The Peter Attia Drive - #399 ‒ The evolution of Alzheimer's disease and dementia care： how early detection, personalized treatment, new therapies, and a multimodal approach are changing the landscape ｜ Gayatri Devi, M.D.

00:00:00.000 Hey, everyone. Welcome to the Drive podcast. I'm your host, Peter Atiyah. This podcast,

00:00:16.540 my website, and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:21.520 into something accessible for everyone. Our goal is to provide the best content in health and

00:00:26.720 wellness. And we've established a great team of analysts to make this happen. It is extremely

00:00:31.660 important to me to provide all of this content without relying on paid ads to do this. Our work

00:00:36.960 is made entirely possible by our members. And in return, we offer exclusive member only content

00:00:42.720 and benefits above and beyond what is available for free. If you want to take your knowledge of

00:00:47.940 this space to the next level, it's our goal to ensure members get back much more than the price

00:00:53.200 of a subscription. If you want to learn more about the benefits of our premium membership,

00:00:58.040 head over to peteratiamd.com forward slash subscribe. My guest this week is Dr. Gaia

00:01:06.980 Devi. Dr. Devi is a neurologist and psychiatrist specializing in memory disorders, cognitive

00:01:13.820 neurology, and women's brain health. She is the founder of New York Memory and Healthy

00:01:20.280 aging services and a clinical professor of neurology at Zucker Hofstra Northwell School

00:01:26.380 of Medicine. Her practice spans the full spectrum of Alzheimer's disease from asymptomatic high-risk

00:01:32.240 individuals through advanced dementia. And she is known for a personalized multimodal approach

00:01:37.560 that integrates biomarker guided diagnostics, anti-amyloid therapy, hormonal factors,

00:01:42.560 and lifestyle interventions. I've wanted to have Dr. Debbie on the podcast for a while now

00:01:47.260 because the field of dementia is changing rapidly. For a long time, Alzheimer's disease

00:01:51.920 was viewed as really kind of a one-way decline, diagnosis with very little room for nuanced

00:01:57.320 intervention. But she takes a very different approach, one that is highly individualized,

00:02:02.060 focused on early detection, careful risk stratification, and matching the right

00:02:05.660 combination of treatments to the right patient. In this episode, we talk about how to think

00:02:10.920 about dementia as a spectrum, including Alzheimer's disease, but also vascular dementia,

00:02:15.420 Lewy body dementia, and many other mixed presentations in between, as opposed to discrete

00:02:20.540 diseases. The evolving biology of Alzheimer's disease, including amyloid, tau, neuroinflammation,

00:02:27.040 and why pathology does not always map neatly onto symptoms. How she evaluates high-functioning

00:02:33.320 patients with very subtle cognitive changes, and how she thinks about biomarkers,

00:02:38.020 APOE4 risks specifically, and testing asymptomatic people. We talk about anti-amyloid therapies,

00:02:44.440 which have become very controversial, although slightly less so today than when they were

00:02:48.780 initially released, including lacanumab and decanumab, their risks and benefits and her

00:02:54.980 approach to reducing side effects, including the most devastating side effects which we talk about,

00:03:00.520 why some patients with Alzheimer's disease may stabilize or even improve with a personalized

00:03:05.160 multimodal treatment strategy, the overlap between Alzheimer's disease, vascular dementia,

00:03:11.320 and Lewy body dementia, including why Lewy body disease is often confused with Parkinson's

00:03:16.360 disease, the relationship between menopause, estrogen, and cognition, including her specific

00:03:23.300 concept of menopause-related cognitive impairment, and finally, how early detection, AI-assisted

00:03:30.080 monitoring, neuroinflammation, targeted therapies, and more personalized treatment may reshape the

00:03:35.980 future of our dementia care. So without further delay, please enjoy my conversation with Dr. Debbie.

00:03:49.020 Gaia, so great to be with you today. Very excited to talk about this. It's a topic that we've

00:03:56.040 discussed a number of times on the podcast, but unfortunately, there's always a lot to talk about

00:04:01.720 here. And I think you're a physician that, of course, is near and dear to our heart in the

00:04:09.200 practice. We co-manage a number of high-risk patients. And I've just always been impressed

00:04:15.060 by the way that you have taken a very personalized approach to managing this condition. But I just

00:04:20.480 want to let folks get to know you a little bit better. So you're both a neurologist and a

00:04:26.560 psychiatrist. Is that correct? That's correct. I mean, I'm boarded in multiple specialties,

00:04:31.320 neurology, psychiatry, brain injury medicine, pain medicine, and behavioral neurology. But

00:04:38.480 my subspecialty is really memory disorders, primarily from a neurological perspective.

00:04:44.060 But all these other areas allow me to better handle people with cognitive impairment.

00:04:50.060 I'm just kind of curious as you think about the evolution of your career when you started, which is obviously not that many years ago, you're still quite young, to where you are today. What has been kind of the biggest shift in your understanding of the dementing conditions that we're going to talk about today?

00:05:12.200 I think two things, Peter. One, that over the years of my practice, I've come to realize that pretty much all dementing conditions, and particularly Alzheimer's disease, can present variably depending on the person who has the condition and also depending on comorbidities.

00:05:39.140 So there really is a spectrum aspect to Alzheimer's and dementias, which people don't recognize.

00:05:46.740 They think of it as an all-or-nothing disease, and that's not true.

00:05:50.240 Much like autism, it really can span the spectrum from very mild impairment that doesn't get worse to severe impairment that kind of deteriorates rapidly.

00:06:01.540 That's number one, that dementias are really a spectrum disorder.

00:06:05.640 And number two, that what's really exciting to me is I feel like I'm in the field now almost as if I was an infectious disease doctor who is practicing medicine before and after penicillin was invented, you know, or discovered, sorry.

00:06:23.620 So basically that there are now so many different ways that we can treat dementias, we can personalize our treatment, we can give each person the right cocktail, if you will, of medications that would best benefit them.

00:06:41.980 And we now have drugs that can alter the course of the disease, which is absolutely, to my mind, fantastic.

00:06:50.080 We still have a ways to go, but I feel like we're on the threshold of something really grand.

00:06:56.660 Well, that's probably the thing I want to talk most about today, but I think there's some groundwork I'd love to lay to bring the listeners along with us before we get into both the treatments, but also the way in which you personally kind of manage some of the risks.

00:07:13.280 So we're going to talk about some of these risks and then the various treatments around them.

00:07:17.000 But before we do that, let's go back to kind of what you opened with, which is that we're not dealing with binary diseases.

00:07:23.420 It's not like you have vascular dementia or you don't, or you have Lewy body dementia or Alzheimer's disease or you don't.

00:07:30.200 Can you go a little deeper into this idea of what the spectrum of these diseases look like?

00:07:34.660 And maybe for the purpose of this first question, let's limit it just to Alzheimer's disease.

00:07:40.920 So maybe define Alzheimer's disease within the broader context of dementia.

00:07:45.500 I think there's probably still some people who use those terms synonymously, even though

00:07:48.840 they're not.

00:07:49.900 And then within Alzheimer's, again, walk us through the different subtypes, if you will,

00:07:55.160 and maybe how a clinician is able to arrive at that diagnosis.

00:08:00.860 Right.

00:08:01.480 So dementia is the umbrella term.

00:08:05.300 So a de-mentation of the brain.

00:08:09.100 and that means that there is loss of connectivity between the brain cells

00:08:16.580 that results in a loss of function, and it's a progressive condition.

00:08:22.080 That's dementia.

00:08:23.460 And there are several different types of dementia, Alzheimer's being the most common.

00:08:28.060 And what Alzheimer's is defined as is a loss of synaptic connections,

00:08:33.320 or synaptic connectivity that results in a loss of function that is driven by the presence

00:08:43.880 of extra neuronal amyloid plaques, intra neuronal or intracellular tangles, neurofibrillary

00:08:54.040 tangles, and increasingly what we now know to be inflammation in the supporting microglial

00:09:01.500 cells and other glial cells of the brain substrate that cause cognitive impairment that progresses.

00:09:09.880 So the question is, how do you make this diagnosis in a person? And, you know, that's really

00:09:16.800 difficult because you have to really have a high index of suspicion, particularly with people who

00:09:26.880 are very intelligent, people who are very functional, people who are able to compensate

00:09:32.440 very well, so that you can have someone who has moderate pathology and even severe pathology,

00:09:40.960 and they can still perform very well in many areas because they have a good brain reserve.

00:09:48.800 They have a resilience in their brain that prevents the kind of cognitive and functional

00:09:53.880 deterioration that someone who has less reserve and less brain resilience doesn't have. And there

00:10:01.480 are many different subtypes of Alzheimer's disease, depending on the part of the brain that's

00:10:07.080 affected. So you can have an Alzheimer's disease that primarily affects memory versus Alzheimer's

00:10:15.620 that can affect more the visual-spatial system. So it depends on the area of the brain where the

00:10:22.360 pathology tends to predominate. And it also depends on secondary conditions, comorbid brain

00:10:30.220 conditions that are present, such as Lewy body disease, TDP-43 pathology, agoraphilic brain

00:10:37.960 disease, all of which can inform how the person presents. So when you're looking at Alzheimer's

00:10:44.880 disease, you want to look at, first of all, you want to establish that the person has these

00:10:50.440 plaques and tangles. And that's something we can get into later because there are two groups

00:10:56.220 of thought on that. There are people who say, well, you can just have biological indices that

00:11:03.520 say that a person, you can diagnose Alzheimer's just based on biological biomarkers alone versus

00:11:10.940 another group that says, well, no, you need biomarkers plus clinical symptoms. You need

00:11:16.500 biomarkers plus functional deficit to make the diagnosis. But the biomarkers are really amyloid

00:11:23.140 tau and increasingly neuroinflammation. Okay. A few things I'd like to unpack there,

00:11:30.860 but maybe I actually want to start with a more basic question, which is we will get to obviously

00:11:38.140 talk about APOE, the genotype, which is probably one of the most common genetic predisposing factors,

00:11:45.160 although not the strongest. And maybe we can even spend a minute talking about the other

00:11:49.040 genetic conditions that are more dispositive in their prediction of a person developing

00:11:53.880 Alzheimer's disease. But pathophysiologically, what is the order of operations? Stated another

00:12:01.520 way, what starts this process? Does it start with the inflammation? Does it start with the

00:12:08.240 deposition of amyloid? Does it start with the tangles? Is it different in every person? Is

00:12:15.340 it different by location in the brain? Maybe talk us through that a little bit.

00:12:20.300 So generally speaking, it used to be thought that really amyloid was the first abnormality

00:12:26.640 that begins in the brain up to two to three decades even before the onset of clinical

00:12:34.520 symptoms. But now we're beginning to understand that changes in the inflammatory pathways may

00:12:41.860 predate amyloid deposition by quite some time. So it may be that first neuroimmunological changes,

00:12:51.500 then amyloid, and then eventually down the road tau, and that ultimately causes synaptic

00:13:00.500 dysfunction where the synapses, the connections between the nerve cells start to go awry and the

00:13:07.420 person then develops cognitive changes. And that process takes decades. So the whole point, I think,

00:13:14.640 with Alzheimer's and other dementias is if you can intervene early enough, even as early as your

00:13:22.080 30s into your 40s by incorporating anti-inflammatory methodology, anti-inflammatory

00:13:29.920 activities into your lifestyle, then perhaps you can prevent the whole cascade from being triggered.

00:13:38.020 On that topic, one of the things we've become much more aggressive in our practice around

00:13:44.140 is managing two viruses in particular, herpes simplex virus and varicella zoster.

00:13:51.020 Now, the data appear much, I think the data appear stronger for the management of varicella zoster through the use of a shingles vaccine in individuals who have been previously infected, which, of course, at this point by the time you're 50 is virtually everybody.

