In this podcast episode, Alzheimer’s disease experts Anton P. Porsteinsson, MD, and Vijay K. Ramanan, MD, PhD, discuss the most recent results presented at CTAD about AD treatments currently available as well as in development.
In this episode of “Cognitive Conversations,” expert faculty discuss breaking data presentations from the 2026 Clinical Trials on Alzheimer’s Disease conference held in San Diego, California, from December 1-4, 2025, and their implications for clinical practice.
Topics covered include:
This podcast episode is part of the CME/CE-accredited program, “Early Clues, Lasting Impact: Detecting and Treating Alzheimer’s Disease in Its Earliest Stages,” supported by an educational grant from Novo Nordisk.
Faculty:
Anton P. Porsteinsson, MD
William B. and Sheila Konar Professor of Psychiatry, Neurology, and Neuroscience, and Medicine
Director, Alzheimer's Disease Care, Research and Education Program (AD-CARE)
University of Rochester School of Medicine and Dentistry
Rochester, New York
Vijay K. Ramanan, MD, PhD
Consultant and Associate Professor
Director, Alzheimer’s Disease Treatment Clinic
Department of Neurology
Mayo Clinic
Rochester, Minnesota
Link to obtain CME/CE credit:
https://bit.ly/4999otE
Dr. Porsteinsson: Hi, everyone. This is Dr. Anton Porsteinsson, and I'm here with my colleague Vijay Ramanan from the Mayo Clinic in Rochester, New York.
Hi, Vijay. How are you?
Dr. Ramanan: Anton, very good to see you. Lots of good things for us to talk about in this space, which is a good problem to have.
Dr. Porsteinsson: And we're going to talk about some of the presentations at the just recently completed Clinical Trials in Alzheimer's Disease meeting that was held in San Diego, December 1st to 4th. Vijay, a lot went on in that meeting, and I think, though, it's safe to say that most of us were most excited. The top of our list was the results of the evoke and evoke+ studies.
For those that are not familiar with those studies, these were sponsored by Novo Nordisk and looked at the GLP-1 agonist semaglutide in Alzheimer's disease. These were two large studies, over 1,800 participants in each study, all of whom had early symptomatic Alzheimer's disease that were biomarker verified. And then they were randomized to either be on placebo or semaglutide in the oral version for two years for the primary outcome measure, and then actually continue in a placebo-controlled fashion for another year.
And fair to say that these studies were large, so we have a lot of power. They were well-designed. They were well-executed. And for me personally, the data was very disappointing, as there was no impact on any measure of Alzheimer's disease. Using the CDR sum of boxes that looks at both cognition and function as a global scale, cognitive outcomes, functional outcomes, yet there was actually a positive impact on biomarkers. But that effect was small, about 10% or so.
We know that people were consistent with taking their GLP-1 agonist because we saw weight loss, we saw improvement in glucose control, but no impact on the course of Alzheimer's disease. I found this pretty disappointing. What's your take on this, Vijay?
Dr. Ramanan: I couldn't say it any better, Anton. I think, as you mentioned, there's so much going on in the field right now, and there's a lot happening at CTAD this year, but I think the results of this work were really first on the agenda for most in the field. And peeling back the curtain, of course, we need all the good approaches we can get our hands on to help manage Alzheimer's and related diseases.
And one big-picture approach is thinking about existing drugs that may be utilized for primarily other clinical indications, but essentially being repurposed to see do they have an impact on Alzheimer's disease in this case. And a lot of interest in these GLP-1 drugs, they're, of course, being very commonly used throughout the country, throughout the world, primarily for weight loss purposes. But a lot of hypotheses coming from observational and retrospective work asking the question, could there be an impact on neurodegenerative diseases?
