for your attention.