00:14:07.680 And so presumably the observations that we see that there's a significant reduction in the risk

00:14:15.620 of dementia and Alzheimer's disease in people who have been vaccinated for shingles and people who

00:14:22.100 have HSV treated suppressively versus those who don't. The thinking is that this works through

00:14:28.880 an inflammatory pathway. What can you say about any of that? Yeah, I really think there's something

00:14:34.980 to be said. I think neuroinflammation is going to be the new frontier in terms of how we're going

00:14:40.600 to deal with and even prevent entirely Alzheimer's disease. I have, for example, like two patients in

00:14:48.400 my practice where the sibling is riddled with Alzheimer's, but these two patients have severe

00:14:57.080 neuroimmunological disease that they've been treated with aggressively for decades, and neither

00:15:03.600 of them have a lick of Alzheimer's pathology. It's quite impressive. So I think there's something

00:15:09.740 to be said for, you know, this kind of treatment with an exposure to various viruses that have a

00:15:17.520 predilection for the brain and the central nervous system, like the herpes zoster virus, for example,

00:15:22.800 that causes shingles. Sorry, just to clarify the case you presented there, are you saying you have

00:15:28.300 three siblings that are genetically comparable? I mean, meaning like they would all have the same

00:15:32.860 ApoE phenotypes across them, they're 3-4s or 4-4s or whatever. One of them has disease. The two that

00:15:38.920 do not have a deficiency of their immune system that renders them less susceptible to mountain

00:15:46.300 immune response? So I have two pairs of siblings. Ah, okay. So in each of which one sibling has

00:15:54.780 significant Alzheimer's and the other sibling doesn't, which is particularly remarkable in one

00:16:02.180 case because that person has two copies of the APOE4 allele, which puts them at higher risk,

00:16:09.540 in fact, definite risk for developing Alzheimer's. And that person doesn't even have any amyloid,

00:16:15.460 not even a touch of amyloid in their 70s. So it gives you pause. It makes you think,

00:16:20.420 well, what is it about this person that protects them?

00:16:24.440 And are you saying that they have a known immunologic condition that is reducing inflammation,

00:16:30.120 or are you saying that's the hypothesis potentially?

00:16:33.300 That's the hypothesis.

00:16:34.540 The hypothesis is, and the fact that they've been treated

00:16:37.300 with various drugs that somehow modulate the immune system

00:16:43.060 may have had maybe factors here.

00:16:46.180 Again, it's speculative, but it's certainly worth thinking about.

00:16:50.000 Yeah.

00:16:50.480 The other thing that I would say that lends some credibility to this argument,

00:16:54.720 and again, unfortunately, we're largely stuck relying on epidemiologic data

00:16:59.920 as opposed to clinical trials, although there are trial data on some of the vaccines,

00:17:04.120 is the relationship between gingival health and dementia, which again points to this idea that

00:17:11.300 you have an area that is in very close proximity to the brain. And as anybody knows who's listened

00:17:18.300 to our podcasts on dental health, when the gums are inflamed, it wreaks havoc on the body through

00:17:24.500 an inflammatory pathway. And so the brain would obviously be highly susceptible to that. So

00:17:28.540 So again, I think all of this makes sense.

00:17:31.160 Yeah, I totally agree.

00:17:32.640 And to your point, even the entire gastrointestinal system, you know, the internal lining gets

00:17:37.640 replaced, what, every two weeks, the entire lining does.

00:17:41.420 So you've got to think what happens in that process and how that drives, you know, that

00:17:46.680 changes with different immunological conditions.

00:17:51.840 So talk to me about how you evaluate a patient.

00:17:54.860 Let's maybe take a step back and talk about your practice.

00:17:56.940 So are patients coming to you more commonly being referred by another physician? Are patients coming to you directly because they find out about your center and they say, look, I, you know, me or a loved one are concerned about some evolving symptoms. But basically, how are patients finding their way to you and what is your process of evaluation?

00:18:15.820 we have i've been in my practice on the upper east side for 26 years so over this time there's

00:18:24.980 a large a referral base of physicians who refer to me i also have kind of developed a niche in

00:18:32.700 terms of the evaluation and treatment of physicians and other professionals so they seek me out they

00:18:40.160 read an article in a journal or something along those lines. And then, of course, referrals through

00:18:46.100 families and friends. The one thing I will say about our practice is most of our patients

00:18:51.580 are highly functional. They're still working, many of them, and they want to stay working.

00:18:58.220 And my great joy in these years is that I've been able to help them achieve that goal,

00:19:06.360 to be able to stay functional and useful to the society and also have a sense of agency

00:19:12.900 for themselves. So the thing, the challenge with all these patients, especially the high

00:19:19.680 functioning ones, is that they are extraordinarily, their brains are extraordinarily smart at

00:19:28.420 compensating. So a lot of the testing that we do has to be fairly rigorous in order to uncover

00:19:35.680 the problems that such patients might have. So if you did like what people routinely know,

00:19:41.800 which is the mini mental status exam or the MOCA exam, many of these patients score a 30 or 29 or

00:19:48.000 28 well along into their condition, even 10, 15 years into their condition because they are so

00:19:55.900 adept. They have such great cognitive reserves. What do I mean by cognitive reserve? The ability

00:20:02.380 of the brain to maintain a large number of connections and to be resilient to disease

00:20:10.480 such as Alzheimer's.

00:20:12.680 So once we put them through, so first of all, we do a very meticulous general history, including

00:20:19.360 history about immunological conditions, past exposure to shingles, things like that, family

00:20:25.600 history, and then after that decide on the kind of testing we want to do, which often

00:20:31.400 involves cognitive testing, which takes several hours. We often will look at the electrophysiology

00:20:37.960 of the brain to see if there's any evidence of early slowing. Look at brain blood flow using

00:20:43.940 transcranial Doppler to see how much blood flows through different areas of the brain. We will do

00:20:49.380 specialized exams, imaging exams, such as MRIs, using special techniques so that we can evaluate

00:20:58.300 the different brain regions, such as the hippocampus, the parietal lobes, so we can

00:21:03.440 see if there's differential changes there, early changes. And oftentimes, especially here in New

00:21:09.700 York, we're able to do amyloid scans and people that we're worried about, and tau scans, which

00:21:17.220 measure brain levels of tau, which is the other abnormal protein that's associating Alzheimer's.

00:21:22.340 And sometimes we will also do what are called DAT scans, which are scans looking at dopamine levels in the brain, which also help us to differentiate if there's another process going on.

00:21:35.060 So it's a fairly rigorous series of tests that's tailored to each individual person.

00:21:40.720 And of course, laboratory testing, looking at APOA, APO little b, inflammatory markers,

00:21:48.220 and obviously the APOE genotype.

00:21:51.200 And certain patients where there's a history, family history of early onset Alzheimer's,

00:21:56.140 we do do early onset genetic screening to see if there's a particular gene that's abnormal.

00:22:02.560 And how routine is the use of a lumbar puncture? Is that something you would do in everybody or

00:22:08.800 has the robustness of the current blood-based biomarkers for AB 4042 and P-tau displaced a

00:22:17.120 lot of the need for lumbar puncture? So I am kind of, I think I'm old school

00:22:22.060 with the blood-based biomarkers. I definitely think they are the future, but I'm not sure

00:22:28.100 they are already the future. I feel that they are standardized against amyloid, for the most part,

00:22:39.020 amyloid scans. And I'm of the opinion that they would probably need to be standardized against

00:22:45.240 both amyloid and tau, because so many of us in our 70s and 80s have amyloid and will never

00:22:53.180 develop Alzheimer's. So I tend to, pretty much in any one that I'm concerned about with Alzheimer's,

00:23:01.760 I will do either tau and amyloid scan or a spinal tap. The advantage to the spinal tap is it's less

00:23:10.940 expensive, and you also are able to detect very early changes in tau, whereas the tau PET scans

00:23:19.320 traditionally don't really pick up early changes in tau. The disadvantage, of course, is no one

00:23:25.320 wants to get a spinal tap. Although, may I just say that all the women who've had epidurals to 1.00

00:23:31.820 have children have effectively had reverse spinal taps. You know, they've had anesthesia injected

00:23:37.220 into their spinal cavity. That workup sounds incredibly extensive. On average, since you're

00:23:44.300 managing this is patients who are outpatients. You can't do everything you just said on a day,

00:23:49.520 I'm imagining. So is that, for a motivated patient, is that a week, two weeks worth of

00:23:55.860 interventions? So we have patients come in from around the world or from different parts of the

00:24:03.200 country, and they're just here for testing and an evaluation. And we're able to pretty much

00:24:08.160 schedule everything over a two to three day period. Okay. That is an exhaustive list of

00:24:14.800 tests. I'd love to kind of walk through the contours of how you piece together results

00:24:22.220 and start to come up with what the archetypes are. So in other words, once you have all of

00:24:29.620 those tests back, how do you start to say, this is the pathology. So let's call it by disease.

00:24:37.060 So I think you're on an Alzheimer's pathway. I think you're this far along on the pathway. And I think it's of this variety. Does what you see in one snapshot give you a sense of velocity? Or can that only be determined over a time course where you repeat the evaluation a year later, look at the change and say, this is a slowly progressing versus a rapidly progressing condition?

00:25:01.480 And basically, how do you take that information and then create a treatment plan?

00:25:07.660 That's such a great question.

00:25:10.220 You have to marry the symptoms that the patient has, the time period over which the symptoms have occurred, how old the person is,

00:25:24.500 their comorbidities, whether it's high blood pressure, obesity, cholesterol, high cholesterol,

00:25:32.940 and other brain comorbidities. For example, how many strokes they've had, where these strokes

00:25:39.320 are located in the brain, which occurs in quite a large percentage of patients with Alzheimer's

00:25:45.600 as they have concomitant cerebrovascular disease. And then you look at the test results. How many

00:25:52.440 copies of the APOE4 allele do they have, or do they have a presenilin 1 mutation, which

00:25:58.360 is 100% associated with Alzheimer's, usually early onset Alzheimer's disease.

00:26:03.880 You also want to look at their cognitive scores to see, is this person has an overall intelligence

00:26:13.040 at the 99th percentile, but their language scores, their ability to come up with words

00:26:19.120 is at the 10th percentile.

00:26:21.020 This is somebody I'd worry about because it means that they're progressively losing their language skills and they're having trouble communicating. And let's say they're working in a job as a lawyer, then it may affect how well they're going to be able to function.

00:26:34.820 And that's one of the ways I assess in a single time point how much of a change has occurred.

00:26:44.080 So someone whose overall ability at the 99th percentile, visuospatial skills at the 10th percentile, language skills at the 10th percentile, I worry that this person's declined quite a bit, but they're compensating overall.

00:26:57.420 And maybe the following, the ensuing year, might be a year where they're losing quite a bit of function.

00:27:03.460 So this might be someone where you want to be more aggressive about treatment.

00:27:08.440 You have a patient who's having problems with language, and they have just in that area,

00:27:13.100 their blood flow is much reduced, or they have some slowing in their EEG, which is worrisome,

00:27:19.440 even though they look clinically fine.

00:27:21.700 All those things go into, all those different facets of their evaluation go into my final

00:27:28.820 determination of how much of each condition. Often it's very rare that the person just has

00:27:36.100 primary Alzheimer's alone. They'll usually have Alzheimer's plus. Alzheimer's plus a little bit

00:27:42.520 of stroke. Alzheimer's plus a little bit of hydrocephalus. Alzheimer's plus a little bit of

00:27:48.540 TDP-43 pathology. So you want to kind of put that all together and see how much each area is being

00:27:55.340 affected. Interestingly, if it's only memory that's being affected, that's actually a better

00:28:02.040 prognosis for patients than if it's primarily language that's being affected. Because we have

00:28:07.980 data to show that the trajectories, the rate of progression varies across depending on where the

00:28:15.560 pathology is. And then I'm not, I'm one of those people, I think, who is very, very, very averse

00:28:23.360 to staging Alzheimer's. I am not, people often will say to me, well, what stage is this person

00:28:30.280 in? And I usually say, listen, this person is at stage zero as far as their ability to communicate.

00:28:37.680 But maybe in memory, they're at a stage three or four. And maybe in language, they are in a

00:28:44.660 different stage. So you really want to look at the person in all their different functional domains

00:28:50.980 and stage them on their different domains, that helps you better than just giving an overarching

00:28:57.480 stage, which I find to be very reductionist, simplistic, and does a disservice to both the

00:29:04.100 patient and the family. I want to ask you a question that I am positive you get asked every

00:29:10.360 time you go out and someone finds out what you do for a living, which is what explains the fact

00:29:16.860 that I might be seemingly in perfect neurological health and yet I find it very difficult to

00:29:26.580 remember names of people who I know and I don't mean like know very well like my wife or my kids

00:29:32.440 but like a person who I've met 30 times I might see them for the 31st time and it might take me

00:29:40.520 a full minute to remember their name so people ask me this all the time because they experience it

00:29:46.760 and truthfully I observe it in myself a number of times and it's not just the name of an individual

00:29:51.820 it could be I don't know like the name of a city or something like that it just happened to me the

00:29:59.840 other day I couldn't remember Ibiza I was like what is that place in Spain again where everybody

00:30:07.620 goes to party and it literally took me it felt like an hour it was probably 15 seconds but it

00:30:14.160 was an eternity to come up with Ibiza. Like, I don't know how I could forget something like that.