And heard a number of colleagues make the argument, could there be potentially anti‑inflammatory effects of these medications that are beneficial in the nervous system? Some active debate about perhaps the results of some of the retrospective work showing patients who were, say, on these medications having a lower frequency of dementia diagnosis, maybe the possible explanation is better indicators of broader vascular risk factors or picking out in those patients who had a lower risk of developing dementia again retrospectively, were there other broader positive health indicators contributing to that signal? And I think as you mentioned very well, we really need high-quality, large, late-phase clinical trials to directly answer these questions and here was the opportunity. And of course, disappointing for all of us in the field, for patients and for care partners, as I think there was some hope, could there be a bit of a positive signal here, a benefit to clinical status, cognitive or functional decline, perhaps slowing, and unfortunately that's not the case, that really these were negative studies.
And as you mentioned, peeling back the curtains and understanding why that is will, of course, be part of the next phase in the field, digging into this data further. Those positive indicators on biomarkers, modest effects, and, of course, not correlated with a clinical benefit in this study, is this related to the timing of use of the medication, the duration of the trial? Pretty standard for most trials, I think, in our field, that couple of years' duration, but might things be different if there was use of these drugs at a different time point in the disease?
These are all questions I think that are out there, but for a very high-quality trial, unfortunately resulting negative, not giving us an indication to really proceed forward at least with any expedition with these medications.
It does raise the question for me, though. I'd love to hear your insights. Do you think this is the end of the story for these GLP-1 medications, or are there other things, other populations, subgroups, different timing of use, are those things that are maybe still out there for the field to answer, or is this kind of the end of that direction?
Dr. Porsteinsson: And that's a great point Vijay. You mentioned inflammation, and in fact, in the group that was on the semaglutide, there was a 30% reduction, mean reduction in high sensitivity CRP. So it had an impact. I think what I take from this is that we have to be cautious about epidemiological evidence that something works in a large population, and we have to do what, like you said, these large well-considered randomized trials.
Now we do know that we're going to see much more details at the so-called ADPD conference, which will be in Copenhagen in March. So, Novo has promised that they will give us a deep dive into the data at that point. But I have to respect the lead physician at Novo who said that these are convincingly negative data and that if there is a subgroup that benefits from the GLP-1 agonist, it's a small one that has a meaningful benefit; otherwise, we would see some daylight between the drug and placebo, and there really wasn't any separation.
To me, though, one of the reports at the meeting was on proteomic signatures in different Alzheimer's populations and the APOE4 homozygous, that is the group that has two copies of the APOE4 gene, that is at a much higher risk of developing Alzheimer's disease and actually has some proteomic changes quite early in life.
One of the proteomic changes was in GLP-1. So when we see the more detailed data, that is where my eye is going to go immediately. Was there a difference in terms of APOE carriers, particularly APOE4 homozygous?
So, I don't think that the story is completely done. Another point that I would like to make is that semaglutide does cross to some degree into the CNS, but there are other GLP-1 agonists in development that have a high CNS penetration.
So, that may potentially matter, but we'll see as we get there. Another major symposium was on lecanemab, the subcutaneous version. And lecanemab, the subcutaneous version, is already approved for maintenance treatment. So after somebody has been on lecanemab intravenously for 18 months, they can use this as a maintenance treatment versus continuing on less frequent IV infusion.
What do you think about this for maintenance? What are you doing in your clinic with that?
Dr. Ramanan: Yeah, gosh, great questions. I think we could chat all day on this topic. Peeling back the curtains a bit, there's still some debate in the field about what's the most optimal approach with these anti-amyloid therapies.
We know, for example, in the donanemab trials and in its drug label, its appropriate use recommendations, really the philosophy there is treat to clear, get folks to an amyloid-negative status, and then would stop the therapy. With lecanemab, there is this option of continuing on some level of maintenance dosing even after an 18-month period, after an individual might be at an amyloid-negative status in some of those cases. And I think we're going to continue to learn as a field what really is the value, the costs of that type of approach.
I think the discussion is so far centered on might there be some action of lecanemab, because it does have a relatively preferential impact on some of the soluble species of amyloid. Could that affect a component of the disease that may contribute to ongoing progression even if plaques are cleared? Could that potentially also minimize the risk of reaccumulation of amyloid plaques in parallel with this?
