The Peter Attia Drive - #381 ‒ Alzheimer's disease in women： how hormonal transitions impact the female brain, the role of HRT, genetics, and lifestyle on risk, and emerging diagnostics and therapies

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

00:00:42.700 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,

00:00:58.000 head over to peteratiyahmd.com forward slash subscribe.

00:01:04.240 My guest this week is Lisa Moscone. Lisa is a neuroscientist, neuroimager, and the director

00:01:09.940 of the Women's Brain Initiative at Weill Cornell Medicine, where she leads research on how sex

00:01:15.380 differences, especially menopause and hormonal transitions, shape brain aging and Alzheimer's

00:01:20.180 risk. She's also a professor of neuroscience and a pioneer in brain imaging approaches that

00:01:24.800 map Alzheimer's disease decades before symptoms appear. In this episode, we talk about why

00:01:29.800 Alzheimer's disproportionately affects women and why women's increased lifespan over men

00:01:35.900 does not fully explain that 2x difference. Talk about the idea that Alzheimer's disease

00:01:41.280 is actually a midlife disease for women beginning long before symptoms, and how menopause is fundamentally

00:01:47.080 a brain event and what happens to brain energy, structure, and immune signaling during that

00:01:52.280 transition. We talk about what advanced imaging reveals about preclinical Alzheimer's disease.

00:01:58.020 We talk about Lisa's work in imaging estrogen brain receptors. We talk about APOE4, specifically

00:02:04.880 other genetic risks and why they impact women seemingly more than men. Some of the nuanced evidence

00:02:11.300 around menopausal hormonal therapy risks, benefits, timing, formulations, and why the WHI caused decades

00:02:18.180 of confusion. We talk about Lisa's new initiative called the CARE Initiative, a global effort to cut

00:02:23.360 women's Alzheimer's risk in half by 2050, along with some practical evidence-based strategies for

00:02:29.040 supporting brain health throughout midlife transitions, including lifestyle, sleep, metabolism, mood,

00:02:34.520 and the involving role of medications, including GLP-1 agonists and CIRMs. So without further delay,

00:02:41.860 please enjoy my conversation with Lisa Moscone.

00:02:49.720 Lisa, thank you so much for coming out to spend time with me today.

00:02:53.440 Thank you for having me.

00:02:54.680 This is actually a wonderful podcast because it combines two topics that we have spent a lot of time

00:03:01.160 on in this podcast. It's two topics that we spend a lot of time on in our clinical practice,

00:03:07.620 but it is probably, at least to my recollection, the first time I've brought them into an intersection

00:03:12.900 here. So one of them is all things that pertain to women's health, in particular, the transition

00:03:17.960 through pre-peri and post-menopause. Again, this is a topic we care deeply about and have very strong

00:03:24.380 points of view on. And then the other, of course, is brain health, which I don't think there's a single

00:03:29.080 person listening to this podcast who doesn't appreciate both the role of the dementing

00:03:34.520 diseases in how they truncate lifespan, but perhaps much more importantly, how they truncate

00:03:39.460 healthspan. And the reason I wanted to talk with you today was because you sit at the intersection

00:03:44.860 of these two, which is you're asking the questions as they pertain specifically to women and brain

00:03:50.860 health. So I just want to maybe start with a bit of your background. So tell me how you came to find

00:03:56.000 this as your focus. It's quite personal for me. So I was born and raised in Florence,

00:04:02.640 in Italy. Which, by the way, I just want to say, literally one of the greatest places on this planet.

00:04:08.780 It's really pretty. I never appreciated how pretty Florence is until I moved. I'm quite proud. Whenever

00:04:16.520 I go back, it's like, oh, this is really nice. As you may know, as people may know, in Italy,

00:04:21.920 families really live together. So I was born and raised in Florence. My parents live in Florence.

00:04:28.020 My grandparents were in Florence. And my parents are nuclear physicists, both of them. I come from an

00:04:34.980 interesting family where half of the family has a PhD, usually in physics. The other half is in the

00:04:41.080 army. So we're very disciplined scientists, some of us. And I grew up in this environment where

00:04:48.640 everything was about physics and biology and studying and learning. And I decided to apply

00:04:55.660 that knowledge to medicine. So I have a PhD in neuroscience and nuclear medicine, which is a branch

00:05:01.740 of radiology. I do a lot of brain scans. Around the time that I started studying neuroscience and

00:05:09.620 specifically memory and language, I was very interested in memory functionalities and cognition.

00:05:16.280 My grandmother, who lived on the same landing as me and my parents, she started showing signs of

00:05:24.060 cognitive decline. My grandmother was the rock of the family. She was this really strong, extremely

00:05:30.780 intelligent woman who went through World War II when she raised the family. My grandfather was

00:05:36.880 a prisoner of war, being in the army. For a long time, nothing broke her spirit until she started

00:05:45.460 losing her memory, until she started losing the ability to communicate, until she started losing

00:05:51.840 the ability of taking care of us. She could not remember how to cook. And that really broke my

00:05:58.280 grandmother and broke us and led to a diagnosis of Alzheimer's disease over time. And what was even

00:06:06.000 scarier is that my grandmother was one of four siblings, three sisters, and one brother. All three

00:06:13.940 sisters developed Alzheimer's disease and passed away from it, whereas the brother did not and was spared,

00:06:20.820 even though they all lived to the same age. Which was what age?

00:06:24.380 My grandmother passed when she was in her late 80s.

00:06:27.700 And how long did she suffer with the disease? When did it start?

00:06:30.880 At least a decade. It was very subtle. Very often, Alzheimer's disease starts in a gradual way. At first,

00:06:39.280 there's some mild cognitive impairment, which she was able to almost masquerade. She had strategies to

00:06:47.740 find the answer, keep going about today without really telling us that she was having a hard time. But then

00:06:54.140 it became quite evident. And in the end, it was very severe because she was healthy otherwise. So her

00:07:01.440 body was healthy, but her mind was not.

00:07:04.360 So all three sisters basically succumbed to this in their late 80s, having the onset in their late

00:07:10.700 70s, which again, I bring that up only to say this is a very typical trajectory.

00:07:14.600 It's very typical. Unfortunately, it is quite typical.

00:07:18.540 This might be a question beyond your research, because I know you're not a clinician,

00:07:22.860 but given that you had such a personal experience, one of the things people often ask me is

00:07:28.380 at what point do patients become aware

00:07:31.720 of what's happening such that it creates enormous distress for them

00:07:37.100 versus

00:07:38.340 when is the cognitive impairment so severe that they are no longer suffering

00:07:44.200 and it's only those of us around them who are suffering,

00:07:47.480 but we could potentially take some solace in the fact that they are no longer suffering.

00:07:51.760 Do you have a sense of that from your experience?

00:07:53.280 Yes. There's research in this showing that we're now able, now that we're getting better

00:07:59.960 at diagnosing Alzheimer's disease, not just using clinical tools, but using biological markers

00:08:06.240 like brain imaging and biological fluids in blood. Now, we can tell when a person is at risk for

00:08:13.780 Alzheimer's or is showing red flags for Alzheimer's fairly early on. And then it's possible to correlate

00:08:19.880 that with what the patient is telling you. Because the tests, the cognitive tests that we've been

00:08:25.880 basing the Alzheimer's diagnosis on for decades are quite...

00:08:30.880 Quite late.

00:08:31.880 They're a little bit late. They're not quite sensitive to the earliest possible manifestations

00:08:37.380 of the disease, which are usually subjective. So what we're learning now is that there's a phase

00:08:42.640 that is a preclinical phase where the disease is underway. You can see the proteins and the lesions

00:08:49.180 either in the brain or in biological fluids. But objectively, there is no impairment. There is no

00:08:57.440 deficit at that point on cognitive testing. Many patients will tell you that they don't feel the

00:09:05.000 same. There's this idea that they are aware that something is changing, that their performance is not

00:09:12.640 the same, that they're not performing as well as they used to, even just a few years prior. And it's

00:09:18.480 very hard to say, is it just aging or is it something more severe? And now we're getting better at doing

00:09:25.360 that. But it looks like the preclinical phase of Alzheimer's disease can last decades, where the

00:09:32.320 disease is underway. It starts with negative changes in the brain that very slowly but surely eventually

00:09:39.320 exceed the brain's ability to compensate. But it takes a long time. The brain is an extremely resilient

00:09:46.220 organ. And unfortunately, many patients are in this gray area, if you will, where they know

00:09:54.540 that something is amiss. But when they come for a neurological evaluation, everything is fine. When

00:10:00.840 they go to see a neuropsychologist, they test with the normative value by age and education. So it's

00:10:07.420 really hard to provide counsel and to offer a treatment plan if we are not able to diagnose the

00:10:14.540 fact that they are, in fact, on the path towards Alzheimer's disease. That could last decades. It could

00:10:21.800 last a really, really, really long time until the deficits are such that there is an objective

00:10:27.080 diagnosis. And usually patients do experience discomfort and depression and anxiety for years

00:10:35.880 until dementia is severe enough that they start forgetting who they are. Or they start forgetting

00:10:42.120 that they have a family, perhaps. Or they start forgetting why they are where they are.

00:10:47.560 So unfortunately, it's a heartbreaking disorder.

