5 Themes Shaping the Future of Dementia and Alzheimer's Care

[00:14 - 01:16]
Introduction
Slide 1

For those of you I haven’t met, I’m Mary Patterson. I am the VP of Clinical & Regulatory here at BrainCheck.

My background is in Neuroscience and medical device development, but my work at BrainCheck really sits at the intersection of cognitive assessment, clinical research, and how to actually translate advancements in Alzheimer’s science into routine clinical care.

Back in July, I had the fantastic opportunity — for the fourth time — to attend AAIC this year in London.

There is always an enormous amount of science presented. So, rather than to tackle the enormous task of summarizing all of the hundreds of studies, today’s focus will be on what emerged as five themes that shaped the meeting.

Then, more importantly, what they really tell us about where dementia and Alzheimer’s care are headed.

I’ll also spend a few minutes at the end looking at the three BrainCheck abstract presentations that we did at AAIC and how they fit into that broader landscape.

[01:17- 03:10]
5 Themes Shaped AAIC 2026
Slide 2

The five themes that we’ll cover — stepping back from the individual presentations, posters, conversations held, I felt that these five themes kept coming up — First being: Patient Targeting.

We have more diagnostic tools and therapeutic options than ever before, which is a really exciting time in this space. But we still have a fundamental problem of identification of the right patients — early enough, and at scale.

The second being: Blood Biomarkers. Science here continues to move extraordinarily quickly, and the conversation is shifting from “Do these tests work?” to “How do we actually deploy them and interpret them?”

Then third: Going Beyond Diagnosis. There was much more attention this year about what happens after a patient or an individual is identified with cognitive concern.

Focus on care pathways, what successful longitudinal management looks like, caregiver support (and the importance of that), and holistic outcomes for both patients and caregivers.

Fourth: Digital Health, obviously relevant to BrainCheck. Digital assessments, AI remote monitoring, and the expanding capabilities there.

And clinical decision support and how they're becoming increasingly integrated into the broader clinical conversation.

Finally: Prevention and Early Intervention —which has been a theme for a very long time, but the field continues to move earlier and earlier in the disease course, including towards identification of risk before any sort of impairment develops.

What ties all of these themes together is a shift from scientific possibility towards what clinical implementation is going to look like at scale.

[03:14 - 06:16]
Patient Targeting Has Become the Next Bottleneck
Slide 3

Starting with probably one of the biggest overarching takeaways — being Patient Targeting — and how that's become the next significant bottleneck.

For decades, really one of the leading and dominant questions in Alzheimer's disease was: “How do we diagnose it accurately?”

With sophisticated answers to that question, with advanced imaging, CSF biomarkers, blood-based biomarkers, and disease-modifying therapies, these tools have highlighted the new bottleneck of who should receive them — and when.

So most older adults are not entering memory clinics and asking for neurocognitive workups. They're sitting in primary care practices, cardiology offices, health system populations, Medicare panels — the list goes on — often with very subtle symptoms that have yet to trigger a specialty referral. The scale of that problem is really enormous.

Data highlighted from the Davos Alzheimer's Collaborative currently suggests that approximately 75% of people globally living with dementia remain undiagnosed.

That bottleneck is sitting increasingly upstream.

There's a health system with hundreds of thousands (potentially even more) patients. How do we efficiently identify the subset that warrants cognitive evaluation? And then from there, who actually needs biomarker testing and who actually needs specialty referrals?

So actually, time to brag on BrainCheck. The solution that we've recently launched to help try and address them is our Population Health Analytics tool.

It's designed to analyze the existing structured EHR data that exists across the health system's population and then organize the patient records using research-derived patterns from research from you and Vanderbilt — or clinician review.

So the goal is not to diagnose MCI or Alzheimer's disease, but rather, from the data that already exists within the EHR, give clinical teams a scalable starting point: Who needs to be called first? And help surface and prioritize the patients that warrant REACH and cognitive evaluation, rather than exclusively waiting on a patient or a caregiver to raise their hand.

Really in practice, the goal is to create an upstream layer to that cognitive care pathway and use the data that they already have to help prioritize where to start.

That's a very different problem than simply building a more accurate diagnostic test.

We heard so much about this at AAIC about population health, risk stratification, brief cognitive assessments, and scalable workflows — with the question becoming less, “Can we detect and successfully diagnose Alzheimer's disease?” but shifting towards: “Can we build a system that reliably finds the people who need that detection at the right time?"

[06:20 - 09:27]
Blood Biomarkers Continue Their Rapid Evolution
Slide 4

Blood-based Biomarkers - and their very, very rapid evolution.

Can't talk about AAIC, really, for the past several years without talking about blood-based biomarkers.

The pace of progress here continues to be remarkable. In particular, p-tau217 continues to build evidence as a highly informative marker for Alzheimer's pathology.

Plasma biomarkers are moving beyond specialized research settings and much more towards a broader patient application.

We've seen a lot of that momentum continue excitingly past the conference. Just last month, there was the FDA clearance of C2N's PrecivityAD2 blood test for use in adults as young as 40 who have signs and symptoms of cognitive decline; it's an incredibly exciting announcement.