I think we've learned from some of the initial data from both lecanemab and donanemab that reaccumulation after therapy is stopped, seems to occur relatively slowly, generally in line with what we know about the disease even absent treatment. But that's where this drug option, I think, sits in the field. An area of active debate, and we're going to continue to learn.
In our practice, we have offered maintenance therapy with the IV form thus far for patients who are at that time point. They've been on the IV form for 18 months. Most of our patients who have gotten to that decision and had that option, I found, have tended to choose to continue on maintenance IV. In that case, it's a once-a-month IV infusion.
I think part of that reflects that for most patients with delivery of the drug and the process and the nuts and bolts of administration and monitoring being tidied up, that the treatment can fit into their broader life, their broader medical management, and actually the opportunity to de‑escalate in terms of the frequency of getting the infusions at that point usually can fit into plans.
I think for some patients, maybe because their general philosophy or their goals of care, they're putting these pieces together, as we discussed, through shared decision-making. They feel these drugs are primarily designed to clear my amyloid plaques from the brain. If I've gotten to that point, maybe, for me, the best choice is to stop and observe and reconsider my options down the road after some monitoring.
I think in a few other cases, particularly when the frequency of returning to the clinic, say, for the infusions, we still think it's important in these patients to have some safety monitoring, not at the cadence that, say, is at the start of therapy, particularly in the first 6-12 months, but still some safety monitoring with MRI scans and clinical follow-up and touchpoints through the clinic.
Sometimes for patients, those things, maybe if social supports aren't quite as fulsome as one would hope, can be a little bit more challenging. You tease through the options. Are there ways that we as a healthcare team can help support some of those nuts and bolts?
Dr. Porsteinsson: You must have, actually, a very nice infusion center at the Mayo Clinic in Rochester, Minnesota, because I think that some patients really create quite a bit of connection with the infusion staff and don't want to see that necessarily go away.
Dr. Ramanan: I think that's absolutely right. Same experience that we've had, yes.
Dr. Porsteinsson: Now, one of the big pending changes with subcutaneous lecanemab is that on November 25th, ASI completed a rolling submission under what is called a fast-track status to the FDA. And this is for, actually, initiation of treatment with subcutaneous lecanemab. The FDA hasn't set what is called a PDUFA date, their decision date, and those seem to have been pretty flexible to the FDA of late. But traditionally, it should be within six months, and actually, the maintenance approval for the subcutaneous formulation was done in about three months. So, we don't know when this is going to happen exactly, but I expect it will be in the first half of 2026.
Now, I hear different feelings about this. ASI reported that satisfaction scores for participants on the subcutaneous injections were very high, and actually, both participants and providers were pretty satisfied, but I also know that us, here in Rochester, New York, and other clinicians are a little bit worried about those first 3-6 months when the ARIA risk is the highest. Do we somehow have a disconnect when people are self-injecting at home?
But how do you see this? What do you think about the approval for initiation of treatment, that there won't be any IV component that we start with subcutaneous formulation?
Dr. Ramanan: Similar thoughts as you, Anton, and I think if this is approved for initiation, it will fall into a similar philosophy as with maintenance. In other words, I think it will be an option, and some patients will choose that option. It's not hard to see the advantages of it, the increased flexibility that they might have, not necessarily having to tie their schedule around returning to an infusion center at every two-week or every four-week cadence from a maintenance standpoint.
But I could imagine, as you mentioned earlier, for some patients, many at our center, they feel that actually the burden of visits is not an issue, that returning to the infusion center can actually be a positive element of their care. And as we know with other disease settings, some patients and care partners are very comfortable using injections at home, such as the auto‑injection process that would be here for lecanemab. Others actually prefer, in a way, the convenience of going to an infusion center and getting the medication.
So I think not every patient, if given the option to initiate with subcutaneous, would choose that versus intravenous. I think there'd be a mix. And I think the most important thing that you mentioned is regardless of the delivery of the drug, having a robust mechanism to maintain the connection so that we are catching potential episodes of ARIA early and managing them appropriately, that's really going to be critical regardless of the delivery of the drug method.