00:10:51.960 Lisa, how much of what you just described through the lens of Alzheimer's disease specifically

00:10:57.220 is comparable in other forms of dementia, such as Lewy body dementia or frontotemporal dementia

00:11:04.340 or vascular dementia? I mean, maybe just for the audience, we can put these all in context.

00:11:09.520 Alzheimer's disease is the most common form of dementia, but it is far from the only cause of

00:11:14.700 dementia. So do you want to maybe put it in the context of these others, both in terms of

00:11:19.020 maybe some of the prevalence of these, but also any of the subtle differences in what you just said

00:11:23.900 as far as onset and presentation?

00:11:26.580 Yes. So dementia is an umbrella term that includes different disorders that are typically categorized

00:11:33.140 in terms of pathology, which is the kind of lesion that every disorder expresses most abundantly,

00:11:39.500 and also in terms of clinical symptoms and sometimes age of onset. Now, most people are familiar with

00:11:46.760 Alzheimer's disease and they usually think that Alzheimer's and dementia may be the same thing.

00:11:51.660 It's a common misconception. Alzheimer's is the most common form of dementia, accounting for about 70%

00:11:57.360 of all dementia cases. There are other types of dementia. Now we hear more and more about

00:12:03.400 frontotemporal dementia, for instance, because Bruce Willis, unfortunately, has been diagnosed with

00:12:09.180 that. In that case, the presentation is a little bit different. When we do brain scans, actually,

00:12:15.360 we can tell whether a patient has frontotemporal dementia or Alzheimer's disease based on the pattern

00:12:22.680 of changes in the brain. Frontotemporal dementia tends to occur a little bit earlier in life

00:12:29.040 and predominantly as associated with aphasia, which is disturbed language production. Whereas Alzheimer's

00:12:37.060 is more about memory function. It's more about forgetting things. However, at the end of the day,

00:12:44.380 once the dementia, the different types of dementia are quite severe, there is a lot of overlap in terms

00:12:50.580 of symptoms, the male core. The pathology, the pathophysiology is different, but the symptoms at the end

00:12:57.080 tend to overlap. So it's quite difficult to do a good differential diagnosis.

00:13:02.020 Lewy body dementia is another form of dementia that is due to mutation in the alpha-synuclein protein.

00:13:09.040 So it's a little bit different from Alzheimer's, where the main problem is an amyloid beta,

00:13:14.480 a fibrillar amyloid beta production in lesions that aggregate into plaques, and also neurofibrillar

00:13:20.820 tangles inside neurons. So each one of this dementia has a slightly different

00:13:26.220 biological substrate. Vascular dementia is also very common and tends to overlap with the other

00:13:33.480 types of dementia. In fact, very often we talk about mixed dementia. It's quite rare for a patient

00:13:40.520 to only have Alzheimer's, for instance, and not some vascular damage. It's rare for a patient to have

00:13:48.720 only Alzheimer's disease and not also some features of Lewy body dementia. And for many,

00:13:55.160 many years, the diagnosis was purely clinical. It was a little bit late. You needed to have very

00:14:01.480 clear-cut symptoms in order to be diagnosed as Alzheimer's or Lewy body or frontotemporal dementia.

00:14:08.300 Now that we have access to biological markers, we're getting better and better,

00:14:13.160 and the diagnosis has been done earlier and earlier. So that is hopefully leading to better

00:14:19.700 therapeutic routes for each type of dementia and development of pharmaceuticals that are able to.

00:14:27.260 Right now, we're trying to reverse the damage that, at least in Alzheimer's, makes results.

00:14:34.640 I think optimistically, we might say we're trying to halt progress.

00:14:38.340 Yes, we're trying to move away from trying to reverse the dementia or very severe lesions. We're

00:14:45.620 trying to move back in time. We're trying to catch people when they're still relatively healthy

00:14:52.080 and the potential for delaying the onset of the symptoms or even preventing. Hopefully,

00:14:58.900 the accumulation of the lesions in the brain is greatest, is feasible. So we're trying to work

00:15:04.700 with people who are fairly young. And that is very new in the field of Alzheimer's. Even when I moved

00:15:11.900 to New York, I was already looking at Alzheimer's prevention, what could be done. That was 20 years

00:15:17.760 ago. And I was working at NYU, New York University, the School of Medicine, with my mentor back then,

00:15:24.640 Dr. Moni DeLeon, who's really a pioneer in the field of Alzheimer's prevention. And his team

00:15:30.360 was one of the very few teams in the world to work with individuals who were younger than 65.

00:15:38.700 Because everybody else was looking at those who were 65 and older. He was like, no, we need to start

00:15:44.740 earlier than that. The point you make about the overlap is really interesting. It means that if you

00:15:50.220 look at the prevalence of each type, it will not total 100%. So you said 70% of dementia will have an

00:15:58.740 Alzheimer's component. What are the approximate numbers of frontotemporal, Lewy body, and vascular

00:16:04.720 in terms of just aggregate presence? It's interesting. I think it's difficult to really

00:16:09.320 come up with specific numbers, especially for vascular dementia, because it's always kind of

00:16:13.960 intermixed. Lewy body is usually around maybe 10, 20% frontotemporal, but it's the same. But it

00:16:20.740 depends. I don't know that we can really say. Yeah. We're going to talk a lot about Alzheimer's

00:16:25.540 disease today. And we know that Alzheimer's disease occurs disproportionately in women. It's

00:16:29.760 about two to one. Do any of the other forms of dementia disproportionately occur in women?

00:16:35.880 No. And that's actually something that was very interesting to us when we started looking at the

00:16:41.720 association between female sex and Alzheimer's disease. Because when I started looking into this,

00:16:47.340 this was a while ago, I would ask the question, is it just my family? Or is there a bigger lesson

00:16:54.020 that we need to learn? And the answer was that we've known since the 1990s that after aging,

00:17:01.020 after getting older itself, being a woman is the strongest risk factor for developing Alzheimer's.

00:17:08.740 But when I asked, can we do something about it? The answer was mostly, well, the point is longevity.

00:17:16.160 It's just aging. The idea is that women live longer than men, and Alzheimer's is a disease of old age.

00:17:23.140 So at the end of the day, unfortunately, more women than men have Alzheimer's disease. But there are

00:17:29.400 two things that contradict, in part, this statement. And clearly, aging is important.

00:17:34.640 Yeah. The first one, by the way, is just a simple actuarial analysis. I did this myself 10 years ago,

00:17:40.680 back of the envelope, because that was the first guess. The first time I thought about that question

00:17:45.560 was maybe in 2015. And the obvious answer was, well, women, on average, live two and a half to

00:17:50.820 three years longer, that must account for it. But if you actually go through even something as

00:17:55.700 rudimentary as the CDC mortality tables, and slice them by five-year increments, you can't explain the

00:18:01.920 increased prevalence by a factor of two to one on that delta in age. But I'm sure you have a much more

00:18:08.160 elegant explanation for why this was not the case. One thing that came to mind, obviously, was that the

00:18:13.520 difference in the longevity gap was not 10 years. It wasn't that wide. It was just a few years. And for

00:18:19.860 instance, in England, the gap is about two years. But Alzheimer's disease and dementia, the whole category is

00:18:26.300 the number one cause of death for women and not for men. But then the other point is that if it was just

00:18:32.960 aging, then women would have a higher prevalence of other age-related disorders and neurodegenerative

00:18:39.640 disorders relative to men, only they do not. Right. Cancer and cardiovascular disease are also

00:18:45.120 age-related. But even within the dementias, right, for vascular dementias, 50-50, Parkinson's disease

00:18:51.420 with dementia is more prevalent in men. Is more in men. Yes. Frontotemporal dementia seems to be more

00:18:56.260 prevalent in men. A Lewy body dementia is above 50-50. So that just doesn't seem to be a good

00:19:03.440 way or good reason to dismiss an important question. Is it safe to say the scientific

00:19:09.580 community today has stopped with that sort of excuse? And we've now fully accepted the fact

00:19:14.620 that there is something biologically different about women that is leading to this enormous mismatch?

00:19:20.040 I wish we could say they would have passed it. We're not. Really? Yes. Yes. It's still

00:19:25.740 a subject of very active debate in my field. Sorry, just to make sure I understand. We're debating

00:19:32.760 why it's happening or we're debating that age is the reason that it's happening?

00:19:37.240 We're debating that still. Yes. We're still thinking that survival and longevity is something

00:19:43.980 that may be driving the higher prevalence. And the argument for this is actually not a bad argument.

00:19:48.720 They say prevalence is something that you look at cross-sectionally. Yes. But what about the

00:19:55.000 incidence? When does it come on? Yes. And the question is, if prevalence is higher, but incidence

00:20:01.820 is not, then it could be aging. It could be age. That's right. So the question is, do women develop

00:20:08.480 Alzheimer's at higher rates than men? And I'll just explain to the listener what that means. You and I

00:20:14.280 obviously understand it. Incidence is the number of cases that occur over a given period of time. So

00:20:20.560 you might say the incidence of this cancer is this many cases per 100,000 people per year. Prevalence,

00:20:26.980 as you stated, is the cross-sectional cumulative number of people at any point in time that have

00:20:32.900 the given condition. And so to your point, if the incidence is identical at every section in time,

00:20:39.960 but the prevalence keeps getting larger as time goes on, then you might have to ask the question,

00:20:46.520 are fewer women dying of the disease and therefore accumulating cases?