00:30:18.840 So again, I don't believe that I have Alzheimer's pathology, but what explains this phenomenon of

00:30:26.340 occasional delays in acquisition of names? This is, you're right, one of the most common

00:30:33.880 questions that I get asked and one of the most common requests. Will this treatment help me

00:30:40.120 remember better, particularly nays. And I always say, get used to calling people darling and dear. 1.00

00:30:47.000 Become more Southern, you know, because the truth is names are inorganic bites of information. Our

00:30:54.280 brain is trained to remember things organically. So a name is really a neologism. It's a made up

00:31:02.520 thing. It doesn't have very much meaning. And we're now, particularly in modern times, obviously

00:31:08.600 inundated with names, whereas 100 years ago, there were very few names that one had to remember.

00:31:16.620 There are certain people who have a proclivity for remembering names, politicians, for example,

00:31:21.580 someone you might know who never forgets a name. But the vast majority of us, our brains are not

00:31:28.320 meant to remember an endless amount of names. We really, maybe, I mean, some people speculate that

00:31:34.680 Maybe we're only supposed to remember 150 names, which is a very small amount.

00:31:40.720 I would say that the majority of us have access to most names, and it's not, to me, a viable thing to try to recover as we get older, because it's an effort that I think is doomed to failure.

00:31:57.840 The other issue, which I think is more problematic, is the ability to recall common nouns.

00:32:04.680 To remember to call something a chair, to say, well, that's a book.

00:32:09.340 I want to know the five different ways you can say green, you know, chartreuse, taupe, taupe, hunter green, et cetera.

00:32:17.640 Those kinds of things, if you're beginning to have trouble coming up with finding words for that, finding words as opposed to trouble finding names, that becomes more of a problem.

00:32:29.700 Because that kind of memory is much more housed in the part of the brain which deals with facts as opposed to names, which I think, unless they're well rehearsed, like the names of your family members, are much more episodic.

00:32:48.420 Their names are more like episodes, triable, things that happen that have a specific time and place.

00:32:55.480 What did I have for breakfast?

00:32:57.000 What did I eat yesterday?

00:32:58.160 what color was the jacket of the person I saw two days ago. None of those things matter. Our brains

00:33:05.600 are designed to forget most of the things we're exposed to most of the time because that's how

00:33:11.660 our brains function. And all these episodes, all these times where we have little micro memories

00:33:19.560 on a daily basis, millions of micro memories, if we kept them in our brain, we wouldn't be able to

00:33:26.180 function. So evolutionarily, we're designed really to forget most names and most episodes.

00:33:34.640 That is very reassuring to hear because even though that has been my intuition and that is

00:33:40.280 how I have generally counseled patients when asked this question, there's always a tiny part

00:33:46.140 of me that gets worried when I can't remember one of these names. Let's talk a little bit about now

00:33:52.400 the evolution of treatment. Actually, no, no, there's one question I want to ask you about

00:33:56.820 before we get there. We haven't alluded to it yet on this podcast, but it's been discussed

00:34:01.400 on other podcasts and we've written about it. And I think many people listening may be familiar

00:34:05.840 with this fact, but there is a significant difference in the prevalence of Alzheimer's

00:34:13.540 disease between men and women. There are lots of speculations as to what those are. I have my own

00:34:18.620 thoughts. I'd like to hear you sort of provide your views on maybe what could be driving that

00:34:24.920 and anything else that is different beyond just, you know, the top level prevalence, right? There

00:34:30.780 must be some biological or nuanced differences in presentation and things of that nature as well.

00:34:36.820 Right. So, you know, let's not forget that compared to 150 years ago to now,

00:34:44.180 our life expectancy has nearly doubled. You know, we've gone from an average life expectancy of 50

00:34:50.260 plus, 50, 52 years to now late 70s, early 80s. And women, 150 years ago, generally were dead 0.99

00:35:00.300 by the time they reached menopause, which is the time period when we stopped making estrogen.

00:35:07.320 And we now know that estrogen receptors co-localize in the brain, particularly in different parts of

00:35:13.760 brain that deal with memory, like the hippocampus. And estrogen, the presence of estrogen drives and

00:35:21.000 drives synaptic sprouting in the hippocampus. Dominique Turan Allerand at Columbia did some

00:35:28.100 of the early work in this. And Barbara Sherwin at McGill did quite a bit of work in terms of

00:35:34.840 what happens to women who go through surgical menopause, who've had, you know, who suddenly

00:35:40.140 stop making estrogen. And their deficits and memory deficits, they start to have trouble with

00:35:46.640 short-term memory. They have trouble with executive function. They have trouble with recall, with

00:35:52.120 word finding. This looks identical to the early stages of Alzheimer's disease. So imagine someone

00:36:00.840 in their 40s who may not realize that they're going through menopause or perimenopause because

00:36:08.020 it's a very gradual process, and they start to notice these changes, then they can, and

00:36:14.000 I've actually had this, I have patients in my practice who've been misdiagnosed with

00:36:20.240 Alzheimer's disease and other dementias, who really only turned out to have menopause-related

00:36:26.120 cognitive impairment, where they have findings that are identical to what you see in early

00:36:32.360 Alzheimer's disease that are objectively seen and that can be treated with hormone replacement and

00:36:38.560 if that's not possible with various other interventions. Now why is it, is it because of

00:36:45.640 this lack of estrogen for the pretty much half their lives or a third of their lives that women

00:36:52.080 are at higher risk? That's possibly one reason. It could also be because men die quicker from

00:36:59.980 cardiovascular disease than women do. Women tend to survive more after cardiovascular insults.

00:37:07.100 And therefore, whenever you have any kind of risk factor for cardiovascular disease,

00:37:13.000 that also puts you at higher risk for Alzheimer's disease. The two conditions are so intertwined

00:37:20.540 that years and years ago in 2002, there was a paper that said, is Alzheimer's stroke published

00:37:26.720 in the journal circulation. Because if you look at the risk factors for Alzheimer's, every single

00:37:31.400 one is also a risk factor for stroke and cerebrovascular disease. So women surviving past

00:37:37.880 a time period where men die may also put them at higher risk for Alzheimer's. There's certainly a

00:37:44.280 huge difference in immunological susceptibility to illnesses between men and women. There's a

00:37:51.600 big difference in multiple sclerosis risk between men and women. There's a big difference in various

00:37:57.060 other rheumatological conditions between men and women as lupus erythematosus. So there may be

00:38:03.500 immunological factors that we don't yet know that are driving a higher risk in women. Reason being, 0.92

00:38:10.020 and of course, people always say, well, women live longer. Maybe that's why they're at higher risk. 1.00

00:38:14.020 But we do know that even when you adjust for that increased life expectancy of women, they still are at higher risk.

00:38:24.780 And the other interesting thing about women who present with Alzheimer's, who have con, is that they present differently.

00:38:31.440 Women often will present with depression.

00:38:34.220 They present with withdrawal.

00:38:36.280 They often have more language difficulty problems than men do.

00:38:41.180 Men present more often aggression with agitation, and so the presentation is different as well.

00:38:48.760 And what about the pathologic findings when you get right down to the assessment?

00:38:54.600 If you took a man and a woman who would ultimately turn out to both have Alzheimer's disease,

00:39:00.220 are you going to see differences in the patterns of amyloid versus tau

00:39:05.100 versus any of the other markers that you're looking at as sort of standard markers?

00:39:10.420 Generally speaking, not, Peter. Generally speaking, it's very hard to differentiate between the sexes based on the biomarkers or based on the differences in testing. But certainly, I mean, in terms of their presentation, I would say very strongly that women are far more likely to present with depression.

00:39:31.620 And also women who present with Alzheimer's tend to generally have gone through earlier menopause, tend to generally not have been on hormone replacement.

00:39:42.360 Yeah. And unfortunately, as we sit here recording this in 2026, we're still sort of dealing with that problem because you again have that generation of women I refer to as that lost generation that were with, you know, denied HRT following the WHI in 2002.

00:40:01.680 And so these are probably the women that are coming fully into the consequences of that at a higher rate because these are women that would now be in their 70s having never had hormone replacement therapy, which again is, I'll get off my soapbox before I even get on it on that topic.

00:40:21.880 We didn't actually, maybe just for the sake of completeness, finish the swing fully, my bad, I took us away from it, on some of the top-level biomarkers. So we mentioned them, but maybe give folks a sense of what you're looking for.

00:40:34.720 So I think people nowadays, meaning patients themselves, are even requesting, hey, you know, Doc, can you look at my AB4240? Can you look at my PTAO217? Can you talk about the most common commercially available validated assays and what the levels mean? And also, I'd like you to, after that, give us your point of view on the utility or futility of testing asymptomatic people.

00:41:02.900 May I talk a little bit about the differences in diagnostic approach here?

00:41:08.380 Absolutely.

00:41:09.380 Okay. So I'll actually tell you a story. I saw a patient in the early part of last year

00:41:15.980 who just had, you know, is a high-functioning man who had, in his 70s, kind of vaguely wasn't

00:41:23.420 feeling well, ended up seeing a neurologist who I really respect at an academic center here in

00:41:30.260 the city and got tested for, he actually happened to get the Quest testing for AD detect tests

00:41:38.140 to see, and it turned out he had abnormal levels of A-beta 42 and therefore was diagnosed with

00:41:48.320 Alzheimer's by this neurologist. And he came to me for a second opinion. My feeling was, you know,

00:41:55.640 after talking to him, I wasn't really convinced that he had really cognitive impairment. I thought

00:42:01.340 he felt a little slow, a little sluggish. Maybe there were physical issues. But be that as of May,

00:42:07.840 we did amyloid scan and the scan was negative. So for about three weeks, he had thought he had

00:42:14.400 Alzheimer's. He didn't have Alzheimer's just based on the scan alone. So I'm saying this

00:42:19.580 as a cautionary tale to say that you want to be careful making a diagnosis of Alzheimer's with

00:42:27.160 just the blood tests alone. I mean, there are certain blood tests, the LumiPulse blood test,

00:42:32.860 the Percivity blood test, the Quest AD blood test, the C2N assays. But all these are generally,

00:42:41.080 all the blood tests are standardized against, for the most part, I think there are a couple that

00:42:46.740 aren't, but the ones I mentioned are all standardized against amyloid beta pathology

00:42:53.100 on PET scan. So because amyloid is present, so if you take all people in their 70s,

00:43:03.720 community-dwelling people, and you image them, about 25% of us will have amyloid in our brain.

00:43:10.240 If you take people in their 80s and you image them, about 30-plus percentage of us will have amyloid.

00:43:17.860 And by the time you get to 90, about 44% of us will have amyloid without symptoms, living in the community.

00:43:25.760 So if you're going to standardize a blood test and say this blood test is sensitive for picking up Alzheimer's and it's specific for Alzheimer's,

00:43:36.880 In other words, when it's positive, it means it really is Alzheimer's, not another condition.

00:43:41.140 But you're validating it against an amyloid PET test, PET scan.

00:43:46.540 You see my problem there, is that because amyloid is positive in so many people as we

00:43:53.660 get older, many of us without any clinical symptoms, what does this really mean?

00:43:58.960 And so this is now a critical juncture, I think, in terms of how we diagnose Alzheimer's,

00:44:05.680 Because there are two groups that have different approaches to diagnosing Alzheimer's.

00:44:11.440 That is the Alzheimer's Association criteria for diagnosing Alzheimer's, which diagnoses Alzheimer's primarily based on the presence of amyloid, whether it's in the blood, whether it's in the brain.

00:44:24.620 And there's the International Working Group, which is mostly European, but international.

00:44:30.460 A member of the National Institute of Aging is a part of it.