Dr. Porsteinsson: So let me change direction here for a little bit. In the CTAD meeting, as well as multiple meetings, there have been numerous presentations on biomarkers, imaging biomarkers, cerebralspinal fluid biomarkers, but particularly blood biomarkers. That has been one of the hottest areas within the Alzheimer's disease field in the last few years.
And I did see that you were co-author on a presentation that really has caused some reverberations. And to some degree, I felt that the information on the blood biomarkers has been very consensus-based. We get yearly, kind of a new publication on the consensus of an expert group, how they should be utilized, and what they bring to the table.
Here, we actually saw a number of different parties present data on real-world use in different settings, in expert settings, in primary care settings. But also, what was intriguing is the different versions of, for example, the p-tau-217 test done in different labs and in different combinations or ratios.
Tell me a little more about this. This is getting a lot of press.
Dr. Ramanan: This is important. I completely agree. And as you mentioned, it's getting a lot of attention, and it's going to continue to rapidly evolve.
It's one of the most exciting but also challenging areas of our field. And maybe taking a step back, A, I think it's wonderful that we have biomarkers that we can integrate to our clinical evaluations. They're important for diagnosis.
If I'm considering, does this individual have Alzheimer's disease as an evidence point to help support that diagnosis, biomarkers can play a role. From the rule-out perspective, if I think it's unlikely that this individual has Alzheimer's disease but it's within the possibilities. I think a negative biomarker result can again help to support a diagnosis potentially of an alternate neurodegenerative etiology or other etiology. So, biomarkers are very important from that standpoint.
As we hinted at earlier, they're also critical in thinking about eligibility for disease-modifying therapies. If you're going target amyloid plaques in an individual, you really have to know, does that patient have an abnormal burden of amyloid plaques in the brain? So, we have right now, clinically available, different ways to get that information.
In general, during life, it's widely considered that the gold standard for identifying amyloid plaques in the brain would be through an amyloid PET scan. Even understanding that that is the current gold standard in vivo, amyloid PET by itself is not perfect. In fact, no biomarker is. They're giving you a readout of something that's happening in the nervous system.
There are cases, for example, of genetic Alzheimer's disease where based on the structure of the amyloid fibrils within the plaques, based on potentially the tracer that's being used, you could get a falsely negative amyloid PET scan, even though at autopsy, you know that those amyloid plaques are there. So, that's not perfect, but it's still the gold standard.
We also have fluid biomarkers. We have a lot of experience in the field with CSF biomarkers, which are at that +90% in terms of sensitivity and specificity and predictive values for guiding, is this individual likely to be amyloid PET positive or negative? We've learned a lot from those CSF biomarkers, including, in that setting, likely related to influence of comorbidities as well as fluid dynamics, that ratios, for example, a ratio of phosphorylated tau to amyloid do better than isolated analytes within the CSF, where sometimes you can't control for those other broader factors.
And now we're starting to add that depth of literature and experience and data on plasma biomarkers. I think one of the most important things that you hinted at is that just because we have blood-based biomarkers for Alzheimer's disease, not every blood-based biomarker is equivalent. There are different assays. They may have differences that are meaningful in terms of the cut points that are utilized to guide whether someone would be abnormal or positive versus negative. Those things matter. And where you provide that data from, which impacts the cut points, impacts the downstream interpretation – are these studies coming from population-based samples? Are they coming from clinical trial samples? Are they coming from clinical convenience samples? – all of those things impact the situation. And that's before you ever get to the possibility of the challenges with acquisition, with collection, with processing and shipping, with reagents that are included.
And so, I might mention, as you highlighted, I'm a co-author on one of the presentations in this space. Dr. Alicia Algeciras-Schimnich from Mayo Clinic was the lead. And this really looked at the now FDA-approved p-tau-217 to A-beta-42 ratio in the plasma and compared it to, for example, just looking at an isolated plasma p-tau-217. And in that particular sample, perhaps due to issues with the stability of the A-beta-42 protein, perhaps due to other factors, these debates continue.