00:20:51.760 Yeah. And I think it's been very difficult to get a good estimate of incidence with the studies that

00:20:58.360 we have because the diagnosis of Alzheimer's disease has changed over time. And we're catching more

00:21:04.740 people now even earlier than we did in the past. So there are more and more studies showing that

00:21:11.340 the incidence is also higher among women, especially in countries with low to middle socioeconomical

00:21:18.900 status. Another way to think about this mathematically would be because, as you said, incidence is very

00:21:25.320 complicated because the diagnosis is so complicated. It might be easier to look at mortality and tally up

00:21:33.500 the mortality differences. Because if you really think that you would reverse test this hypothesis,

00:21:40.520 if you think incidence is too high, you should see men dying at a much higher rate at a comparable

00:21:47.100 stage of disease to women. In other words, the women should be outliving the men with Alzheimer's

00:21:52.780 disease if they're all getting it at the same rate, which I don't think is the case, of course,

00:21:56.720 but that would be a way to test that hypothesis in the negative.

00:21:59.460 Yes, that could be. And in fact, in some countries, Alzheimer's disease is actually the number one

00:22:05.420 cause of death for women over 65.

00:22:07.760 That's staggering.

00:22:08.640 Whereas heart disease, yes.

00:22:10.000 What countries?

00:22:11.060 European countries and also in some parts of the United States. So the data is just coming out

00:22:16.020 because many scientists are in a way puzzled that longevity has been the only explanation for the

00:22:24.040 disparity. And we're now looking at risk in ways that they think are doing more

00:22:29.080 justice to the question that we could do before, especially by looking at biological markers,

00:22:35.740 because we can see that, and this is a lot of my research, when we do brain imaging or we look at

00:22:42.200 other biofluids in midlife and we compare, let's say that we have a population of men and women who are

00:22:49.100 like 45 to 65. And all these participants, all these people have a family history of Alzheimer's or

00:22:56.340 perhaps the APOE4 genotypes. So they are technically at higher risk for Alzheimer's than the average

00:23:02.700 person, than people who do not have these risk factors. If you look at the brain scans of men and

00:23:08.140 women, at least in my work, but it's been replicated, many other scientists and other teams,

00:23:13.700 the women tend to show more red flags for Alzheimer's disease in midlife as compared to

00:23:20.840 men of the same age. And this is quite consistent. And we've also seen the progression of lesions in

00:23:28.280 the brain tend to be faster in women. So then when you compare men and women who have the same symptoms

00:23:35.940 and the same level of dementia severity, the women's brains actually harbor more pathology.

00:23:43.340 So what seems to happen is that we start developing the lesions of Alzheimer's, the pathology in the

00:23:49.400 brain earlier on than men, starting in midlife. And we live longer with it, but we're able to compensate

00:23:57.380 more in that the tests that we use to diagnose Alzheimer's are heavily reliant on things like verbal

00:24:04.560 memory, which women have a little bit of an advantage in. And so it's more difficult to

00:24:09.440 diagnose Alzheimer's in women early.

00:24:13.040 Because they have a higher cognitive reserve in the metric you're using to test.

00:24:16.740 Yes. So in a way, women are masking the fact that there is Alzheimer's in their brains,

00:24:23.740 but not necessarily overperforming.

00:24:26.540 But starting at a higher reserve.

00:24:27.820 Yes. Starting at the higher level. So the idea is that we live with Alzheimer's longer.

00:24:33.540 And that may lead to a higher number of Alzheimer's patients among women down the line.

00:24:39.200 So that really changed the whole question about women's brain health and Alzheimer's risk. Because

00:24:45.060 what we and others have shown is that Alzheimer's is not a disease of old age. It's a disease of midlife

00:24:52.920 with symptoms that start in old age. Alzheimer's starts in midlife with negative changes in the

00:24:59.780 brain and that later on lead to the symptoms and the clinical diagnosis of dementia. But then that

00:25:06.180 changes the question, right? Because if Alzheimer's is not a disease of old age, but it's a disease of

00:25:11.400 midlife, and women have a higher risk of Alzheimer's disease, a higher long-life, long-term risk of

00:25:18.700 Alzheimer's disease compared to men starting in midlife. Then the question that we should be

00:25:24.140 asking, I believe, is, well, what happens to women and not to men in midlife that could then

00:25:31.160 potentially explain the higher risk of Alzheimer's down the line?

00:25:34.980 Yeah, I just wrote that down. I think that's such a profound statement, Lisa. And it actually reminds

00:25:38.280 me of a statement I've made many times on the podcast, quoting one of our guests about something

00:25:42.760 totally different, which is osteoporosis. The guest said that, look, osteoporosis is a childhood

00:25:48.520 disease. It just doesn't manifest in childhood. But you reach your genetic ceiling of your bone

00:25:55.360 density by the time, in the case of a woman, by the time she's 18 or 19 years old. So if a woman

00:26:00.380 isn't able to create enough deformation in her bones and all of the things that lead to strong

00:26:05.360 bones by the time she's 18, 19, her risk has already started, even though that disease won't

00:26:10.620 manifest until she's 60. And I think your example is, frankly, even more terrifying, but it's well stated.

00:26:17.120 So let's talk a little bit about some of the theories for this. I think you and I are in

00:26:21.480 pretty strong agreement that age alone cannot explain this, even if it partially contributes

00:26:26.440 to it. There must be some tail effect of age. Well, let's accept that. The first idea that I

00:26:32.220 think would pop into anybody's head if they're thinking about this for more than two minutes

00:26:35.540 is a fundamental difference between men and women is that men have a very gradual loss of

00:26:43.400 androgens throughout their life. But they're never shocked with androgen deprivation. Whereas

00:26:49.960 women have a sudden and shocking loss of androgens at about the time you're talking about in the

00:26:57.480 middle of their life, they will lose their sex hormones. Of course, the question is, does that

00:27:03.120 play a role? So where does that idea fit into this? And what other ideas fit into it that could

00:27:08.800 explain this? Again, this is a subject of very active debate in my field to the point that people

00:27:15.940 almost feel like they have to take sides. It's very interesting. It's an interesting time to be doing

00:27:21.140 this work. So I think a good way to answer is by, for instance, doing brain imaging, which is what we

00:27:28.020 have been doing. And back in 2017, we published the first study, which is bizarre to think about it.

00:27:35.460 It was 2017. But it was the first study showing the brains of women before and after menopause.

00:27:43.420 Everything we had up until that point was done after menopause, looking at menopause more like an

00:27:49.420 outcome. Whereas we were looking at what happens during the transition to menopause, which is the

00:27:56.460 most neurologically active phase, if you will. And so we had, in the first study, we had three groups

00:28:02.740 of women, premenopausal with a regular menstrual cycle, perimenopausal with regular menstrual cycles,

00:28:09.980 and postmenopausal up to age 65. So no more menstrual cycles for over a year.

00:28:16.040 No hormone replacement therapy in that group?

00:28:17.440 No, no hormone replacement therapy. And then we had age-matched men, because premenopausal women

00:28:24.220 tend to be younger than the postmenopausal ones. And what we found was that before menopause,

00:28:30.320 at the premenopausal stage, regular menstrual cycles, there were barely no differences between

00:28:35.880 women's brains and men's brains.

00:28:37.600 And sorry, Lisa, what type of scan are you using for this?

00:28:40.440 So we're doing different brain scans. We use MRI to look at brain volume, to look at presence of

00:28:48.980 lesions in the brain.

00:28:51.300 Let's be very technical, because I think this really matters. So we're going to talk about lots

00:28:55.760 of different types of scans today. But the way I typically try to explain this to my patients and

00:29:00.120 to the listeners of this podcast, and feel free to correct this oversimplification as someone who's

00:29:06.060 trained in a radiographic field. I always want patients to understand the difference between

00:29:10.760 imaging modalities and functional modalities. So when you look at a test, like an MRI, you are doing

00:29:19.000 it for anatomical resolution. And again, you can choose how you do that. You can T1 versus T2,

00:29:26.720 this flare versus that. You can highlight white matter versus gray matter. You can highlight the

00:29:32.400 vascular system, but you're looking for anatomic resolution. Conversely, if you do something like

00:29:37.720 an FDG PET scan, you're not looking at anatomic information. You're looking at functional information.

00:29:43.520 You want to understand how metabolically active, in the case of FDG PET, the cells are. CT scans tend

00:29:50.300 to be much more anatomic, etc. Would you agree with that way to think about them? And how do you

00:29:55.260 think of the suite of different radiographic studies that can be used?

00:30:00.100 Yes, I think what's most important is what you're measuring. You can use different tools to measure

00:30:05.440 different things. And the really good definition is whether you're looking for structure,

00:30:09.520 information for functional information, for biochemical or pathological information.

00:30:15.440 With MRI scans, what you can do is look at the anatomy of the brain for sure, is to look at

00:30:21.520 volumetrics. You want to make sure that some parts of your brain are really nice and dense with neurons.