00:44:34.240 a few American major centers are a part of it as well. That group feels that there are too many

00:44:42.940 people who have amyloid so that when you make a diagnosis, you really want to be sure that you

00:44:49.720 have amyloid and tau and symptoms. So it's a clinical biological condition rather than a

00:44:58.260 primarily biological condition. So I'm careful about testing people who are asymptomatic unless 0.99

00:45:05.180 they have a family history. So for example, I have a young physician in her 50s, both her parents had

00:45:11.700 Alzheimer's, confirmed Alzheimer's, and she has two copies of the four. She's somebody where you

00:45:17.920 really want to do testing, even though she has no symptoms, right? And she turned out to have

00:45:23.560 amyloid in her brain and she's being treated to clear the amyloid. So there is, I think, a role

00:45:30.220 for preclinical testing in people who are at high risk, but the vast majority of people, we should

00:45:36.320 be careful. And if the blood tests are positive and it's surprising, you may really want to get

00:45:42.940 further testing to confirm the diagnosis because otherwise, as the International Working Group

00:45:49.060 very eloquently put it, you have a large population of, quote, patients and waiting,

00:45:55.660 unquote, people waiting to become patients, but maybe in the process changing their whole approach

00:46:00.860 to living. You've got to be careful. Yeah, this is, to me, what separates this type of screening

00:46:06.700 from cancer screening. Obviously, I'm very vocal about my view on early and aggressive cancer

00:46:12.580 screening, and I certainly take a lot of criticism for it, but I find it very easy to defend the

00:46:17.320 position and I won't do it now. However, I think the type of screening you're describing is

00:46:23.440 fundamentally different, not only because of the pretest probability, but because of the implications

00:46:30.260 of what to do about it. And so I think your approach makes a lot of sense. And unfortunately,

00:46:37.920 yeah, go ahead. Just to go back to your cancer analogy, when PSA screening first came out,

00:46:45.140 you know, a while back. The incidence of patients being diagnosed with prostate cancer just rose

00:46:51.840 dramatically. And it was only a few years later that we were able to approach it in a more nuanced

00:46:58.300 way and understand that PSA levels rose as we get older. So you really want to plug all the factors

00:47:05.900 in rather than just knee-jerk, oh, high PSA equals prostate cancer. You got to figure out age, etc.

00:47:12.280 And similarly, I think we will probably arrive at a more nuanced approach to the presence of amyloid and equating that with a diagnosis of Alzheimer's disease.

00:47:23.980 So let's talk about the patient you just mentioned, only in the sense that it's a great foray into maybe something that is not the standard way we would treat.

00:47:36.700 So the example you gave is a highly functional woman in her 50s who carries two copies of

00:47:44.500 the E4 gene.

00:47:45.560 And again, I think just maybe for the sake of ensuring everybody is familiar with what

00:47:49.800 the statistics are, if you're carrying two copies of the APOE33 gene, so if you're the

00:47:56.960 wild type, and you don't have a particularly noteworthy family history, what is your lifetime

00:48:03.300 incidence as a male and a female for developing Alzheimer's disease?

00:48:07.560 I'm going to let you come up with those numbers.

00:48:10.380 Okay.

00:48:10.800 You know, because I'm not.

00:48:13.040 No, I mean, I think for, you know, it's in the ballpark of, gosh, I don't even know anymore.

00:48:18.800 I was going to say it's probably about 6% for a man, 10 to 12% for a woman, if you're 0.81

00:48:25.160 a 3-3. 0.88

00:48:26.300 Yeah.

00:48:26.660 But it could be a little less than that.

00:48:28.780 Yeah.

00:48:29.280 Yeah.

00:48:29.620 Now, that said-

00:48:30.320 For me, yeah.

00:48:31.240 Yeah, go ahead.

00:48:31.620 The way I think about it is I say, if you have two copies of the 4-4, that kind of is akin to

00:48:37.380 you having BRCA, the BRCA gene for breast cancer. So it increases your risk by 60% as opposed to

00:48:44.100 the wild type. So this woman has two copies of the E4 gene, which again is not deterministic

00:48:51.140 the way the PSEN1, PSEN2, or APP genes that you loosely referred to earlier may be closer to

00:48:57.800 deterministic. But again, it is a heavy, heavy risk modifier. And she's totally fine, but she's

00:49:05.240 saying, look, I don't want to wait until I'm not fine. So is there something we can do? And of

00:49:09.740 course, everything is predicated on the belief that the answer to that question is yes. If the

00:49:13.680 answer is no, then why would we do anything about this, right? But the answer is yes. So she comes

00:49:18.700 in. Did you go straight to the amyloid brain scan or was that just the first thing that came up

00:49:24.640 positive in the suite of testing that you did? That was the first thing that came up positive

00:49:30.180 in the test. Actually, she was a person where we did a spinal tap. So the level of amyloid in the

00:49:37.940 brain was determined by the presence of low amyloid because most of it is getting deposited

00:49:43.500 so that there's very little in the spinal fluid amyloid. And she also had tau that was kind of

00:49:50.300 borderline, and she had two copies of the four. And, you know, this is very important. Actually,

00:49:57.340 people will come to me almost every day who is a new patient and says, oh, you know, I have a

00:50:03.340 family member with Alzheimer's. And I always ask them, how was that Alzheimer's diagnosed?

00:50:08.840 Was it diagnosed by autopsy? And the answer is almost never, yes. And up until 2007 or 8,

00:50:16.240 When we had the spinal tap, we really had no way of diagnosing people anti-mortem before

00:50:21.900 death.

00:50:22.560 So whenever people say they have Alzheimer's, I actually, in my family history, write down

00:50:27.560 that the person has a family history of dementia, but we don't know what type it is because

00:50:32.820 most often they may not have had Alzheimer's.

00:50:35.680 But in this physician's case, because I also happened to take care of her father, I know

00:50:41.000 that he really did have Alzheimer's because we've done all the tests, the amyloid and

00:50:45.420 the PET scan. And the mother who had died did not have an autopsy, but most likely had Alzheimer's

00:50:52.340 as well, based on her presentation. So in her case, even though she did very well overall,

00:50:58.940 she may have had a couple of areas that I was not quite happy with on the cognitive scores,

00:51:05.480 you know, things that gave me pause. Her language wasn't quite as fluent. Maybe there were a couple

00:51:10.300 of areas where her memory could have been better for a physician compared to her overall ability,

00:51:15.600 which was quite high. And therefore, we opted to start her on a whole slew of preventive

00:51:23.500 treatments to help her reduce weight. She went on a GLP-1, but she also went on a monoclonal 0.97

00:51:30.480 antibody to remove the amyloid that was beginning to build up in her brain. And I think that was

00:51:36.440 something that I was very excited to do with her, that she was very excited to embrace because now

00:51:43.480 there was a possibility that we could change her destiny, you know, and, and that's really the kind

00:51:49.980 of thing I find wonderful. Okay. So let's, let's talk about these anti-amyloid therapies because

00:51:56.800 they're a little bit of a mixed bag. In fact, I don't know the order, but was aducenumab the

00:52:03.980 first one that was approved? Yes. Okay. And gosh, it's, it's hard to believe how fast time has

00:52:09.400 flown, but I want to say it's been, was this pre, this is probably about six or seven years ago.

00:52:15.900 So adacanamab, I can tell you exactly when adacanamab was approved in July of 2021.

00:52:24.500 Oh, wow. It was post COVID. For some reason I thought it was pre COVID, but,

00:52:27.500 But it was a very controversial approval in the scheme of things because there was an advisory board to the FDA that recommended against approval.

00:52:38.900 I'll let you explain why.

00:52:40.920 The FDA decided against, entirely within the purview of the FDA, to go against its advisory board.

00:52:46.620 But it's just not often the case.

00:52:48.140 But in this case, it did, and it went ahead and approved the drug.

00:52:50.680 So walk us through the controversy there.

00:52:52.700 Why did the advisory board say one thing and the FDA another?

00:52:55.140 The advisory board was not—so the adecanumab, which is a drug that clears amyloid plaque in the brain, so it's a monoclonal antibody against amyloid, as are the two other drugs subsequently approved, lakanumab, lakembi, and donanumab, kisunla.

00:53:16.460 All these three drugs reduce and eventually eliminate brain levels of tau so that when you, the person's been on the medication for some time, there's no longer any plaque in the brain.

00:53:30.460 The problem is that it's not necessarily associated with clinical benefit, and that's really the crux of the problem that the advisory board had, was that even though the drug was effective in clearing plaque, it didn't necessarily translate into significant clinical benefit for the patient, and it was associated with serious side effects, including brain bleeding and brain swelling.

00:54:00.460 And it was not insignificant in cost.

00:54:04.000 When it was first approved, I think it was somewhere in the mid $50,000 range per year

00:54:08.660 for treatment.

00:54:10.220 So for those reasons, there was a department and I think several members left the advice.

00:54:15.760 There was a lot of brouhaha around it, but it was approved.

00:54:19.040 And I actually was one of the first adopters of adicanumab because I thought that despite

00:54:25.500 the controversy, I felt that if we could move the needle a little bit for patients with Alzheimer's,

00:54:35.320 it was worth trying. So I started using it. And because the drug had a greater than 40%

00:54:43.100 incidence of abnormal brain bleeding and brain swelling as a result of the use of the drug,

00:54:51.200 I developed a very, very slow titration protocol. My reasoning with this protocol was, well,

00:54:57.520 the amyloid's been developing over decades in these patients. There's no reason to go in and

00:55:04.180 quickly accelerate the dosing. We could go at it slowly. So what if it takes two years to clear

00:55:11.340 the plaque as opposed to a year and a half? And another big problem with these drugs as a group,

00:55:17.040 all three drugs is that they are not generally used for patients who have two copies of the

00:55:24.240 APOE4 allele. The feeling being that these patients who have two copies of the 4 allele

00:55:29.720 are much more likely to have significant aria, which is amyloid-related imaging abnormalities,

00:55:38.020 edema, and hemorrhage in the brain. But the problem is the patients who have two copies of

00:55:42.840 the four, like my physician patient, are the ones who are most likely to progress and also the ones

00:55:49.140 most likely to need this drug. So I've developed a protocol where we start and titrate patients up

00:55:56.500 very slowly. And in fact, recently for the drug donanumab, which is a third drug in this group,

00:56:02.900 they have the FDA has approved a slightly slower titration protocol than initially used because

00:56:10.740 they too have noticed that a slower protocol is helpful in drastically reducing risk of

00:56:16.800 side effects.

00:56:17.900 The one thing I will say is that if you look at the amount of change, what's called the

00:56:25.020 CDR sum of boxes change between patients on drug versus patients on placebo on all these

00:56:33.180 three groups, lacanumab, which is a second drug, aducanumab is no longer on the market.

00:56:39.020 It was taken off the market.

00:56:39.940 So we're really only talking about lacanumab and donatumab.

00:56:43.760 Lacanumab is a once-every-two-week drug intravenously.

00:56:48.020 Donatumab is a once-a-month drug.

00:56:50.440 If you look at the changes on the CDR sum of boxes scores, which is, you know, changes in memory, ability, etc., the changes are small.

00:56:59.720 It's 0.3 or 0.4 points out of 18 points on the CDR sum of boxes score.

00:57:07.220 This is a very small change.

00:57:09.160 And in fact, in the early data on donanumab, there was really no change on the CDR sum

00:57:14.580 of VOX scores between patients on drug versus patients not on drug.

00:57:19.800 But I do believe that many of these patients had moderate to severe disease.

00:57:25.400 So if you treat patients early on, before there's a lot of tau pathology, before there's

00:57:33.480 too much synaptic loss, then you are able to, I think, have greater benefit. And that's been shown

00:57:41.160 to be the case. And then, of course, there's the question of what if you treat people before they

00:57:45.560 ever get the symptoms? What if they have the amyloid, they're at high risk. Like my physician

00:57:50.420 patient, are you even able to prevent the disease? So these are questions that we don't know. But

00:57:55.740 certainly in my patients, even what I would say to your listeners is if they have two copies of

00:58:01.220 the 4-4, they may benefit from a very, very slow titration protocol. And we've published on it,

00:58:07.460 and our data is we find about a 4% risk for amyloid-related imaging abnormalities in such

00:58:14.460 patients. Can you tell me what is the sort of boilerplate incidence of ARIA, these abnormality,

00:58:21.680 these amyloid-related imaging abnormalities in 4-4s to contrast with the 4% you're seeing in

00:58:26.940 your series? So generally, what happens in patients who have two copies of the 4-4 is they

00:58:34.920 have a lot more cerebral amyloid angiopathy. To distinguish it, so amyloid deposited in the blood

00:58:44.640 vessels of the brain. And this amyloid can sometimes cause a break in the continuity of

00:58:56.500 the blood vessels going into the brain and cause small microhemorrhages when there's bleeding,

00:59:02.340 or it can cause white matter changes. It seems as though people who have two copies of the 4-4

00:59:09.100 have higher prevalence, more severe cerebral amyloid angiopathy, and are more prone to

00:59:18.040 developing ARIA-E and ARIA-H, amyloid-related imaging abnormalities with edema or with

00:59:24.460 hemorrhage, and are also more severe.

00:59:27.240 So they're much more likely to have larger bleeds, more edema, and more problematic.