And we really saw that these nuts and bolts matter. Where cut points are really matters, else you might be utilizing a test that for all other purposes is a high-quality test. But because of the small variations that we're dealing with can impact whether someone's classified as amyloid-positive or amyloid-negative. And so I think the data collectively showing all of these pieces really matter. Analyte, cohort, interpretation, cut points, they're really important for how we deploy these tests in real-life clinical practice.
Dr. Porsteinsson: So complex isn't always better. And I think that when we see these ratio tests, maybe one of the other critical parts is that a lab doesn't just give us the ratio, but gives us the individual values for the components that go into calculating the ratio. So here, for example, in the test that you mentioned, it's a p-tau-217 over A-beta-42 ratio.
And if the A-beta-42 is so kind of sensitive to collection and processing and storage and all of that, that it might influence the ratio, we have to be able to see the individual values as well. So more to come here. And, you know, actually, you and I will talk about this in more detail come January in a webinar. So, let me shamelessly kind of highlight that for the audience.
Now, in the time that we have left, I wanted to hit a few more highlights that I thought were interesting. So there are really great expectations of a humanized monoclonal antibody targeting a beta amyloid called trontinemab.
Vijay, you tell us, how is trontinemab different than both lecanemab and donanemab?
Dr. Ramanan: Great point, Anton. So, trontinemab really involves an essentially repurposed drug that may be familiar to the audience, gantenerumab. And the key difference here is in how the drug is delivered to the central nervous system.
It relies on some technology known as Brainshuttle technology. And what this is trying to get at is when we are dealing with a blood-brain barrier, pending that drug of interest to an antibody that can go bind a receptor on that blood-brain barrier allows that package to essentially act as a shuttle. And when that binding happens, this package is then able to cross that barrier. And then the shuttle can deposit its intended package, in this case, the medication being trontinemab, an anti-amyloid antibody, and therefore get potentially higher penetrance into key central nervous system tissue that we want to while also potentially allowing for lower systemic dosing being needed in order to get that target of CNS penetrance.
And some of the early data on trontinemab is very exciting for the field, both in showing a high potency of amyloid clearance. In fact, within six months, a very, very large proportion of the sample getting to an amyloid-negative status, but also potentially with lower rates of ARIA or other side effects related to these therapies, possibly related to that balance of being able to use a lower dose administration systemically while still getting very high potency penetration into the nervous system.
So really the bigger point is, as we deploy these anti-amyloid therapies into clinical practice, there continues to be development on the next generation of these therapies where maybe we are changing how we administer these drugs and their underlying structure in order to achieve that potency of amyloid clearance and optimize the potential for side effects. So there'll be more to come about this drug, including in additional clinical trials, even including testing these medications in individuals who are amyloid-positive but currently don't have any cognitive symptoms, therefore, as a sort of prevention study approach.
Dr. Porsteinsson: And so we're going to have the Phase III TRONTIER 1 and 2 studies for people with early symptomatic Alzheimer's disease, that's MCI due to Alzheimer's disease and very mild Alzheimer's disease. And next year, I think we will see the PREVENTRON study, which will look at people, like you said, that are preclinical.
The Brainshuttle technology increases the percentage that goes from plasma to the CNS space tenfold, so from about 0.3% to about 3%. But interestingly as well, it's distributed more evenly in the brain tissue, so this may definitely be an advantage over the currently available agents. And there will be other Brainshuttle connected antibodies coming, so we'll be keeping an eye on this.
But talking about antibodies, there were some disappointing news about tau antibodies that had emerged just before the CTAD meeting. Johnson & Johnson highlighted that their mid-domain antibody failed in Phase II, and that comes on the heels of multiple shots at so-called N‑terminus antibodies, such as semorinemab and posdinemab, that failed. And actually, another mid-domain antibody called bepranemab had some unclear data a year ago, and we saw a little bit more details on that.