00:30:27.980 Whereas if we find signs of atrophy, that could be a risk factor for future dementia. There's one part

00:30:33.860 of the brain that we always look at is the medial temporal lobe, which is a combination of structures

00:30:40.460 that are quite primitive, if you will, and that are highly involved in memory function and also

00:30:45.540 emotional regulation. And there's one structure in particular, the hippocampus, which is considered

00:30:51.540 a biomarker for Alzheimer's risk. Because the hippocampus, you really want it to be as big as

00:30:58.160 possible. You want the volume to be really nice and large. But when we find reductions in volume and

00:31:04.600 thinning of the structure of the hippocampus and the parahippocampal gyrus, which is right below,

00:31:10.420 that is a risk factor for Alzheimer's. It doesn't mean you have Alzheimer's. It means that that is a

00:31:15.800 red flag for potential Alzheimer's risk down the line. And then like you mentioned, this is called

00:31:21.180 the T1 MRI. We usually use it for volumetrics, also to make sure that there are no brain tumors,

00:31:26.760 there are no obvious vascular damage strokes. So that's a good first line. It's a good first

00:31:32.620 baseline. But we also do usually T2 and FLIR scans that give you additional information on other

00:31:39.760 parameters that are important. Like if you have gliosis in your brain, which is a bit nonspecific,

00:31:46.940 but it is a sign of white matter integrity damage. It's like little punctuations in the brain

00:31:53.720 that tend to emerge with aging, but could also be a sign of inflammation, of vascular insults.

00:32:01.320 So it's good to monitor that. We can look at the vascular system in the brain. What I also like to do,

00:32:08.000 and we have it in all our biomarker panels, is that we use MRI with some modifications. So we can do DTI,

00:32:15.920 diffusions tensor imaging, where you can see the structural connectivity of your brain,

00:32:21.360 all the different fibers that connect the different neurons, and you can extract a lot of information

00:32:27.220 from those images. And then we also use a modified version of an MRI to look at blood flow with ASL,

00:32:36.920 arterial spin labeling, which is completely non-invasive and is also really quick. But it is

00:32:42.200 helpful to look at whether the brain receives enough blood flow at any given time.

00:32:48.300 And then we use petroscopy. We do 3D1P, phosphorus 3D1 magnetic resonance petroscopy,

00:32:56.160 to look at ATP production in the brain. So far, it's the only technique, except there's one

00:33:02.900 potential with PET that's still being established. But this technique can give you a good read on the

00:33:10.000 ratio of phosphocreatine to ATP production, which we find to also be a potential biomarker

00:33:17.240 for brain stress, almost. When the brain is in a state of energetic damage or crisis,

00:33:25.340 that could signify that the neurons are under metabolic stress. So we do that too. And this

00:33:31.180 is all MRI. So we can do everything in less than an hour.

00:33:35.380 And you will get all that information on multiple sequences of one scan?

00:33:39.380 Yes. We can do everything. Well, it's different scans.

00:33:43.000 Yeah. I mean, you run the patient through different sequences, but under one table time.

00:33:47.220 Yes. One table time. We do need to switch the coil for this petroscopy scan. So we bring you

00:33:52.700 out to the scanner for just a minute. We switch the coils and go right back in.

00:33:56.480 How many coils do you need for that scan?

00:33:58.820 At least 36, but I think more potentially. You switch from the hydrogen to the phosphorus,

00:34:06.340 phosphorus. But that's just something the technician does.

00:34:09.560 This is something you would have a really hard time doing outside of the brain because of motion.

00:34:13.940 Yes. Oh, yes.

00:34:15.180 One advantage of the brain is you lock that head in place and it's a short distance under the magnet.

00:34:20.540 Yes. So those are really good images. And then we also do positron emission tomography

00:34:25.500 or PET scans. We use FTG, like you mentioned, that looks at metabolic activity in the brain.

00:34:31.720 And we also use another tracer. It's called C11-PIB, Pittsburgh Compound B,

00:34:37.460 which shows Alzheimer's plaques in the brain.

00:34:41.680 Is that what we would colloquially refer to as an amyloid pet?

00:34:44.640 Yes. It's the amyloid pet. It's just the tracer. It's the carbonated. It has a C11,

00:34:49.400 which means that you can do the FTG and the PIB right away, back to back. You don't have to wait

00:34:56.360 and bring the patient back the next day. So you can do everything quickly. And another advantage

00:35:02.540 is that PIB has a very clearer signal than the fluorinated tracers, which is helpful.

00:35:09.860 The signal to noise ratio tends to be a little bit higher. So you get a clearer read, which is

00:35:15.220 helpful for people who are younger because we're not using it diagnostically. We're using it for

00:35:21.380 research and the read is a little cleaner. So we get a better signal to noise ratio, I think,

00:35:27.980 but other people think it too. It's all a matter of whether or not you need to make it. You need to

00:35:32.380 have a cyclotron right there. You need to have a chemist that can make it for you and then just

00:35:36.800 run upstairs and you inject because it decays really quickly. So you need to have this big nuclear

00:35:42.480 medicine capability on site. If you don't, or if you prefer to use a fluorinated tracer that you can

00:35:49.660 buy commercial, then it's perfectly fine. But this is all the scans that we are doing. And now I'm

00:35:55.880 really excited about this. We're also doing brain estrogen imaging. So this is the first time that

00:36:02.640 people have been trying to measure estrogen and hormones in the brain for a really, really long

00:36:07.640 time. It's very hard to do it for a number of reasons. Back in 2019, I went to my radiochemistry

00:36:13.860 department and I said, we think that menopause is very important for Alzheimer's risk for women.

00:36:20.960 And we assume that is the decline in estrogen levels that drive the increased cellular aging

00:36:29.340 and biomarker risk of Alzheimer's in women. But that needs to be proven because all the information we

00:36:34.940 have is from rats. So we need to see what happens in women. And we also need a tool to measure what

00:36:42.500 hormone therapy is doing in the brain. And they said to me, well, that sounds really, really great,

00:36:47.700 but we don't have it. And sorry, you wanted to measure estrogen or estrogen receptor density?

00:36:53.120 I would like to measure both. What I can measure today is estrogen receptor density.

00:36:58.180 Okay. So what we do is that we have estradiol and we label estradiol, the hormone,

00:37:05.040 with a fluorinating molecule. It's just attached to the estradiol. Then there's an injection.

00:37:10.640 Estadiol, the estradiol goes in the body, but accumulates in the brain. And the way it works

00:37:17.020 is that this little molecule mimics estradiol itself and looks for the target. I think your

00:37:23.940 listeners know this, but the way the hormones work is that the hormone is like a key that needs to open

00:37:29.320 a lock. And the lock is the receptor. And every type of hormone has a specific receptor. So estrogen

00:37:36.140 has estrogen receptors, progesterone has progesterone receptors. The way that this tracer works is called

00:37:41.920 fluorine-18, fluoroestradiol, that it goes up in the brain and it looks for the estrogen receptors.

00:37:48.720 It binds to the receptors and it works like it kind of jams the lock. So the receptor just is almost

00:37:57.240 frozen in time for the period of time that the estrogen is there. And then the F18 molecule

00:38:03.840 starts shooting out gamma rays and we can take a picture of that from the outside. And then we use

00:38:10.820 filter back projection and other techniques to get an image of the brain. And we can use that with

00:38:17.860 kinetic modeling to get a measure of estrogen receptor density in every part of the brain.

00:38:25.400 So we can finally do that. And just in 2024, we published the first proof of concept study

00:38:34.320 showing that we can get a signal, especially in the pituitary gland where the signal is specific,

00:38:41.240 like it's not confounded by metabolize or by blood beam barriers.

00:38:45.180 And this is not counterintuitive. I mean, we would expect to see a high density of estrogen receptors

00:38:50.560 in the pituitary gland, independent of the disease we're talking about, just because we want to see

00:38:55.480 the feedback. I mean, I guess we'd expect to see it in the hypothalamus even more.

00:38:58.920 Yes.

00:38:59.400 Because we would want to get feedback for FSH and LH. Is that your thinking?

00:39:04.140 Yes. Totally my thinking. The blood-brain barrier is a big issue.

00:39:08.200 But the pituitary is slightly...

00:39:10.660 Outside.

00:39:10.700 Isn't the hypothalamus is outside, right?

00:39:12.380 The hypothalamus is inside the blood-brain barrier.

00:39:13.780 The pituitary is outside.

00:39:14.620 The pituitary is half and half.

00:39:15.820 Yep.

00:39:15.960 So the back is protected by the blood-brain barrier. The anterior part is not. So the tracer

00:39:21.860 goes in really easily, which is helpful to us.

00:39:24.900 Yep.

00:39:25.280 So for now, this is what we're doing. We're able to measure estrogen receptor density in the pituitary.

00:39:30.480 Just go back to that for a second. How is estrogen getting across the blood-brain barrier

00:39:35.460 outside of that access to half of the pituitary?

00:39:38.660 There are transporters.

00:39:40.300 And what's the time course? So if you injected me in the arm through my IV,

00:39:45.960 how long until it traverses? And does it only traverse in the free component?

00:39:52.360 Does the estrogen have to be unbound or does it bind to albumin or something else?

00:39:56.460 It does bind. Yes. That's why we need kinetic modeling. So the timing is relatively fast.