00:59:33.600 This is also one reason people often get confused.

00:59:36.380 They say, well, why is it that people on placebo develop microbleeds, as did people on drugs?

00:59:44.500 That's because microbleeds can happen in people who have cerebral amyloid angiopathy, regardless of anything.

00:59:50.880 So the fact that you get the monoclonal antibody, it increases your risk.

00:59:57.480 But you may have a baseline risk anyway because you have two copies of the fluoroalele or even a copy.

01:00:03.580 Okay, so let's go back to the very beginning on that.

01:00:05.480 So adikinamab is approved.

01:00:08.440 The controversy is basically it improves the biomarker.

01:00:11.300 It doesn't improve the disease.

01:00:13.240 The example I would give here is it's a drug that lowers LDL cholesterol but doesn't prevent heart attacks.

01:00:20.200 You decided, and I don't want to sort of paraphrase or misrepresent, but I think what I've heard is, look, you said there's an alternative explanation here.

01:00:30.000 A, we could be looking at too short a period of time.

01:00:32.380 B, we could be looking too late in the history of the disease. If we intervene earlier, give

01:00:39.240 ourselves more time and mitigate the side effects by ramping the drug up slower, we are after all

01:00:45.160 removing amyloid and amyloid is playing a role in the disease. Is that a fair assessment of the

01:00:50.480 nuance that you sort of applied to the first drug, which is no longer there, but the logic still

01:00:55.140 applies to the second and third? That's correct. Yeah, exactly right. And the other thing, Peter,

01:01:00.740 I always ask myself anytime a new drug comes out is, would I want this drug? If I had Alzheimer's

01:01:07.860 now, knowing everything I know about this drug, would I want to be on this drug? And I felt that

01:01:13.100 with adecanumab, my answer was yes. So that's why I started giving it to my patients. Now,

01:01:19.080 was it ultimately pulled off the market because of area or was it pulled off the market because

01:01:24.440 of lack of efficacy? It was pulled off of the market because they required post-monitoring

01:01:33.780 of the drug because of the controversy, and that became financially for the company, I think,

01:01:39.200 unfeasible. Got it. What is the annual cost of lacanumab and decanumab?

01:01:45.080 The drugs themselves cost about $26,000, but then there's the administration fees,

01:01:52.300 which can range, you know, sometimes in institutional settings, they can be as high

01:01:56.900 as $10,000 just to infuse the drug versus you go into a infusion center, it could be $400. So

01:02:03.760 there's always that spread per infusion. And then there's a cost associated with the MRIs that the

01:02:10.520 patients might need, which they need, and then they need amyloid imaging. And one of the things

01:02:17.100 that I unfortunately have discovered is the insurance companies, especially with lukanumab,

01:02:22.560 which starts at 10 milligram dosing right away, 10 milligram per kilogram, they will not reimburse

01:02:28.880 patients if we start them at a lower dose and titrate them up. So in other words, your efforts

01:02:35.680 to protect the patients from area-related imaging abnormalities is something the insurance companies

01:02:42.760 are saying, hey, we don't have a protocol for this. So the patient's going to have to pay out

01:02:46.720 of pocket if they want a safer protocol. That's correct. Yeah. That sounds about right.

01:02:51.720 We haven't run into, yeah, it's very unfortunate. We haven't run into that with the donatumab so

01:02:56.920 much because donatumab does have a 350 to 700 to 1050 to 1400 milligram titration schedule.

01:03:05.440 And so we've been able to like keep people at the 350 for like four or five infusions without

01:03:11.040 anybody noticing. But unfortunately, we haven't been able to do that with lakanumab because it

01:03:15.660 just starts at 10 milligram per kilogram. And we've had patients, Peter, who are 4-4 patients

01:03:22.780 who've had severe aria on just three milligrams of lacanima, three milligrams per kilogram,

01:03:30.520 as opposed to the 10. So can you imagine what their side effects would be on 10 milligrams

01:03:35.320 per kilogram? And the reason for this is simply the higher initial risk. In other words, it's

01:03:42.240 It's that the APO, if you're a 3-4 and then ultimately a 4-4, with each of those steps,

01:03:47.140 you're moving towards a higher susceptibility for the edema and or the hemorrhage in response

01:03:54.260 to the drug.

01:03:55.180 Can you say a little bit more about what it is the, and I know you kind of alluded to

01:03:58.560 this already, but what is it that the drug is doing in the process of clearing amyloid

01:04:05.080 or tau that is resulting in edema and hemorrhage?

01:04:10.480 So the monoclonal antibody, whichever one it is, the drug goes into the brain and it

01:04:18.100 starts to clear amyloid in the substrate of the brain, but it also removes amyloid from

01:04:25.760 the tunica media or the middle coat of the small arteries in the brain.

01:04:32.060 That's where amyloid is.

01:04:33.700 So what that does then is it disrupts the lining of the arteries, and it causes leakage

01:04:42.940 of initially fluid, and that's called edema, ARIA-E.

01:04:49.220 And then eventually there's extravasation of actual blood because there's a rupture

01:04:53.680 of the lining.

01:04:54.760 There's a tear in the lining of the blood vessels, if you will.

01:04:57.940 And that then causes ARIA-H or hemorrhage.

01:05:01.380 And if it's a large enough disruption, then you can actually have a massive brain bleed.

01:05:06.920 And there are cases of death in such patients, particularly if they're on a blood thinner.

01:05:12.460 You know, we have patients who could be on a blood thinner for atrial fibrillation or

01:05:16.720 something like that.

01:05:18.060 Alternatively, also, you know, for example, one of my younger patients who's on it for

01:05:24.080 prevention. She's 51 or 2, and she's developed significant ARIA-E on both sides of her brain

01:05:33.000 on 3 milligrams per kilogram of lakanumab. And she's been on this dose now for six doses,

01:05:38.380 which is almost, you know, it's three months. And she still has ARIA-E in different areas

01:05:44.140 because her brain is, you know, there's little tears in the fabric of the blood vessels as the

01:05:52.420 monoclonal antibody removes amyloid from the lining of the blood vessels.

01:05:57.300 Now, is she having any symptoms or is this something you're only able to know because

01:06:01.780 you're doing the concomitant MRIs along the way to monitor progression?

01:06:06.760 Is this something you never want to wait until it shows up symptomatically?

01:06:10.340 You want to know that it's there radiographically and pull back in response to that?

01:06:16.180 That's a great question.

01:06:17.300 And the answer is absolutely.

01:06:18.800 So we have a very rigorous monitoring protocol with imaging.

01:06:24.300 And I will say I've been using monoclonal antibodies since 2021, so it's been over five years.

01:06:30.280 And I would say I cannot remember, perhaps in one 4-4 patient who was in her early 60s with Alzheimer's, she had some visual findings.

01:06:40.800 But everybody else, we only discovered the brain bleeding and the brain swelling on MRI.

01:06:48.800 Sorry, you're saying that in the five years of your very liberal use of monoclonal antibodies, your incidence of ARIA is 4%, but symptomatic ARIA has only been one case?

01:07:04.500 Symptomatic aria has been one case.

01:07:06.360 I did have one case, which was a drug.

01:07:09.540 And all the patients have stayed on the monoclonal antibody, including one patient who had significant aria, where he had arias defined in multiple places, some of them rather large, aria-y.

01:07:25.680 And he had absolutely no symptoms.

01:07:28.500 And this was one of, he was on adecanumab.

01:07:31.260 He was a 4-4 patient. He was in his early 60s, and he had been on adecanumab, a slow titration,

01:07:41.080 and he had gone up to 8 milligrams per kilogram. And at month 19 or 18, he developed on MRI

01:07:49.540 significant ARIAE. We're talking fairly significant, with some microhemorrhages,

01:07:55.900 without any symptoms, including on a careful neurologic exam.

01:08:00.340 And how far do you pull back the drug when you see that?

01:08:04.000 So that patient was actually quite scary.

01:08:06.920 I mean, that was my first real, I mean, that was maybe 2024 or so before the other drugs

01:08:13.160 that really come out and early 2024.

01:08:17.000 And he, what we did was we realized that the aria had happened a few weeks before because

01:08:27.220 of the characteristics on the MRI, it wasn't acute. And he actually happened to, wanted to go to

01:08:34.040 Europe the following week. His exam was normal. So he went to Europe and then he came back and we,

01:08:40.680 I think we cut down to a lower dose. We went down to six milligrams and then went back up again.

01:08:46.580 So it was a scary moment. What I do do now, what I've learned to do is when you have patients who

01:08:53.840 have people with CAA, cerebral amyloid angiopathy, can have ARIA without, as I said, without any

01:09:01.880 impetus. They can just have spontaneous ARIA A and E. And generally the way you treat it

01:09:09.560 is you treat it with steroids. So what I do in my patients who have, like for example,

01:09:16.840 this young woman who's got 4-4 and 3 lakanumab, 3 milligrams of lakanumab, who has ARIA-E,

01:09:25.560 which seems fairly refractory, I treat her with steroids as you would somebody who has

01:09:31.460 cerebral amyloid angiopathy with the inflammation. And we just pre-medicate with steroids,

01:09:37.980 and that seems to control it. Ah, pre-medicate. So in other words,

01:09:42.500 you're saying you prophylactically manage with a small dose of a steroid before each subsequent

01:09:47.400 dose once you've made the diagnosis of RNA. So are there any clinical trials that are going on

01:09:55.200 to take this protocol of yours and do it in a randomized fashion? Because obviously you're

01:10:00.920 doing it in what we would call open label, meaning everybody is being treated this way. Nobody's

01:10:05.300 getting either a placebo or a standard approach. To me, an elegant trial, quite frankly, although I

01:10:12.260 think it would be cost prohibitive, and that's probably why it's not going to be done, would be

01:10:16.480 taking your low and slow approach, what I would call the low, slow, early approach, right? So we

01:10:24.120 take these high risk patients that are in their fifties or sixties, who are probably 10 years

01:10:29.680 away from clinical disease or five years away from the first symptom. We treat them in a way

01:10:35.000 that is very low, very slow, very safe. And we ask the question, are we bending the course of

01:10:41.480 their disease. Because to me, that's the interesting question, right? I think what

01:10:45.280 these drugs have demonstrated is if you take a person who is in florid dementia and you slam

01:10:52.200 them with these monoclonal antibodies, you are not bending the arc of their life. And in fact,

01:10:59.080 if they are E44, you are introducing a staggering amount of risk. In fact, I assume many neurologists

01:11:06.160 would not even treat an E44 late enough in the game with a high enough dose because the risk and

01:11:11.780 reward isn't there. So it's a shame that, you know, the companies that make these drugs wouldn't

01:11:18.640 subsidize the cost in a way that would allow a more longitudinal study in what I think is the

01:11:23.840 appropriate patient population. But, you know, I'm not going to diminish the cost of monoclonal

01:11:29.420 antibodies. They're certainly more expensive than your typical small molecule, but they're probably

01:11:33.700 not the sticker price. And I think there's more people to be helped if we could do the right

01:11:38.880 study and again, kind of replicate what I think you're doing clinically. Because again, the

01:11:43.520 approach makes sense. If you think about most chronic diseases, time and area under the curve

01:11:50.040 is the issue, right? We know this with atherosclerosis. If I take a person for primary

01:11:56.720 prevention, I'm not going to move the needle treating them for a year. It's just not enough.

01:12:01.900 I have to treat them for years to move the needle. So anyway, I think that's hard to imagine. So tell me what's on the horizon as far as the monoclonal antibodies. Like where do you see, you know, with every drug, whether it be GLP-1s or statins or any drug, they always get better with subsequent generations. So where do you see them going next?

01:12:23.620 I think we're going to have drugs that are monoclonals that are able to like ameliorate this risk of, you know, edema and hemorrhage, drugs that can be taken as pills, drugs that can be given at home.

01:12:39.580 And I do want to say the most exciting thing that happened to me, I have a woman in her 70s who came to me with, you should see in a, you know, a television show in which we were discussing Alzheimer's.

01:12:55.280 And so she came to the office and she turned out to have Alzheimer's as she's very, very, very functional, not particularly.