It's also an important milestone, not simply because we have another blood test available, but because it illustrates just how quickly this landscape is changing.

We're moving toward a world where clinicians may have substantially greater access to blood-based tools to help evaluate Alzheimer's disease much, much earlier in that diagnostic process.

Greater access to biomarkers also makes that upstream clinical pathway even more important and relevant.

Blood biomarkers are intended to tell us whether a patient has cognitive impairment in the first place. The evaluation helps establish that clinical context and determine whether further workup is appropriate.

Again, the question becomes who should actually receive one. That brings us back to that patient targeting challenge.

We saw evidence at AAIC of how significant that impact could be once those tools are implemented in routine care, particularly in a primary care setting.

A real-world study that I found to be the most compelling from Lund University included about 1,300 patients of family physicians and had approximately a 65% diagnostic accuracy before reviewing blood biomarker information. Then, post-reviewing the blood test results, that accuracy increased to 93%, which is nearly identical to the specialist accuracy at 94%, which could be absolutely transformative for primary care.

The takeaway isn't that we should simply start taking Alzheimer's blood tests and performing them for every single adult. It reinforces the need for structured diagnostic pathways, identification of patients with cognitive concern, and then how to objectively assess cognition and determine who warrants that further etiological evaluation.

I really think the biggest shift in the biomarker conversation from AAIC was less about proof, as it has been in former years, and more about deployment in a primary care setting because we know that these biomarkers can perform extremely well. But figuring out how to use them efficiently in a real-world clinical practice is a huge challenge.

[09:30 - 11:04]
Diagnosis Alone Is Not Enough
Slide 5

Moving on to our third theme, which is the increasing recognition that diagnosis alone is not the endpoint.

Historically, so much of Alzheimer's research has understandably been on diagnostic accuracy: sensitivity/specificity of digital cognitive assessments, Biomarkers, imaging, and classifications — and while those things remain critically important, a lot more of this year's discussion was about what happened after we identify somebody:

How do we manage that patient longitudinally? 
How do we support the caregiver?
How do we address safety function and all of the neuropsychiatric comorbidities, advanced care planning, and modifiable risk factors?

Even with the fantastic advancements in biomarkers, implementation really isn't simply a matter of making tests available.

Going back to some of the information presented from Davos and their implementation experience of about 3,000 blood-based biomarkers across 130 clinics: the lessons extended well beyond the assay application itself into clinician education, the EHR integration, result disclosure workflows, and having a standard of care for that conversation and coordination across these, and how much that all matters with great impact.

We're seeing the field really mature from “How accurately can we diagnose disease?” to “How can we create this effective care pathway?”

[11:08 - 12:42]
Digital Health Continues to Gain Momentum
Slide 6

Theme four: Digital Health and its gaining momentum alongside blood biomarkers.

Digital assessments, remote monitoring — particularly in a research setting — the integration of AI, and clinical decision-making support were really everywhere and broadly discussed at AAIC this year.

And there's an important distinction in how these technologies are being discussed right now.

A few years ago, digital cognitive assessment was often discussed primarily as how it's in the alternative and a more scalable solution to a paper test, and what we're seeing now is a broader role for digital tools within the clinical pathway itself.

In addition to having more and more digital tools available, there were a lot of groups presenting data that suggested brief digital tools helping identify patients that need to move forward and establish medical necessity for biomarker evaluation is where the focus needs to be.

It's important because cognitive testing and biomarkers answer two very different questions.
Biomarkers tell us whether AD pathology is present, but the cognitive assessment tells us whether, and how, that pathology is manifesting itself quickly — and neither can really exist in isolation.

So the opportunity to connect population identification, cognitive assessments, biomarkers, interpretation, and then the holistic longitudinal management into a coherent workflow was, again, tying the themes together and really where a lot of the discussion was.

[12:48 - 14:37]
Prevention and Early Detection Continue to Expand
Slide 7

Finally, theme five: Prevention and Early Detection.

We have continued movement towards the earlier stages of the disease, and as I think many know, that's where the conversation has existed for many years.

But there's significant attention to lifestyle intervention, the modifiable risk factors and the role that they play (and to what extent), personalized prevention, and identifying individuals early in the disease trajectory is where a lot of the science and plenary sessions were focused.

One particular notable announcement was the 100 million dollar funded Protect-Cog trial.

This study is incredibly exciting. It's going to study and evaluate whether combining multi-domain lifestyle interventions with metabolic targeting therapy (the GLP-1s) can further reduce cognitive decline, MCI, and dementia.

Regardless of what that study's eventual result will be, the study itself is going to tell us something important about where the field is heading: no longer thinking about Alzheimer's care as exclusively something that begins once dementia is clinically obvious.

Again, moving upstream to risk prevention, the earliest signs of cognitive change and potential interventions that combine both pharmacologic and non-pharmacologic approaches.

Again, making early identification even more important, with interventions most valuable earliest in the disease course. Then the need for health systems to be capable of identifying patients before they arrive at the setting

Prevention and detection aren't really separate conversations anymore. They're becoming much more increasingly dependent on one another.