00:40:01.920 If we inject now, we can see uptake within minutes. And then what we do is that we keep

00:40:10.020 seeing the tracer accumulating up in the brain. So we do a time activity curve,

00:40:15.260 a tracer uptake in the brain relative to the tracer kinetics in blood. And we need both to get a good

00:40:24.460 sense of how much is actually sticking to the receptors and for how long and how much is just

00:40:30.940 pushed back into the circulation. So the whole scanning time is 90 minutes,

00:40:39.040 but the peak of uptake is within 30 to 35. So I would say between 30 and 15 minutes is when you

00:40:47.280 get the most signal. And then you start to saturate.

00:40:50.820 Yes. Also, there's a component of blood flow that you need to disentangle. There's a whole

00:40:55.940 mathematical model that we're using. It's called the Logan plot. And I mean, I don't know if your

00:41:02.000 listeners want to know this. I don't know if they do, but I do.

00:41:04.820 Okay, good. So what we do is then we have spent so much time trying to find a good reference region

00:41:11.580 for the modeling. Because if you have a reference region that you know to be free or almost free

00:41:18.960 of estrogen receptors. Right. It's your negative control. You can subtract it out.

00:41:22.740 Yes, that's exactly the problem that it took a long time. We had to talk to preclinical scientists,

00:41:30.520 to cell biologists, to pathologists, to people who really specialize in the estrogen receptors.

00:41:36.200 And I've been working with Dr. Roberta Diaz-Brinton, who's a legend in my field. She's been doing this

00:41:42.060 for, I don't know, 40 years. She knows everything about estrogen receptors. And working with her and

00:41:48.280 looking at all the postpartum studies, we found that there's a very specific part of the cerebellar

00:41:53.740 cortex. It's this part of the brain that people say is mostly involved in movement control, but has a

00:42:01.400 number of different functionalities. And there are estrogen receptors deep in the white matter

00:42:05.860 of the cerebellum. But if you look at, let's say this is the cerebellum from the side. If you look

00:42:12.080 at the inferior most part of the gray matter of the cerebellum, like the thinnest layer possible

00:42:18.800 towards the posterior inferior part, that seems to be consistently void of estrogen receptors.

00:42:26.680 Whatever receptors are found there tend to be beta receptors. So there are three types of estrogen

00:42:33.040 receptors, alpha, beta, and GPR. And this tracer that we use is more specifically looking for estrogen

00:42:40.660 receptor, alpha. Oh, interesting. So you're not using 17 beta, you're using the 17 alpha estradiol?

00:42:48.460 Yes. Oh, that's interesting. Why? I would have guessed you use the beta.

00:42:51.960 I can talk about this for a very long time. Number one is not being developed. Yes.

00:42:56.920 Even though it's the biologically active estradiol, right?

00:43:00.260 Brain. Yeah. But not in tumors.

00:43:02.440 Interesting. In tumors is the alpha. And these tracers were developed for oncology.

00:43:08.220 Ah. Yes. Got it. Okay. Yes.

00:43:10.680 Because they, of course, care about breast tissue.

00:43:12.920 Absolutely. Yeah. Of course. Yep.

00:43:14.960 Yes. Yes. So it's hard to make ligands for PET. It takes years and years.

00:43:21.880 You're working with what was developed off the shelf.

00:43:24.060 Yes. We basically repurposed a tracer that is commonly used now in oncology to see if we could

00:43:30.640 just apply it to the brain, which I think is a win-win situation because we don't have to reinvent

00:43:35.700 the wheel. So by using this reference region, the other thing that needs to happen is that

00:43:41.600 the signal needs to be the same. Let's say if you're looking at women who are premenopausal,

00:43:46.220 perimenopausal, postmenopausal, the signal in that reference region needs to be invariant,

00:43:52.180 which we demonstrated. Therefore, we were able to do a kinetic modeling using the cerebellum,

00:43:58.180 that specific part of the cerebellar cortex as the reference.

00:44:01.700 And by doing that, we show the estrogen receptor density in the pituitary gland

00:44:07.960 starts increasing during the perimenopausal window, but is actually higher after menopause,

00:44:16.560 which goes completely against whatever knowledge we had from preclinical studies.

00:44:22.820 Although, let's think about it for a second.

00:44:24.860 Let's think about it.

00:44:25.500 Okay. I don't know that I would think that that's counterintuitive because as estrogen levels

00:44:31.260 decline, you would almost expect the pituitary in a greater and greater appetite for estrogen to

00:44:37.940 upregulate expression of receptors to say, I want more, I want more, I want more. And we know that

00:44:45.000 it's screaming for estradiol because it's secreting more and more FSH and LH. So is that effectively what

00:44:52.040 you think is happening?

00:44:52.800 I think that's what's happening, but that does not happen in rodents. All the models that we had

00:44:59.180 for menopause are based on preclinical work and animal models. And what happens in rats is that

00:45:07.220 most studies utilize an ovaryectomy. So it's a surgical removal of the ovaries of the female rat.

00:45:13.980 So you induce menopause surgically.

00:45:15.800 Yes. And what the studies have found is that there is an initial overexpression of the estrogen

00:45:21.640 receptors, but then there's a sudden crash. So the window of opportunity is very narrow.

00:45:29.140 And when you translate into human ears, there's like an inverted U shape, but it's more like a

00:45:34.300 little, like a Gaussian curve. It's so narrow that within no more than five years after the final

00:45:41.420 menstrual period, the idea is that the estrogen receptors have declined to half the density that

00:45:47.040 they used to have prior.

00:45:48.600 And that's the prediction you would have had in women.

00:45:51.160 Yes, but we didn't find that at all. We found that up to age 65, estrogen receptor density was

00:45:57.900 still nice and high.

00:45:59.400 So this is totally off topic, Lisa, but I just want to park this on the side so that we can come

00:46:04.600 back to it. And hopefully between the two of us, we'll remember that what I'm about to suggest.

00:46:08.280 Would this not potentially suggest that a woman in her 60s who went through menopause 10 to 15 years

00:46:18.460 sooner, who was not treated with menopausal hormone therapy would still be a candidate given that she

00:46:27.140 clearly has upregulated her estrogen receptors in her CNS and therefore at least physiologically

00:46:35.080 suggests an appetite for estrogen?

00:46:37.500 Yes.

00:46:38.760 Okay. We'll come back to that in detail because as you know, we just keep banging on all the

00:46:43.020 greatest hits of the mantras of modern medicine, which says, even if someone has finally come

00:46:49.800 around to say maybe menopausal hormone therapy is not the worst thing you can do to a woman,

00:46:54.680 I'm being facetious, you better give it to her the day she enters menopause. God forbid we take

00:47:00.940 all of these women who are out there who were in their 60s who were deprived of hormones 10 years

00:47:06.760 ago and give them hormones. Their window is closed. The door is shut.

00:47:11.540 That was the concern. And in fact, when we were writing the protocol for this study, I said to my

00:47:17.960 team, we're going to do 35-65. And they're like, we should do maybe 55? I said, no, no, no. We're going to do

00:47:25.120 65. We're going to try and map the whole window of opportunity. And they're like, I don't think

00:47:30.280 that's a good idea. It's a little, it'll be far ahead. And when my, and I said, well, we're just

00:47:35.240 going to do it. When the results were coming in, you know, and sometimes you have a feeling.

00:47:40.700 No, yeah. Sometimes you have to trust your intuition and it's a more interesting question.

00:47:44.560 It's a more interesting question.

00:47:45.960 I think it's worth it. Maybe obviously we will concentrate around age 52, 53, but let's try to

00:47:52.780 map the extremes because we keep talking about this window of opportunity. Like we know what it is,

00:47:59.680 but it's speculative at this point because we have not been mapping it using biological indicators.

00:48:07.080 So anyway, we did it. And I think it paid off because obviously all the women in the study

00:48:13.840 are naive to hormone therapy. So no one was taking hormones of any type.

00:48:19.560 Does that mean that the pre-menopausal women obviously were off oral contraceptive?

00:48:23.800 And how long had they been off oral contraceptive?

00:48:26.300 In the very first study, it was interesting that most of them had never used birth control. It was

00:48:31.540 very interesting.

00:48:32.100 So really, really a naive population.

00:48:33.520 At least three years. That was the exclusionary criteria. And for the post-menopausal women,

00:48:40.060 they were all never users of hormone therapy. Now that we have hundreds of women in the study,

00:48:45.980 we can be more flexible and account for different things statistically or try to stratify between

00:48:52.240 past users of hormone therapy, never users, current users. We also have users now, which is very

00:48:58.860 interesting.

00:48:59.320 And what are you seeing in users?

00:49:01.660 It's not published yet. So I'm not sure that I'm allowed to talk about it. But just

00:49:06.160 anecdotally or descriptively for now, we do see that the window is shift. The curve is shifted.

00:49:13.320 So we now have women who are older than 65 and we're starting to see where the estrogen receptors

00:49:19.000 are starting to come down in terms of density. But in the hormone therapy users for now,

00:49:24.260 it seems like the curve does not stop at that age. It looks like maybe there is preservation

00:49:31.400 of density. And then the question is, is this a good thing or not? Because we don't know if the

00:49:38.400 estrogen receptors are functional. We don't know if the transcriptional pathways are still working the

00:49:44.360 way they're supposed to do. Like, are we stimulating receptors that are not functioning?

00:49:48.940 We should maybe explain to people what we mean. So let's riff off each other on this. But steroids

00:49:55.240 work by driving these transcriptional factors. So when estrogen or testosterone binds to the

00:50:01.980 receptor, what it's really doing, what matters is what it's doing inside the cell.