01:13:03.260 I mean, I think her overall ability was at the 70th percentile, so not, you know, one of these

01:13:08.960 uber high-achieving people. And we started her on medication after a spinal tap showed evidence of

01:13:17.500 amyloid and tau. And she was on the CanMab. She eventually cleared the amyloid. I decided to do

01:13:26.360 a tau scan. She no longer had tau. And, you know, there is downstream reduction in tau in all these

01:13:35.220 different, with all these monoclonal antibodies, because even though they target amyloid, they also

01:13:40.560 clear some tau. But she actually had no tau on her tau scan. So now we're waiting on her spinal tap

01:13:47.640 because I refuse to believe that we've basically erased, removed her pathology. Clinically, she's

01:13:54.720 done remarkably well. So again, speaking to the spectrum of Alzheimer's, there are some patients

01:14:00.220 who respond dramatically. And who knows, maybe her immune system is just different from someone

01:14:06.680 else. I have a pair of identical twins, and I take care of one of the two. And the one, she just,

01:14:13.340 you know, she was on a monoclonal antibody, initially on adecanumab and then on donanumab.

01:14:19.260 And she did much better than her identical twin, who was in a facility.

01:14:25.400 And my patient actually also had heart disease.

01:14:28.120 She had rheumatological heart disease, which destroyed her valve.

01:14:31.920 And she had an artificial heart.

01:14:33.760 And she was on Coumadin. 0.99

01:14:35.640 And we still decided to put her on a monoclonal and watch her like a hawk.

01:14:39.960 She never developed side effects.

01:14:41.740 And she cleared her amyloid.

01:14:43.380 So I'm just saying that I think it's an exciting time.

01:14:47.040 I really do.

01:14:47.740 And I think early diagnosis really does make a difference now in terms of changing the trajectory of someone's cognitive health.

01:14:58.240 So you've been in practice for 25, 26 years, and these drugs have only been available for five years.

01:15:04.280 So prior to the approval of the first monoclonal antibody, what were you doing for this type of a patient?

01:15:12.140 And what were you doing for the later stage patients?

01:15:14.860 And how much of those things are you still doing today?

01:15:17.740 So I've always, you know, I actually trained in the era of HIV and AIDS.

01:15:23.380 So I, and you know, we, all my early patients who died, died from HIV.

01:15:28.740 You'd see them one day, they were dead the next week.

01:15:31.020 And I, for the first time a few weeks ago, saw a patient who actually had not just HIV,

01:15:37.980 wasn't just HIV positive, but actually had clinical, had had clinical AIDS with PCP

01:15:43.700 and Kaposi-Sarcona, who had survived going on immunotherapy and IVIG and all of this,

01:15:50.840 and now is in his 70s and concerned about his memory. So, I mean, it was the first time that

01:15:56.500 I'd ever encountered that. And the way I've approached Alzheimer's disease is the same way.

01:16:02.140 We don't have a cure yet, but we can give each person a cocktail of medications, just like they

01:16:09.440 do with patients with HIV. And in this particular patient, he was on IBIG. So God knows what it did

01:16:15.720 to his immune system. And therefore, you can keep the person alive and functional, not just alive,

01:16:21.980 but actually functional. And that's important. So before the advent of these monoclonal antibodies,

01:16:28.220 and to this day, I still have patients on a combination of a cholinesterase inhibitor,

01:16:33.900 such as donapazole, galantamine, et cetera, memantine, which is an NMDA receptor antagonist

01:16:43.360 that basically, I just say, reduces signal-to-noise ratio in terms of brain neurotransmission,

01:16:51.260 and it also may prevent apoptosis or cell death in patients.

01:16:55.880 Depending on the patient, we may have them on something like valocyclovir.

01:16:59.740 Some patients benefit from other immune-modifying treatments when needed and better control

01:17:07.560 of their comorbidities.

01:17:09.540 Many patients may benefit from interventions like ventricle peritoneal shunt for hydrocephalus.

01:17:18.480 And the other area that I feel is important is maintaining a person's neuronal connectivity.

01:17:26.240 And by that, I mean how well the networks function.

01:17:30.820 And for that, in 2008, I started using transcranial magnetic stimulation, targeted TMS, to help maintain function.

01:17:39.920 And TMS works by stimulating the brain using magnetic stimulation that then creates a small area of current just under the area of the coil.

01:17:54.340 So an area about one centimeter squared of the cortex gets stimulated, and then that creates a stimulation in the circuitry to which that area of the cortex is attached and connected.

01:18:06.740 And the hope is that by that method, you're able to keep those circuits alive regardless of the pathology, regardless of whether the pathology and the cognitive impairment is from stroke, from menopause, from Alzheimer's.

01:18:22.460 And over time, we've kind of refined this process, and so I continue to use TMS in patients, and I find it particularly helps people with language problems, for example.

01:18:35.360 Are you using TMS with EEG guidance?

01:18:39.180 I know that that's something that's becoming a little more popular with psychiatric disorders, such as recalcitrant depression and things of that nature, but not sure about your application.

01:18:49.180 I really always use TMS using neuronavigation guidance of anatomical structures. So we actually have kind of a, we plug in the patient's MRI and we kind of create a GPS of that person's brain. And then we mark those areas that we think need to be stimulated.

01:19:09.160 And I usually stimulate the dorsolateral prefrontal cortex, Broca's area, the precunias for parietal and visuospatial skills.

01:19:18.680 And some patients, we also stimulate the auditory Wernicke's area.

01:19:24.760 And again, this is off-label because TMS is only approved in the United States for treating depression and for certain electrodiagnostic purposes.

01:19:33.380 But we've been using it off-label since 2008 for treating our patients with dementia.

01:19:41.020 How optimistic are you on the causal nature of the relationship between GLP-1 agonist

01:19:49.240 use, dementia?

01:19:50.860 Obviously, the observational data appear favorable.

01:19:54.260 The challenge with data of this nature is it's a little difficult to establish perfect

01:20:00.200 causality.

01:20:00.820 You have obviously the healthy user bias of people who can afford and use and comply with these drugs. But of course, you have the bigger issue, which is any drug that is resulting in weight loss and an improvement in metabolic health is going to improve risk. And that might alone be a reason to use it.

01:20:18.920 But the real question is, do these drugs offer a benefit in risk reduction through an inflammatory pathway or a pathway that is not so directly tied to weight loss of insulin resistance?

01:20:32.760 I believe so. I think the GLP ones offer an advantage. And even the older drugs like metformin, you know, may offer benefit in this area. And so I do use it in my patients where I think it's necessary.

01:20:47.940 And, you know, the other thing, too, is obesity is finally coming down in the United States, you know, tax the GLP-1s.

01:20:56.120 And I'd be very curious to see what happens as a result of that with so many diseases associated with the brain, many of which are driven by obesity and cardiovascular disease, which is a direct consequence, diabetes.

01:21:11.580 We've talked a lot about Alzheimer's disease.

01:21:13.600 I do want to just give a little bit of time to discuss vascular dementia and Lewy body

01:21:18.900 dementia, and I'm assuming you see quite a bit of those in your practice as well.

01:21:22.780 Can you walk us through, A, how those diseases present, and B, how you diagnose them?

01:21:29.320 It's very rare.

01:21:31.280 Alzheimer's almost never presents in isolation.

01:21:35.160 So they've done autopsy studies in patients and find that between 98% to 99% of patients with Alzheimer's have some kind of concomitant primary brain pathology.

01:21:50.500 And the most common is vascular disease, vascular brain disease.

01:21:56.060 But in terms of the primary vascular dementia, I find that rarer than you might think because so many of us as we get older will have some vascular disease in our brain, what I call white matter schmutz in the brain, and used to be thought of as rather benign.

01:22:15.780 but now we know that it is not benign, that it interrupts the connectome of the brain,

01:22:21.180 and it's progressive and causes problems that perhaps interfere with the resilience of the

01:22:27.680 brain to pathology and therefore needs to be addressed specifically. So I'm fairly rigorous

01:22:34.620 about treating patients with vascular dementia, often with concomitant other conditions as well.

01:22:41.540 And one of the things that I do do in my patients who say have multiple strokes in their brain that I believe are the primary cause of their cognitive impairment is, you know, aside from putting them on statins and all of that, is that they do have atrial fibrillation or some kind of a heart problem that is the cause of the strokes.

01:23:04.340 I recommend a watchman so that they can then go off the anticoagulant medication,

01:23:12.860 Eliquis, Xeralto, et cetera, because many of these patients are at higher risk for falls

01:23:18.980 just from their vascular pathology.

01:23:21.620 They have dementia, so it just prevents an added problem down the road.

01:23:27.660 Along those lines, in patients who are on blood thinners for atrial fibrillation or another condition, who then have Alzheimer's disease and therefore are precluded from going on a monoclonal antibody because of the blood thinner, I recommend them getting a procedure like a Watchman so then they can go off the anticoagulant and then get on the monoclonal.

01:23:54.940 So I've done that with a few patients.

01:23:57.320 Lewy body disease, you know, if you look at patients with Alzheimer's, about 40% of

01:24:02.140 them eventually will have some level of Lewy body pathology.

01:24:05.700 If you look at patients with Lewy body, about 30 to 40% of them have Alzheimer's pathology.

01:24:11.700 And so some people think even a higher number.

01:24:14.040 So the two conditions coexist quite a bit.

01:24:16.900 What I find the most troubling in my practice is the number of patients with Lewy body who get diagnosed as having Parkinson's and therefore are treated with Parkinsonian medications, levodopa and carbidopa, the dopamine agonists like Meropex, et cetera.

01:24:38.840 And that actually worsens their Lewy body.

01:24:41.960 They get more confused.

01:24:43.860 It doesn't really help with their symptoms.

01:24:45.820 And I have quite a few of those patients.

01:24:48.660 But in terms of a diagnosis, for me, again, the big, it's not as straightforward as you

01:24:55.200 might think it is.

01:24:56.760 I do almost, in all my patients in Lewy body, I do work them up also for Alzheimer's because

01:25:02.900 of the vast comorbidity overlap between the two.

01:25:06.980 And if they don't have any of the Alzheimer's biomarkers and they have only Lewy body biomarkers,

01:25:13.680 which is primarily we do dopamine scans.

01:25:17.000 We can also do skin biopsies to see if they have alpha-synuclein pathology.

01:25:21.700 You can also do a spinal tap, send the fluid off for alpha-synuclein.

01:25:26.420 You make the diagnosis.

01:25:28.120 The one thing I will tell people is there is, I think, a fairly prevalent misconception

01:25:35.140 that Lewy body disease is rapidly progressive and causes death in five to six years.

01:25:41.660 and I'm here to say that that is absolutely not true. Patients with Lewy body, in fact,

01:25:47.200 respond fairly well to treatment and sometimes better than patients with Alzheimer's. Some

01:25:52.500 patients actually get dramatically better. And as far as I'm concerned in my practice,

01:25:59.200 the life expectancy is not that different from patients with Alzheimer's.

01:26:03.600 Can we talk a little bit more about the pathology? So how much is Lewy body dementia

01:26:07.100 impacted by APOE4, first of all? Everything's impacted by APOE4, unfortunately. And I think

01:26:15.820 it has to do, so much of what happens in the brain has to do with robust clearance of

01:26:23.500 pathogens. And whenever you have E4 pathology, it interferes with that. So I think there is

01:26:33.800 a slight increase. I mean, obviously, it's not as dramatic as for Alzheimer's pathology,

01:26:39.280 but it does increase risk. So tell me or tell us the role of alpha-synuclein in the pathology

01:26:46.620 progression. Is it a target of therapy? Does it play a role similar to amyloid? Obviously,

01:26:52.820 I mean, maybe start with the easiest case, which might not be that realized, but a pure case of

01:26:58.940 Lewy body dementia without a significant overlap of Alzheimer's disease. So what's happening?

01:27:04.800 What's the order of events in pathology? So, you know, the problem is that Lewy body disease

01:27:12.720 and Parkinson's disease share the same pathology, which is abnormality in the alpha-synuclein

01:27:19.080 protein. And the difference is, sadly, a difference of time. So if you have motor symptoms for

01:27:28.140 less than a year, during which time you then develop cognitive symptoms, then it's Lewy

01:27:34.580 body disease. Whereas if you have motor symptoms and then after a period of time of over a year,

01:27:40.680 and some people say two years, then you have Lewy body disease. It's kind of an arbitrary

01:27:45.800 distinction. In addition, with Parkinson's disease, the pathology, alpha-synuclein pathology

01:27:52.900 is primarily confined to the area of the basal ganglia, whereas in Lewy body disease, it's more

01:27:58.920 widespread. But any disease will eventually, and then Parkinson's, how do you differentiate

01:28:04.320 Parkinson's disease-related dementia? So people now, I think, more and more people are, again,

01:28:11.960 going back to the spectrum concept, think of the alpha-synuclein spectrum disorders of which

01:28:17.980 Lewy body and Parkinson's disease, all of them are on that same spectrum and the presentation

01:28:23.780 may be different. One way I distinguish between Parkinson's Lewy body, because sometimes it's

01:28:30.720 very hard to tell exactly when somebody begins to have cognitive changes, as we've discussed,

01:28:36.320 the smarter the person, the easier it is for them to compensate, is to look for a rest tremor.