Now, there are tau antibodies that remain, and other methods of tau targeting. Obviously, tau, and particularly pathological tau, is incredibly complex. If you think that amyloid isn't complex, tau has even more complexity, and the toxic isotypes are very dynamic, and they're cut in various spots, and they're hard to target, hard to identify. The most excitement, though, is about so-called MTBR, or microtubule-binding region antibodies, and there are three of them that are currently in clinical trials that we're waiting for results on.
And then there's the potential for C-terminus and tau antibodies as well. But in addition to the tau antibodies, there are also genetic approaches, so basically what we call ASOs, or antisense oligonucleotides, RNAi, or interfering RNA, and even viral vectors with payloads targeting the messenger RNA for tau.
What do you think? This is getting more intriguing and complex by the minute. Intrathecal injections versus viral vectors, there's a lot happening in Alzheimer's disease, and the treatment isn't getting any simpler.
Dr. Ramanan: I couldn't agree more, and I think the bigger point that you're highlighting, Anton, is that yes, a lot of the field's energy has focused on anti-amyloid therapies, the late‑phase trials of recent years, early implementation in clinical practice, but it's a diverse array of research approaches, absolutely including those targeting tau, despite some of the disappointing results so far. We know that the challenge in targeting tau is different than the challenge in targeting amyloid, and there's still excitement and enthusiasm about the possibility of some of those agents that you mentioned targeting different aspects of that MTBR region, potentially the antisense oligonucleotide approach.
So, definitely more to come, and I think all of this is leading toward a future where we have the ability, I hope, to provide combination therapies on an individualized basis, similar to how we approach other neurologic and systemic medical conditions.
Dr. Porsteinsson: And multiple studies targeting neuroinflammation as well. I won't go into that today, but I was hoping to kind of wrap up by asking you about an intriguing finding with multiple vaccines. You know, vaccines have come under a lot of scrutiny lately, but here we have pretty solid evidence for certain vaccines, such as the shingles or herpes zoster vaccine, as well as the BCG vaccine that is used worldwide as a vaccination against tuberculosis, and that those that get vaccinated may have a lower risk of developing Alzheimer's disease.
So, you know, what goes here?
Dr. Ramanan: It's a great point, and I think you mentioned earlier in our discussion, always worth keeping a grain of salt in mind when we're looking at epidemiologic studies, because those studies are typically not designed to conclusively assess and rule out for any confounders that might alternatively explain the findings.
But particularly, you mentioned the herpes zoster study that was mentioned at CTAD and got a lot of attention over the days afterwards. This was a very well-designed epidemiologic study, including, through some external factors, having the ability to look back at individuals who really seemed to be pretty well-matched over a long period of time, and due to availability and access of the vaccine in one cohort and unavailability in another, could really study what appeared to be a fairly well-matched sample and compare the two, and finding that administration of the vaccine was associated with lower incidence of dementia.
Obviously, we need more, just like we discussed earlier, and I think the plans for additional randomized-controlled trials to really put these ideas to the test are going to remain important, but encouraging and opening potentially some additional questions in the field that are worthwhile investigating.
Dr. Porsteinsson: And I think that what we come away with is really that what we need, both for the BCG vaccine as well as the shingles vaccine, is that we need large, well-designed, double-blind, randomized trials to see, does this really matter? And here is then potentially another good reason, besides the prevention against what the vaccines are protecting you against, shingles or tuberculosis, that are there additional benefits to that?
Well, this has been a power ride through some of the main results that we saw at CTAD. There were multiple others. We can't fit them in today, but hopefully we'll have a chance to chat about that at our webinar in January.
So, Vijay, as always, such a pleasure to talk to you and so informative, and you've got your finger on the pulse of basically what's the cutting-edge in Alzheimer's disease.
So, thank you. Any final words?
Dr. Ramanan: So wonderful to talk with you, Anton, as well, and thank you to our listeners. This is a rapidly-evolving space. You're hearing two folks who are really excited and enthusiastic about the future in this space, and we're lucky to get a chance to chat with you about our thoughts here.
Dr. Porsteinsson: Thanks, everyone. It's been great.