00:50:05.860 Yes, exactly.

00:50:06.320 It has to go into the cell. It has to go to the nucleus. It has to bind to the DNA. And it has to

00:50:11.200 say, hey, start making RNA that's going to make protein that's going to do those things. And it's

00:50:14.640 that process of transcription and translation that matters. And so what you're saying is, hey,

00:50:20.160 don't get too excited, Peter. All we're able to check with this assay is, does the hormone bind to

00:50:25.800 the receptor? The assay can't measure whether the mechanism of that is translated all the way through

00:50:31.720 to protein. Yes. Because the idea is then there's a system, there's a supply and demand system,

00:50:38.880 which is like the brain is calling for hormones and the ovaries are delivering the hormones.

00:50:44.160 As long as the feedback loop is stable, we know that usually the estrogen receptors are doing what

00:50:52.780 they're supposed to do, which is more blood flow to the brain, more energy production in the brain,

00:50:59.040 a stronger immune system, more neuroplasticity, more synaptic growth. But we also know that with

00:51:06.640 age and with disease, the estrogen receptors, as many other receptors, may start to malfunction.

00:51:13.820 They may also go through conformational changes. That means that the output may not be as good.

00:51:21.140 Say more about that. I mean, if I'm going to be honest with you, that's a terrifying thought.

00:51:24.920 What can we point to in the periphery to help us understand that, where it's easier to study this

00:51:31.020 question? I was thinking oxidative stress. Okay. Right. So estrogen, one functionality that

00:51:37.920 estrogen does is to attach itself to estrogen receptors in the mitochondria. And the mitochondria

00:51:44.420 are the energy factory of every cell in the body, including neurons. What the mitochondria do is that

00:51:51.200 they transform energy into ATP, where they take the byproduct of glucose metabolism. And there's a

00:51:59.120 structure called the elytrone transfer chain that produces oxidative stress and free radicals at the

00:52:05.960 same time that they're making ATP. Usually the balance favors ATP. But if the estrogen receptors change

00:52:15.400 that may lead to a less favorable balance where more oxidative stress is being produced relative to

00:52:24.440 the amount of ATP that is being made. So yes, there's still energy that's being produced, but there's

00:52:30.640 more oxidative stress. And this is an issue in the brain.

00:52:35.000 But couldn't there be other explanations for why we see the inefficiency of the electron transport chain

00:52:41.940 and they're like, I'm thinking of something even more basic. Couldn't we do a similar experiment

00:52:45.840 of 35, 45, 65 year old women and look at the periphery and look at mRNA expression of something

00:52:55.420 very straightforward in response to estradiol administration. So you take hormone naive women,

00:53:01.640 inject all three of them with estradiol and measure for equal amounts of estradiol,

00:53:07.400 how much mRNA gets produced for something that we would predict.

00:53:11.020 That's a clinical trial.

00:53:12.580 Yes. But according to this hypothesis, we would expect to see declining mRNA,

00:53:18.500 which would suggest at least possibly that something, and of course, to make it a really

00:53:23.520 cool study, you'd still want to do the labeling study to assume you're getting at least equal

00:53:29.280 amounts of binding. You would normalize. You would basically say, look, I'm going to take the

00:53:32.960 strength of the binding signal and I'm going to normalize it to the mRNA that comes out.

00:53:37.780 Yeah, you could if you had the money and if funding agents.

00:53:42.580 Well, not for this.

00:53:44.080 No, but that's an interesting question.

00:53:45.640 There are so many interesting questions that are not being asked.

00:53:47.520 But this is a jugular question.

00:53:49.120 Yeah.

00:53:49.560 This question implies, can we throw more estrogen at the problem?

00:53:54.740 In fact, I don't know about the periphery, but for the brain, studies have shown that timing

00:54:00.860 is really important. So if you have tissues, neuronal, tissues that are healthy, and you

00:54:10.180 introduce estrogen, estrogen is supportive of the neurons. But if the neurons are diseased,

00:54:17.600 if there's ischemic damage, there's amyloid pathology surrounding the tissues or tangles

00:54:24.640 inside the neurons, then estrogen makes it worse.

00:54:27.280 What's the evidence for that?

00:54:29.200 It's the studies that Dr. Brinton has done many years ago looking at how estrogen impacted

00:54:35.820 mitochondrial function. This is specifically mitochondria, but there seems to be evidence

00:54:41.500 for that in clinical studies as well, like the Women's Health Initiative, which you have

00:54:47.440 very elegantly unpacked. There's clearly an age-related benefit-to-risk ratio when it comes to hormone

00:54:57.140 therapy and brain health. And many people have argued that the women in the Women's Health

00:55:05.880 Initiative, the memory study component that looked specifically at dementia incidents, those

00:55:11.380 women were potentially too old to start taking therapy, hormone therapy at that age. And granted,

00:55:19.740 different formulations, higher doses of hormones, it's not what we do clinically today.

00:55:24.820 Nonetheless, it confirmed this kind of timing hypothesis, especially for those whose MRI scans

00:55:33.120 showed evidence of an existing, not the pathology necessarily, but for instance, vascular lesions

00:55:40.640 or white matter hyperintensities. In sub-analysis, the idea is that women who already harbor damage

00:55:48.820 in their brains may not be responsive to hormone therapy the same way that women with healthier

00:55:56.280 brains would be. This is completely to be demonstrated.

00:56:01.860 So two comments. The first is, do you think we have a sense of the difference between the two

00:56:06.680 variations on that theme? One variation is once disease has set in, estrogen is unlikely to reverse it,

00:56:15.480 but that's different from estrogen will exacerbate it. Yes. Let's unpack that.

00:56:19.300 It's different. The second is, as you pointed out, in the Women's Health Initiative, we were dealing

00:56:24.480 with oral conjugated equine estrogen, which is known to actually slightly increase coagulation,

00:56:32.980 which of course would be an enormous concern for exacerbating the vasculopathy that would accompany

00:56:39.320 this disease. And therefore, whereas we don't see that at all with topical estradiol, we don't see

00:56:46.660 any evidence of an increase in vasculopathy. We don't see any increase in ASCVD risk. So therefore,

00:56:53.520 we might assume that, hey, topical estradiol is much safer than, much is a strong word, but is safer

00:56:59.280 than oral estradiol. Potentially.

00:57:00.320 And that we might not see that risk. So given those two comments.

00:57:05.380 And also the progestin.

00:57:07.060 That's exactly right. The progestin as a whole, I mean, my belief still remains that if there is

00:57:11.640 some meaningful, clinically meaningful uptick in the incidence, though not mortality of breast cancer,

00:57:17.140 the progestin is the most likely culprit.

00:57:19.480 But also for vascular damage. So the MPA, the kind of progestin that was used in the Women's Health

00:57:25.760 Initiative, has later on been shown to potentially increase the risk of vascular damage. And that's

00:57:32.780 the reason we don't use it.

00:57:33.960 Yeah. So couple that with conjugated equine estrogen taken orally.

00:57:38.200 Those are getting a bad rap. They still serve a purpose. I would like to see more research done

00:57:44.400 on the very specific types of hormone therapy, because there are so many different options that

00:57:49.660 one can work with. And what I would really like to see is what these therapies do.

00:57:55.760 In the brain. Because everything you said makes perfect sense. But it's not been seen.

00:58:02.400 I want to see it. We can do it. We have the tools now.

00:58:07.000 In your study, obviously, you have to be clean and as neat as possible. So you have to normalize

00:58:11.240 everybody to the same point. I assume you were injecting a bioidentical. Well, actually,

00:58:16.700 did you inject?

00:58:17.640 We just inject the ligand.

00:58:19.420 Yeah, you've never.

00:58:20.420 Okay, I got it.

00:58:21.340 And then we work with the menopause clinic at Wildcarnet Medicine, which is where I work

00:58:26.540 or the OB-GYN department. And we have women who are now going on hormone therapy for menopause.

00:58:33.780 The vast majority use transdermal.

00:58:36.140 Yes.

00:58:36.820 Estradiol with or without micronized progesterone.

00:58:40.240 And that's kind of standard of care today. It's not necessarily always. But the women that we tend

00:58:46.320 to recruit for the study, obviously, like you said, need to be similar in terms of what kind

00:58:52.520 of therapy they're taking. I also want to see the CEE's compound. I would like to see oral estrogen

00:58:58.320 and the other formulations and see if we get a differential signal or not.

00:59:04.020 Yeah, I think that would be very interesting. But I think the most important question would

00:59:09.280 probably be answered through the lens of the formulation of the day, which is going to be

00:59:15.620 transdermal estradiol and, as you said, oral micronized progesterone.

00:59:21.340 I've done a bad job of navigating on this journey because I've taken us so far off the path into these

00:59:27.060 details. But look, I think you have to sort of follow your bliss. This is incredibly fascinating.

00:59:32.140 Let's bring it back up for people to kind of the surface level from the ocean floor.

00:59:37.980 We've established through this discussion that something is happening in the brain of a woman.

00:59:44.300 Oh, by the way, I meant to ask one final question on that topic. You had male controls.

00:59:48.520 Age matched?

00:59:49.480 For which study?

00:59:50.580 For the estrogen ligand study?

00:59:52.540 No, we're doing only women.