01:28:43.020 I have, in fact, never seen a patient with Lewy body present initially with a rest tremor.

01:28:50.960 So that classic pill-rolling tremor of Parkinson's disease, I've never seen in a Lewy body patient

01:28:57.320 ever. And we're talking...

01:28:59.320 thing. So does that mean that by definition, Lewy body, while the alpha synuclein is diffuse,

01:29:08.080 it's sparing the basal ganglion? No, because in Lewy body disease, by definition, you've got to

01:29:15.100 have motor symptoms. So why do you think they don't pill roll? That I don't know, Peter. This

01:29:21.500 is one of the mysteries of neurology that I haven't figured out yet. But that's what I've

01:29:26.800 notice, and that's what I use to make my distinction. Obviously, when you do a DAT

01:29:32.820 scan, there are classic patterns that you see in Lewy body that you don't see in Parkinson's

01:29:38.760 disease. But again, that is not quite as rigorous and as distinctive as you'd like. But it's

01:29:45.440 important from a clinical perspective, because if you have Parkinson's disease, you would want to

01:29:52.940 treat with drugs that increase brain levels of dopamine. Whereas if you have Lewy body disease,

01:29:59.300 adding dopamine can cause serious negative side effects, including psychosis, worsening the

01:30:07.180 psychosis in patients with Lewy body. So Lewy body disease clinically is diagnosed by the presence

01:30:12.480 of motor disorder, Parkinsonian features, fluctuating consciousness, cognitive impairment,

01:30:18.340 and often visual hallucinations.

01:30:21.780 And these can be quite dramatic.

01:30:23.800 And the one thing also about the visual hallucinations

01:30:26.320 of Lewy body disease is patients have insight into them.

01:30:31.280 So for example, I was talking to the mother of a physician

01:30:34.620 who is my patient who has Lewy body disease.

01:30:37.620 And she's a wonderful woman.

01:30:39.620 She's in Florida now.

01:30:40.680 And I was talking on the phone and she said,

01:30:42.620 yeah, you know, I don't understand.

01:30:44.580 I was very, very clear that I was not in my house, and everyone else told me I was in my house,

01:30:52.800 and I kind of knew I should feel like I was in my house, but I really didn't think I was in my house.

01:30:59.920 Now, Alzheimer's patients rarely ever say that. They don't have that insight, usually. Patients

01:31:06.180 with Alzheimer's can also have those kinds of hallucinations and delusions, but their insight

01:31:12.580 is not as crisp. Or else I'll have a patient say, you know, I called the cops because I saw these

01:31:19.000 people on the street. And I knew because I've done this before that they were not real, but they

01:31:26.660 absolutely looked real to me. And even though my wife told me not to call the cops, I called them.

01:31:33.600 I mean, that has to be beyond terrifying. I mean, that's almost like, even though it's a

01:31:39.160 completely different disease. I mean, that almost feels like schizophrenia minus the voices.

01:31:44.620 Yeah, yeah. Yeah, it's rare to have auditory hallucinations. In fact, I've never ever had

01:31:50.400 a patient with that in my practice. So what are the differences once you have the diagnosis made?

01:31:57.080 What are the differences in your playbook of treatment? And are you catching these Lewy

01:32:01.380 body patients later in the disease? Do you have fewer tools to identify early alpha-synuclein

01:32:08.500 relative to amyloid beta? We have fewer widespread tools, right? So we don't, I mean,

01:32:15.640 skin biopsies now are great, easy to do. And pardon my ignorance, that is news to me. I didn't

01:32:22.460 realize you were biopsying the skin for alpha-synuclein. I thought you were getting it out

01:32:25.740 of CSF. So why does this same protein that is involved in neuronal geography making its way

01:32:34.780 into the skin and, and how does that, how do you link what you're seeing in the skin to what's

01:32:39.900 happening in the brain? So, you know, they travel along the neural pathways. So because your skin

01:32:46.500 is full of neural, of nerves, so you're really, when you're doing a skin biopsy, you're really

01:32:52.080 not doing a nerve biopsy, you're doing a nerve biopsy, but you're doing a biopsy of the small

01:32:57.960 cutaneous nerves in your skin. So you do punch biopsies on your neck, further down in your knee,

01:33:05.300 and then further down. So you kind of see how much synuclein there is at each level. And the

01:33:11.700 more synuclein you see further from the nervous system, the more likely it is to be widespread

01:33:16.480 disease. There are so many conditions that can mimic motor disorders that are not caused by

01:33:24.040 synuclein so what the skin biopsies do is that they distinguish between them but that's only

01:33:29.960 giving you information above sensory nerves correct yes it gives you information about yeah

01:33:36.000 so that's you know relatively speaking in the grand scheme of the cortex that's relatively

01:33:40.440 small real estate so the other stuff you're doing clinically so you'll you would infer

01:33:45.840 based on visual hallucinations that the occipital lobe is involved and things like that right no so

01:33:51.320 So for dementia with Lewy bodies, again, as I say, I work patients up, even if they

01:33:58.620 have Lewy body dementia, I always work them up for Alzheimer's because I know that there's

01:34:02.980 such a comorbidity between the two.

01:34:05.380 But the thing that I often will do is I will do a DAT scan.

01:34:10.120 So when you have a dopamine uptake scan, you can tell if there is abnormalities in the

01:34:15.180 basal ganglia that would be consistent with alpha, could be consistent with Lewy body disease.

01:34:22.020 But then you can't really be absolutely sure until you do the biopsy to verify that it's

01:34:27.320 alpha-synuclein that's driving the pathology. In terms of treatment, one of the most important

01:34:34.280 things that I do is get them off their Parkinson's medications. And that can be very

01:34:39.780 difficult because patients, I found, love the high of the Parkinson's medication. I saw a patient

01:34:47.720 yesterday, actually, a new patient who was in her 70s, had been having what's called an essential

01:34:57.660 tremor, which is a fine tremor that many of us have, but that can also be present. There's a

01:35:05.320 comorbidity between essential tremor and the tremor of Parkinson's disease, but she never

01:35:09.920 had a rest tremor. She had, despite a negative DAT scan, she had been put on levodopa carbidopa

01:35:20.380 three tablets three times a day of 25-100 at an academic institution for treating

01:35:28.340 presumable Parkinson's disease despite her negative DAT scan. And the thing she told me

01:35:36.300 is, you know, she says, listen, these pills really help me because right before I'm ready

01:35:43.460 to take the next set of pills, I start to feel a little bit down. So you have an internal kind of

01:35:50.420 low and the pills kind of reset you because of the dopamine high. So patients are often averse

01:35:57.500 to going off the pills. So one of my tasks with her, because she's been on these nine pills a day

01:36:02.120 for the last year, is trying to get her off those pills. And she really didn't have the tremor that

01:36:08.860 she had was in this central tremor and not the tremor of Parkinson's. So what are the

01:36:16.540 predisposing factors for Lewy body dementia where it's the dominant form of dementia? So yes,

01:36:22.180 there may be some overlapping vascular and or Alzheimer's dementia, but I assume family history

01:36:30.780 poses a risk. What are some other known risks for Lewy body? So family history poses a risk,

01:36:36.640 but the problem is... I mean, let's assume it's diagnosed correctly. I know that that's its own

01:36:42.320 problem with diagnosis. Yeah. Right. No, the problem is that with, yeah, if it's diagnosed

01:36:48.320 those correctly. What are the risk factors? I mean, Lewy body usually occurs in patients a

01:36:53.080 little bit younger. Again, vaster risk factors increase risk for Lewy body disease. Alzheimer's

01:36:58.760 disease increases risk for Lewy body disease. Parkinson's disease increases risk for Lewy

01:37:03.920 body disease. And actually essential tremors also will increase risk because they are on that

01:37:10.180 spectrum of possibly kind of morphing into... But do we think that those things are increasing

01:37:17.480 risk or their presence is a demonstration of other factors that are underlying the risk? In other

01:37:23.600 words, whatever it is that's causing everything you just said happens to increase the risk of

01:37:29.680 Lewy body. But what are those other things? In other words, does metabolic dysfunction, just as

01:37:35.360 we believe metabolic dysfunction is truly causally related to the risk of Alzheimer's disease,

01:37:41.480 do we think it is? And by the way, the important point there is therefore it is treatable, right?

01:37:46.040 Therefore, we target aggressively hypertension, dyslipidemia, and insulin resistance.

01:37:51.860 Because those three things, we have such conviction, are causally driving Alzheimer's disease.

01:37:57.640 Do we have the same level of conviction around Lewy body dementia?

01:38:01.740 I feel anything that's good for the heart is absolutely good for the brain, regardless

01:38:08.280 of the pathology in the brain.

01:38:10.700 And having said that, I feel like we don't yet have enough knowledge about Lewy body disease because we've just, we don't have the.

01:38:22.320 We just don't have the data yet.

01:38:23.880 We don't have a robust enough data set.

01:38:25.620 Yeah, we just don't have enough data.

01:38:26.180 Yeah.

01:38:26.660 Yeah.

01:38:26.980 Yeah.

01:38:27.800 By the way, female-male difference in Lewy body, is it one-to-one, two-to-one?

01:38:34.020 How does it compare to Alzheimer's?

01:38:35.280 I think there's, again, a slight, there's more men with Lewy body disease, but...

01:38:44.020 But that's also because we've got more men with Parkinson's and we might be capturing,

01:38:47.780 yeah, we might be doing this thing incorrectly. Got it. Let's go back to females with menopause.

01:38:52.720 You coined a term, I'm blanking on it, like mercy menopause-related cognitive impairment.

01:38:58.620 Did I get that right?

01:39:00.180 You did. I'm very impressed. How did you remember that?

01:39:03.420 I have a great team of analysts that remind me of these things. So is there ever a scenario where you have a woman who's come to you, she's got a history of breast cancer, and you're just saying it's not worth the risk, or she's too afraid of the risk?

01:39:22.260 you may or may not be familiar with my views on the relationship between estrogen and breast

01:39:26.240 cancer. But let's just say, let's take an extreme case, right, which is she is actively

01:39:30.620 undergoing treatment for breast cancer, in which case if it's estrogen sensitive, you clearly

01:39:34.060 wouldn't give her estrogen. But outside of an extreme situation like that, do you consider

01:39:38.440 estrogen and then by extension progesterone if needed to oppose it sort of mainstays of

01:39:44.140 maintenance therapy for women in either prevention or even the treatment of disease?

01:39:49.260 Yeah, I mean, I think those, you know, in patients where there's no contraindication, we do recommend estrogen.

01:39:58.640 Another thing that really works, Peter, for women like this for menopause-related cognitive impairment is just targeted brain exercises.

01:40:05.980 You know, the brain is fantastic at the concept of learned non-use so that you could have a stroke and suddenly not be able to move your right hand.

01:40:15.520 And this could go on for 10 years.

01:40:17.800 But let's say I restrict my right hand, I put it in a sling.

01:40:22.660 I'm sorry, I restrict my left hand, I put it in a string,

01:40:25.180 and all of a sudden I don't have my healthy left hand to help.

01:40:29.760 All of a sudden, 10 years later, your right hand starts to move.

01:40:33.380 That's the whole concept behind constraint-induced movement therapy for stroke.

01:40:39.780 And I find that the same can happen in menopausal women.

01:40:43.740 So the same problem happens in menopausal women where if they start to have trouble with coming up with words, the very common problem, or they're having trouble multitasking, they just stop doing it. 0.64

01:40:58.800 There is some disuse, dysfunction, and so when you actually force them to use those areas, when you do targeted brain exercises, even just as little as once a week, forcing them to use those areas, they recover function quite dramatically.

01:41:18.620 You can also, in such women, use drugs like cholinesterase inhibitors because the underlying mechanism of cognitive impairment is ultimately loss of acylcholine in the nucleus basalis and areas of the brain in these perimenopausal women.

01:41:35.900 And so those drugs can work. We actually, years and years ago, did a double-blind placebo-controlled trial of donapazole versus placebo in women going through menopause with cognitive impairment and found a trend in 21 patients, 21 on drug, 21 on placebo, for improvement in cognitive function in the donapazole group.

01:41:59.560 So I think those, and then brain stimulation,

01:42:02.720 transbrainial magnetic stimulation

01:42:04.100 seems to help women in this situation as well.