[14:39 -  15:01]
Research Spotlight
Slide 8

All right, so research spotlight on the BrainCheck studies presented at AAIC this year. Shifting gears, we're going to spend a couple of minutes going over the research that our own team presented at AAIC.

Rather than looking at these as three isolated BrainCheck studies, I think they've mapped quite nicely into the themes that we kind of just reviewed.

[15:03 - 16:58]
Validating a Digital Cognitive Assessment Against the MoCA
Slide 9

Our first abstract examined the relationship between BrainCheck Assess™ and the Montreal Cognitive Assessment, or MoCA.

In this analysis, we looked at 176 older adults across multiple clinical sites. The observed correlation was 0.80 between BrainCheck and MoCA scores.

We've presented on our correlation between the BrainCheck Assess™ and MoCA before, but it really demonstrates the strong relationship between the two measures.

Although there will always be an intrinsic difference between the two, as BrainCheck was not created to digitize the paper version of the MoCA, but other neuropsychiatric tests that are standard of care in a Neuropsych setting and include more reliance on our normative database and reaction time — things that digital assessments are able to capture.

Having the strong discrimination scores across normal cognition, MCI, and dementia, and then developing a score mapping to really allow the translation between the 2 instruments is really important for primary care.

So the clinicians have something that is paper-based that they can compare to, that they already know, so that when we introduce the digital assessment, clinicians can understand:“How does this relate?” and “What are the cutoff scores used for these different diagnostic categories?”

This hopefully will help answer that question.

Within the broader AAIC story, it really reinforces the idea that scalable cognitive assessments don't necessarily require sacrificing clinical validity.

If we're gonna move cognitive evaluation upstream and reach substantially larger populations, having digital tools to enable that is going to become increasingly important.

[17:02 - 19:06]
Longitudinal Outcomes Following Digital Cognitive Care Planning
Slide 10

The second presentation was on the longitudinal outcomes of our BrainCheck cognitive care plan.

It focuses on the move beyond identification to ask a different question, and that's: “What happens to patients and their caregivers when they're entering the cognitive care pathway?”

In this research, we included about 2,500 patients over the age of 50 across about 65 practices between 2022 and 2025 and looked at the longitudinal outcomes associated with having the availability of a cognitive care plan and their engagement with that care plan.

Across the population, we observed improvements in several clinically meaningful areas.

Anxiety, measured using the GAD-7, decreased significantly over time.

Depression symptoms improved among patients and caregivers.

Activities of daily living and functioning (we used the Katz Index to inform that) also improved.

For us, the study connects directly back to diagnosis alone not being enough. Identifying cognitive impairment is important, but the ultimate goal isn't to generate a score and attach the subsequent diagnosis.

It's to care for the patient and their caregiver.

How do we identify needs? How do we support caregivers successfully and address behavioral health, safety, functioning, and care planning? And can we successfully do those things and build on that over time?

This real-world data set suggests that a structured longitudinal care pathway really can produce meaningful outcomes across very different clinical environments.

The 65 sites that were included here were for primary care settings, geriatrics, neuropsych, Neurology — and the meaningful results were apparent kind of broadly across these different clinical settings.

Hopefully that's increasingly where this field is headed; away from isolated cognitive testing and towards continuous cognitive care.

[19:06 - 21:02]
Translating a Digital Assessment for European Populations
Slide 11

The last presentation focuses on another major challenge in cognitive assessment, and that's generalizability and accessibility of testing.

This work was conducted through the Predictum Consortium and included 416 participants across 5 European countries.

Through our participation in this study, BrainCheck Assess™ was translated and then culturally adapted into 5 languages (more to come).

The 5 included in this presentation: Norwegian, Dutch, Spanish, German, and French.

I think what is important to understand is that these weren't simply direct word-for-word translations. We had native speakers review the assessments for linguistic and cultural accuracy and had a whole panel of reviewers to confirm the translations and the nuance that existed intrinsically between the different populations and how the user interacted with BrainCheck Assess™.

Across these populations and devices, we saw a 98.3% completion rate, which is really important because one of the challenges of scaling cognitive assessments is that the population we need to reach is increasingly heterogeneous; language, culture, education, intact familiarity, sensory limitations, and the setting of testing all influence whether the results of an assessment are actually usable and reliable.

Demonstrating that BrainTrack Assess™ can be successfully adapted and completed across countries, languages, and multiple devices is an important piece of research and component of scalability and the necessity of that.

Connecting back to our broader themes, being serious about early identification at the population level, our tools have to work for the entire population and not simply highly selective for research participants.

[21:04- 21:47]
Conclusion
Slide 12

All right, so bringing everything together: I left AAIC this year feeling that we're entering really a different and truly exciting, innovative stage of Alzheimer's and dementia care with increasingly powerful biomarkers and digital tools, disease-modifying therapies, learning more about prevention and modifiable risk, and really the ways to potentially reach populations that traditionally specialty center models haven't been able to access.

The next breakthrough might not be a single test or technology, but rather how to connect all of these advances into clinical pathways that work — and that's really where we're focusing a lot of our work at BrainCheck.