00:59:54.300 Got it. Okay, so I was going to ask. So we don't know if in a man's brain,

00:59:58.820 the estrogen ligands remain constant or I would predict will go up slightly as he ages because

01:00:04.720 his estradiol is going down with testosterone.

01:00:06.980 Yeah. So the idea is that the brain compensates for changes in estradiol levels and other hormone

01:00:12.960 levels by increasing the density of the estrogen receptors. So when usually the brain really loves

01:00:20.880 stability, the human brain is built for stability. So when hormones are fluctuating,

01:00:28.500 throughout the menstrual cycle, the concentration overall is still predictable. So the brain needs

01:00:34.600 to make very little effort to maintain a certain number of receptors. This is another thing that

01:00:40.160 is very interesting. The receptors are not just there. They just happen to be in the membranes or in the

01:00:47.100 the brain needs to make them. So it's an active process. And when estradiol levels increase,

01:00:55.580 then the brain needs to make fewer receptors. So we see this decrease in estrogen receptor density.

01:01:01.240 But when estradiol levels come down, then there is this compensatory adjustment where the brain

01:01:07.900 will overexpress or make more of these receptors in order to just grab every little bit of estradiol

01:01:14.740 that is in the circulation. The question is, when does this mechanism crash? Eventually,

01:01:20.840 estradiol levels will be permanently low and the brain is going to have to give up because making

01:01:26.920 receptors is a very metabolically expensive process. So eventually there will be a state or a stage where

01:01:34.280 estradiol is low and the estrogen receptors are low or gone. But when does that happen?

01:01:40.880 It's after 65, it seems like.

01:01:43.220 In our studies, it seems to be after 65. And what I'm trying to do now is to get more people

01:01:48.600 to also use ligand. And we're also working to make new ligands that could look at the better

01:01:54.720 receptors that can give us better signal. In other parts of the brain, like we want to look at the

01:02:00.180 hippocampus, the amygdala, the frontal cortex, the serocingal cortex, we hire specificity and better

01:02:06.560 signal-to-noise ratio.

01:02:08.520 How quantitative is your assay?

01:02:10.360 It's fully quantitative.

01:02:11.280 You know what would be so cool? First of all, how much radiation does it expose?

01:02:15.500 Very little. Very little.

01:02:17.780 Okay. How many millisieverts?

01:02:19.180 It's six millicuries.

01:02:20.740 Is that the equivalent to six millisievert?

01:02:22.920 No.

01:02:23.580 That's less. That's 0.6 millisievert?

01:02:25.640 It's less than one.

01:02:26.560 Yeah. Okay. This would be a very cool study, just out of pure curiosity. Very expensive. So you

01:02:32.520 might not do this. I would love to take a group of 35-year-old women and scan every one of them

01:02:39.740 the day they get their period. And then every five days for 30 days.

01:02:44.360 Oh, yes.

01:02:45.160 And just, because that's your natural experiment of exactly what you just described. That is going

01:02:50.900 to be the absolute highest, absolute lowest level of estrogen and progesterone in the brain

01:02:56.480 in a 30-day window. And the fact that it's quantitative means you can now really develop

01:03:02.540 a sense of how quickly can this compensation occur and what's the highest high and the lowest low.

01:03:08.300 Yes. And then compare that to estrogen levels in blood.

01:03:12.320 Yes.

01:03:12.880 I think it's so important to clarify that estrogen levels in the circulation have nothing to do or

01:03:19.780 very little to do with estrogen levels in the brain.

01:03:22.500 Really?

01:03:22.760 Yes.

01:03:23.500 Oh, okay. So say more about that.

01:03:25.380 That's the problem that we're having, I think, clinically, is that we can measure estrogen in

01:03:30.560 blood. But that will tell you nothing about whether or not you're having heart flashes or

01:03:36.940 forgetfulness or any of the neurological symptoms of menopause because...

01:03:41.480 Because we don't know the receptor density.

01:03:43.120 We don't know the receptor density. And also, the brain levels of estradiol are very highly regulated.

01:03:51.320 And it's basically all the hormones in the brain are sheltered from changes in the circulation.

01:03:59.060 So these transporters are active.

01:04:00.780 They're active. Yes. The brain calls for hormones.

01:04:04.160 Let's talk more about that. I thought this was like a passive diffusion.

01:04:07.340 No, not necessarily. There are periods of time where it could be and times where it's not,

01:04:13.480 which is why you can't just push stuff inside the brain. It's so hard to get a tracer that goes in

01:04:20.360 because the brain doesn't want a lot of molecules, a lot of things just can't come through.

01:04:26.520 This is unbelievable.

01:04:27.820 Yes. It's so difficult to come up with this brain.

01:04:30.720 So you're telling me that if we did my thought experiment of every five days or every day,

01:04:35.840 it's just a thought experiment. Every single day you draw a woman's blood throughout her cycle

01:04:39.880 and you're going to see estradiol go from next to nothing to 200 and back down.

01:04:45.240 You see that in blood.

01:04:46.480 Yes. That's what I'm saying. Yes.

01:04:48.120 In the brain, we don't know.

01:04:49.480 And so you're saying in the brain, it could be uncorrelated.

01:04:52.100 It could not be partially, partially correlated. There has to be a response,

01:04:56.900 but it can't be as dramatic.

01:04:59.340 So what do you think is driving it? Let me ask a question a different way.

01:05:03.560 This is now becoming a very complicated thought experiment. If you did this on a woman every

01:05:08.560 single day for a year, and she had say 12 normal cycles throughout a year, but one of those months

01:05:14.760 she had the flu and one of those months she was sleep deprived because, you know, whatever was

01:05:21.080 happening. One of those months, she was under a lot of emotional stress. One of those months

01:05:26.900 she was eating well. One of those months she was eating garbage food. One of those months,

01:05:33.420 you see where I'm going with this?

01:05:34.580 Yes. You will need a control for every single one of those months.

01:05:37.720 Of course, because they're twins. So I'm going to give you a, we have identical twins.

01:05:40.940 I see.

01:05:41.460 And so you have one of her sisters is perfectly doing the same thing every time, but also she

01:05:46.500 kind of serves as her own control in a way, right? Because

01:05:49.080 I mean, there must be a month that she's okay.

01:05:50.920 Yes. There's a month when she's perfectly okay in January.

01:05:53.420 In January.

01:05:54.280 I don't know why I think January. But do you see where I'm going? What I want to sort of

01:05:57.480 understand is how much do the externalities of her life, which obviously impact her peripheral

01:06:03.440 physiology and must impact her central physiology. How much do you think, if you had to predict,

01:06:09.140 how much would you guess those are the drivers of the brain's demand for estrogen?

01:06:15.320 Hopefully they're not the drivers. Hopefully hormonal production and hormonal demand is skewed

01:06:24.700 mainly by hormonal needs for the brain, because if that weren't the case, we would have a lot

01:06:32.000 of trouble thinking straight. And I think that's one of the reasons that the brain really very

01:06:38.640 tightly regulates entry of nutrients or chemicals from the circulation, because if the levels of

01:06:48.600 receptor activity were to fluctuate too quickly or too frequently, that could lead easily to cognitive

01:06:56.680 impairment or to mental confusion or to an inability to just function. But at the same time, it is

01:07:04.040 important. Your lifestyle has an impact. I don't know that the impact would be visible on a month

01:07:09.740 by month basis, hopefully not. But over time, the poor lifestyle, sleep deprivation, high stress levels,

01:07:18.680 that in theory would negatively impact the brain itself, making perhaps the receptors are not as

01:07:25.980 functional as they used to be. Or estrogen uptake is not as tightly regulated or carefully planned as it

01:07:32.680 used to be. And then there could be glitches that are more long-term. Does it make sense?

01:07:38.600 It makes sense. It basically says we are only at ankle deep water at this point in terms of our

01:07:45.460 understanding of this process.

01:07:47.080 Yes. Yeah. It's really just the beginning. And this is a reason that I launched CARE,

01:07:53.160 which is my new program of research. Can I mention?

01:07:55.680 Of course.

01:07:56.040 So I just launched a $50 million program of research sponsored by Wellcome Leap,

01:08:02.340 which is an independent subsidiary of the Wellcome Trust. It's called CARE,

01:08:07.040 Cutting Alzheimer's Risk Through Endocrinology. As I mentioned, it's $50 million unrestricted,

01:08:12.640 which is amazing for this specific question. And CARE is effectively the largest research program

01:08:20.820 on women's brain health, menopause and Alzheimer's disease ever attempted. And what we're doing with

01:08:27.340 CARE is, it's like the movie Oppenheimer. I think people are familiar with that movie where Dr. Oppenheimer

01:08:34.060 was in charge of designing a research program and then basically inviting other scientists from all over

01:08:41.820 the world to work with him on a sprint. It's called a sprint. So it's a three year, just three

01:08:48.420 years. It's a high risk, high reward research initiative, which for them ended up with the

01:08:54.820 atomic bomb. With us, we'll hopefully end with a means to half the risk of Alzheimer's disease for

01:09:02.320 women by the year 2050. That is our target. We estimate that if everything goes according to plan

01:09:08.860 and we hit all our marks, then we should be able to reduce the risk of Alzheimer's for an estimated

01:09:14.520 330 million women globally. And given current global conversion rates to Alzheimer's, we could

01:09:21.780 potentially prevent 55 million new Alzheimer's patients among women in the next, hopefully, 25 years.