01:42:07.600 So there are multiple different ways that we can help.

01:42:10.660 On the other side of the coin,

01:42:13.560 I have patients who are on an anti-estrogen

01:42:16.980 for treating breast cancer or ovarian cancer,

01:42:20.860 you know, anastrozole, tamoxifen, et cetera,

01:42:24.860 who are coming in with severe cognitive impairment

01:42:27.860 because of that who actually want to go off of the drug regardless of personal risk because they

01:42:36.420 just can't stand how their brain's functioning. Yeah, I think this is one of the real pleasures

01:42:44.920 of being a clinician is your patients teach you a bunch of stories. And I think back to one of

01:42:49.860 the first women that taught me that lesson in 2014 or 2015. This was a woman who also came to

01:42:56.960 see me, status post, mastectomies, buferectomies, hysterectomy for, I think, primary breast 0.99

01:43:05.640 cancer, but bad genetics. 0.63

01:43:08.980 Anyway, she was taken off all of her hormone replacement therapy, and the course of chemotherapy

01:43:16.180 in her life turned out to be devastating.

01:43:17.880 So she was really far down that spectrum of mercy in the sense that basically functionally

01:43:24.580 debilitated.

01:43:25.680 And she was a young woman, right?

01:43:26.800 she was in her fifties. So went from being completely normal to, uh, you know, couch ridden 1.00

01:43:33.760 effectively due to brain fog and just inability to do anything. And she came to me with the 1.00

01:43:40.060 suspicion that I think this is due to the loss of hormones. I think this is due to the, the hormone

01:43:45.860 withdrawal. And, you know, after many, many discussions, not only with her, but with some

01:43:51.680 other physicians, we decided to take the risk and put her back on her hormones. And the recovery was

01:44:00.980 remarkable. I mean, it wasn't subtle, right? So it was very quick that she regained her cognitive

01:44:06.360 faculties. And so what the takeaway for me was both the physiology, but also the choice, which

01:44:13.520 is for some people, and I would probably put myself in that situation, I'd rather deal with

01:44:20.280 the risk of cancer than cognitive decline. But again, the bigger question is it should be a

01:44:27.080 choice. And I also have a little bit of an issue with how aggressively aromatase inhibitors are

01:44:34.420 used for things like DCIS, which I think is, I don't think patients fully understand just how

01:44:43.020 small the risk is that they're mitigating with an aromatase inhibitor for five years in the context

01:44:48.760 of a precancerous lesion. And so I think there are these edge cases where we're doing a lot of harm

01:44:54.880 without a very clear benefit. I can't agree more. And I totally, I can't tell you, I mean,

01:45:02.720 patients that I've worked with who are also being seen by folks at Sloan Kettering, for example,

01:45:08.540 in New York, and have decided with me, to your point about choice, that they would rather risk

01:45:15.620 the possibility of recurring cancer than risk the ongoing ineptitude and inability to function in

01:45:25.740 their lives. And, you know, again, I saw a young girl, this can happen, you know, the lack of

01:45:32.680 estrogen driving brain fog, a 21-year-old girl that I see for migraines who's in college,

01:45:39.680 and she spoke to me earlier this week and she said for the last six months she's been in a

01:45:45.800 complete fog, complete brain fog. She can't function. She's in a good university. And

01:45:51.520 I had her go see her gynecologist. You know, she hasn't had her period. And so the estrogen level

01:46:00.920 for her came back at 22, you know, picograms per decet. It's very low. As you know, menopausal

01:46:08.200 women, it's less than 30 or 35. So hers was extraordinarily low. And we don't really know 1.00

01:46:14.400 why that is. And was her FSH or LH elevated as significant? They were not. So it was very

01:46:19.880 interesting. Her FSH and LH were normal, but it was almost like primary ovarian failure of some

01:46:25.480 sort, you know. But in the meantime, she's not able to function. She cannot work at school.

01:46:32.500 And so I told her, I said she may benefit from some kind of low-dose oral contraceptive while they're trying to figure out what they should do because she needs to be able to finish school.

01:46:48.160 And in the cases where you're managing the hormone, do you prefer a transdermal estrogen to an oral one given the other changes with the increase in SHBG and binding capacity and things like that?

01:47:02.500 Yeah, I mean, I actually prefer transdermal in either a gel or a patch because, yeah, you have lower risk in my, for me, the big consideration is lower risk of strokes and venous clots.

01:47:18.360 So you've been on this really incredible journey. You're at the forefront of what can only be described as a very, very personalized, nuanced approach to the early treatment of dementias across the board. What do you believe today that you absolutely flat out didn't believe 10 years ago? Like, what have you actually changed your mind on?

01:47:45.680 I never thought that patients with Alzheimer's could get better.

01:47:50.220 And you really believe that that is the case?

01:47:54.380 Oh, 100% and 20%.

01:47:55.000 I mean, that is just not conventional wisdom, right?

01:47:57.960 Yeah.

01:47:58.360 No, I never, never thought.

01:48:00.180 And even before, even before we had the more clonals, I had patients who got better with

01:48:06.000 treatment, which just didn't make sense.

01:48:08.340 And I used to think before we had the biomarkers, before we had the spinal taps in 2007, I used

01:48:14.800 to think, maybe I made a misdiagnosis. Maybe this person didn't really have Alzheimer's

01:48:18.920 because the teaching is that every person with Alzheimer's inexorably declines. And we also know

01:48:26.940 that 30% of patients in drug trials with the drug trials for Alzheimer's didn't actually have

01:48:33.840 Alzheimer's by pathology, right? They were misdiagnosed clinically. So I used to think

01:48:39.580 that. And then I realized when I had the biomarker ability that actually patients with Alzheimer's

01:48:45.320 could get better. And that was, it took me, I think I've been in the field specializing in

01:48:52.220 this area since 1994, Peter. And I think for the first six, seven years, I was just learning about

01:48:58.520 it. And then I started seeing patients where it would get better with treatment, sometimes something

01:49:03.780 as simple as just brain exercises, which didn't make sense. And I always thought we misdiagnosed

01:49:09.060 them. And then when 2007 came along and we had the actual ability to diagnose pre-antimortem,

01:49:16.760 I realized that patients could get better. Then it took me another five, eight years before I

01:49:23.200 truly believed it, truly believed it because it's so ingrained, was so ingrained that that was not

01:49:31.380 even a possibility. And now I really do believe it. When you think about the next 10 years of

01:49:38.640 your career, what do you think is going to have the biggest impact on your ability to

01:49:48.140 reverse Alzheimer's disease or prevent Alzheimer's disease?

01:49:53.180 I think for preventing Alzheimer's disease, what's going to have the biggest impact is AI

01:49:58.920 for early detection of changes in our patterns of thinking, in addition to a cocktail of

01:50:08.340 medications designed to ameliorate first inflammation and then pathology in the high

01:50:14.000 risk group patients. And in terms of treatment, I actually think targeted neuromodulation will

01:50:21.600 be important in addition to drugs to reduce pathology. Do you know if there are any drugs

01:50:29.660 specifically in the pipeline for neuroinflammation? Obviously, the field of immune modulators is a

01:50:36.360 popular field, though it's often around autoimmune diseases and oncology. But do you know if there

01:50:43.540 are people specifically targeting neuronal inflammation? I mean, there are, I actually

01:50:50.760 can't, I can't, I know there are, but I can't think of any right now. But there are, there are

01:50:56.000 drugs in pipeline? Yeah. Yeah. Okay. And I will tell you years ago, Norm Relkin's group at Cornell

01:51:02.660 had the IVIG, which is pooled autoimmunity infused into patients who had Alzheimer's disease.

01:51:11.740 And I actually used IVIG in a whole group of my patients with Alzheimer's diagnosed

01:51:18.040 by spinal tap analysis.

01:51:20.800 And two of my patients became plaque negative.

01:51:24.820 And one of them stayed stable for 17 years, and the other one is still stable.

01:51:33.140 So eventually, the feeling was that IVIG had no place in treating Alzheimer's disease.

01:51:41.740 But if you looked at the analysis, people who had the APOE4 allele were far more likely to benefit from the drug than people who did it.

01:51:52.680 So that was an early...

01:51:54.800 Which is really not surprising.

01:51:56.440 I mean, we see this over and over again.

01:51:58.060 I know you and I have spoken about Obacetrapib and the profound impact it's having, at least in a pilot study when you look at the biomarkers.

01:52:05.580 But the impact is most profound, the more severe your risk is. So the four fours actually have reversal. The three fours have a dramatic blunting.

01:52:17.780 So, again, I think what I hope comes out of this and where I hope the world is in five years, I don't even want to wait for 10 years, is I hope that more people are taking an approach like you are through clinical trials where we're doing better identification and stratification of risk and tailoring treatments. 0.99

01:52:38.980 Because as you said, the biggest risk of the current landscape of these treatments is that we are trying to take a very, very, very heterogeneous group of genetic predispositions, other comorbidities, disease manifestations, and we treat them with one hammer.

01:53:00.360 We hit everybody with the exact same hammer and we blunt all of the potential responses.

01:53:06.340 Now, I think that's true, unfortunately, in most clinical trials for most diseases. It's true in the way we manage diabetes and heart disease and cancers. Although with cancer, at least you now see gene targeting and mutation drivers and things like that.

01:53:20.980 So my hope is that that's where the field goes and that we start saying we're not treating Alzheimer's disease. We're treating this subtype of Alzheimer's disease in this subset of patient. Just as we don't treat breast cancer, you know, it's are you triple negative? Are you ER positive? Are you HER2 new positive or negative? Those are completely different diseases for all intents and purposes.

01:53:44.340 So I think the work you're doing is important because I think it's helping people start to hopefully align towards that path of both diagnosis and treatment.

01:53:53.860 Yeah, I think you're 100% right.

01:53:56.600 There is no more heterogeneous disease I can think of, no more than Alzheimer's disease.

01:54:04.640 because it's each person with Alzheimer's disease has their own private version of Alzheimer's

01:54:12.640 disease that is different from everyone else who has the disease because that person has their own

01:54:18.760 individual brain. So that, you know, what's really startling, you know, is you take two patients,

01:54:26.720 identical twins with Alzheimer's, and you find, there are studies, publications, that the risk

01:54:34.020 for Alzheimer's is dramatically different in identical twins who are raised in the same

01:54:41.220 environment. Sometimes one of a pair of twins doesn't get the illness for 10 to 15 years after

01:54:46.800 the other one. How does that happen? Because their brains, although as identical as they are,

01:54:54.360 are still different. And that is really what drives the illness, as well as inflammation,

01:55:01.660 epigenetics, so many things. And you absolutely need a nuanced approach.

01:55:11.480 Well, Dr. Devi, this has been a fantastic discussion. And I learned a lot, but more

01:55:17.560 importantly, I think everybody listening did as well. So thanks for taking time away from

01:55:21.600 your clinic to join us. And I'll look forward to many more discussions with you on these topics.

01:55:27.160 Thank you very much for having me, Peter. That was great.

01:55:29.660 Thank you for listening to this week's episode of The Drive. Head over to peteratiamd.com

01:55:36.800 forward slash show notes if you want to dig deeper into this episode. You can also find me

01:55:42.820 on YouTube, Instagram, and Twitter, all with the handle peteratiamd. You can also leave us

01:55:48.400 review on Apple Podcasts or whatever podcast player you use. This podcast is for general

01:55:54.420 informational purposes only and does not constitute the practice of medicine, nursing,

01:55:58.320 or other professional healthcare services, including the giving of medical advice.

01:56:03.140 No doctor-patient relationship is formed. The use of this information and the materials linked to

01:56:08.780 this podcast is at the user's own risk. The content on this podcast is not intended to be

01:56:14.220 a substitute for professional medical advice, diagnosis, or treatment. Users should not disregard

01:56:19.460 or delay in obtaining medical advice from any medical condition they have, and they should seek

01:56:24.040 the assistance of their healthcare professionals for any such conditions.

01:56:28.120 Finally, I take all conflicts of interest very seriously. For all of my disclosures and the

01:56:33.260 companies I invest in or advise, please visit peteratiamd.com forward slash about where I keep

01:56:40.520 an up-to-date and active list of all disclosures.

The Peter Attia Drive - July 13, 2026

#399 ‒ The evolution of Alzheimer's disease and dementia care： how early detection, personalized treatment, new therapies, and a multimodal approach are changing the landscape ｜ Gayatri Devi, M.D.

Episode Stats

Summary

Transcript