01:09:31.080 And one of the things that we're doing with CARE is, so we have three different components.

01:09:36.620 Trust One speaks a lot to what we've been discussing so far. We want to understand how neuroendocrine

01:09:44.560 aging and specifically hormones really speak to Alzheimer's risk for women because all the

01:09:51.240 predictive models that we have so far are sex aggregated. So we look at things that risk factors

01:09:57.480 that work for men and women, but they're kind of genderless. The fact that sex has been removed

01:10:03.500 statistically, but there are things. How is that even possible?

01:10:06.640 That's the vast majority of models. So when we say that we do talk about this now, that Alzheimer's risk

01:10:14.320 is multifactorial, but potentially preventable in about 45% of cases. The 45% comes from studies that have

01:10:23.040 looked at all sorts of risk factors in cohorts that combine men and women. And more often than not,

01:10:31.560 the predictive models adjust for sex as a covariate. So you want to statistically remove the effects of

01:10:39.580 sex and see whether diet is associated with Alzheimer's risk.

01:10:43.720 We can't get the raw data and reinsert?

01:10:46.480 Yes. This is what we're doing with CARE.

01:10:48.140 Okay. All right. Okay, good.

01:10:48.820 Yes, yes, yes. So whatever we know about Alzheimer's risk so far is genderless. It works for men and

01:10:55.040 women, which is still, it's a wonderful start.

01:10:56.540 To give you one example, you're saying when we sit around and say that having an APOE3 and an APOE4

01:11:03.180 gene increase your risk of Alzheimer's disease by about 2x, that should be more nuanced. We should

01:11:09.160 say in a man, it increases it by x, in a woman, it increases it by y.

01:11:13.880 Sixfold.

01:11:14.620 It's a sixfold increase. So that's something I did not know. We usually talk about this,

01:11:18.960 at least I usually talk about it, without differentiating between sex. Clearly, that's a

01:11:22.420 mistake. So if we're talking about comparing people who are heterogeneous, so one copy three,

01:11:29.420 one copy four, versus two copies three, what's the relative risk increase for women specifically?

01:11:36.260 So women who are heterozygous for the APOE4 allele have a fourfold increase in dementia risk as

01:11:44.120 compared to non-carriers. But women who have two copies of the E4 allele, then the risk is between

01:11:51.400 12 and 15 times higher relative to non-carriers. Yeah. So this is about twice the risk of men.

01:12:01.240 Yeah. Yes.

01:12:02.040 Okay. Let's now talk a little bit about the role that hormone replacement therapy or menopausal hormone

01:12:10.220 therapy can play in women with or without an E4 allele. Let's also talk about it in the context of

01:12:20.160 initiation of therapy at an appropriate time just to start. So what do we know about this?

01:12:25.520 Oh, this is a tortured part of our field of research because unfortunately there's only one

01:12:33.160 clinical trial that's ever looked at hormone therapy and Alzheimer's incidence in women. And

01:12:39.740 that's the Women's Health Initiative memory study. Which of course had a lot of hair on that dog.

01:12:44.540 Yeah. Yes. In that specific study, the risk of dementia was increased for women who were taking

01:12:53.660 the combined estrogen-progestogen therapy, which in that case, like you said, it's oral conjugated

01:12:59.940 equine estrogen and MPA. And the risk was also 50% higher for women who were taking only oral

01:13:07.900 estrogens following a hysterectomy. However, that risk increase was not significant. Now,

01:13:14.720 that's the only clinical trial we have looking at the incidence of Alzheimer's disease relative

01:13:20.120 to hormone therapy use. And all the women were postmenopausal and by a long shot. Unfortunately,

01:13:26.680 we don't have clinical trials where hormone therapy is given in midlife for relief of menopausal

01:13:35.000 symptoms, which is the appropriate indication, where we also measure the incidence of dementia

01:13:40.180 because those trials are just not feasible. It would be like a 20-year, 30-year trial and it's

01:13:47.360 just not possible to do that way. In that case, observational research offers more information

01:13:55.180 about whether there is a differential beneficial effect relative to initiation timing. And we do know

01:14:02.420 that observational research is subject to bias. So this is just more descriptive than definitive.

01:14:09.580 However, it is interesting that meta-analyses, which are statistical integration of all available

01:14:16.480 data, do show the timing of initiation matters and also the type of formulation. So women who do not

01:14:24.440 have a uterus, who have received a hysterectomy, which is the surgical removal of the uterus with or

01:14:29.640 without the ovaries, are typically treated with estrogen-only therapy. They don't have to. You

01:14:35.980 can also have a progesterone, but generally practice says, let's just go with estrogen. Whereas women with

01:14:42.580 a uterus need a progesterone, whether a synthetic progesterone or micronized, or what people say

01:14:50.000 bioidentical progesterone. Now, if you look at these two factors, when you start, which is within 10 years

01:14:56.860 of the final menstrual period, or over 10 years of the final menstrual period, and whether you have a

01:15:02.320 uterus or not, this is what the observation of research so far shows, a 32% reduced risk of

01:15:11.600 Alzheimer's or dementia for women with a hysterectomy who have undergone a hysterectomy and

01:15:17.180 they're taking estrogen-only therapy. Significant, very consistent risk reduction across all the

01:15:24.020 studies available. Almost all the studies. We have now one from Northern Europe that does not show

01:15:29.220 that protective effect. This is when hormone therapy is initiated within 10 years of the

01:15:35.980 final menstrual period. For women with a uterus starting hormone therapy within 10 years, there's

01:15:41.700 a 23% risk reduction, which is, however, a trend level. Some studies show an increased risk.

01:15:48.620 Most studies show a reduction in risk. So we need to better understand what's happening there.

01:15:55.740 When we look at starting hormone therapy more than 10 years after the final menstrual period,

01:16:01.400 there is no obvious benefit for estrogen-only therapy for women with a hysterectomy,

01:16:07.140 and there is an increased risk for women with a uterus who were taking estrogen and progesterone of any

01:16:14.780 type. We do not yet have enough studies. I'm preempting your question. I can see it forming.

01:16:20.380 We can't yet separate progestins from micronized progesterone. We cannot yet look at each specific

01:16:29.720 type of hormone therapy because the data is just not there. We need to do more research.

01:16:35.080 Also, I will add, this is a fairly old-fashioned way to look at this question. I would argue that

01:16:43.600 today with the tools that we have and the data that we're able to collect, what would make more

01:16:49.140 sense, and I'm sure you were going to say, wouldn't it be better to, yes, it would be better,

01:16:53.200 to start hormone therapy today and look at biological markers of Alzheimer's as the therapy

01:17:00.880 is progressing. In fact, what would be ideal is to start hormone therapy today, look at the estrogen

01:17:07.940 receptors, use brain estrogen imaging to monitor whether the therapy is doing what it's supposed

01:17:14.620 to be doing, and also look at biological markers of Alzheimer's to make sure, or at least to test,

01:17:22.120 whether they're either not showing up or they're being delayed in their progression or evolution

01:17:29.140 relative to a placebo group, which is what we're trying to do now with care.

01:17:33.820 Fantastic job anticipating both questions I had and saving me from answering them. Does this mean

01:17:39.980 that one of the initiatives in care is actually a prospective randomized trial that will administer

01:17:49.060 MHT at the appropriate time during perimenopause? I have a whole soapbox on why we have to start this

01:17:56.960 in perimenopause. You don't wait till menopause, and then we prospectively follow the various markers?

01:18:06.080 Half is yes. So we only have three years. Running a clinical trial in three years is just not feasible

01:18:12.460 from start to finish. So what we're going to do in the three years, our goal is to provide evidence

01:18:18.660 a convincing scale that hormone therapy has or doesn't have, but we're hoping it might have,

01:18:25.140 a beneficial effect on biological markers of Alzheimer's by working with women who spontaneously

01:18:31.500 decide to start hormone therapy. So you don't have to enroll and randomize. You follow women who do

01:18:39.260 go on hormone therapy and women who do not. And that's what you can do in three years.

01:18:44.980 How will you match them for health consciousness? There's an inherent, I think, bias that slips into

01:18:51.560 women who opt into hormone replacement therapy because the barrier to entry is high. Most women

01:18:57.760 who want hormone replacement therapy are going to face an uphill battle with their doctor who,

01:19:03.460 no disrespect, but their doctor is ignorant, busy, just believes it's bad. And so the women who

01:19:10.140 ultimately end up on hormone replacement therapy had to jump through a few hoops. They're also probably a

01:19:14.740 little bit more health conscious on average because they're willing to go through the brain damage of

01:19:19.580 having to beat the system and fight and make sure that they can get what they rightly deserve.

01:19:25.020 Whereas the woman who says, I'm not going on hormone replacement therapy, it's less of an

01:19:29.240 active decision. It's probably more of a passive decision. Now, of course, I can come up with some

01:19:33.520 examples. Maybe you have women with a very strong family history or a personal history of breast

01:19:37.840 cancer that might be equally health conscious, but they decide to opt out for reasons that have to do

01:19:43.800 with that. So that might be one way to match them. But otherwise, you have to be very careful with

01:19:47.460 this type of analysis. Yes. Because the healthy user bias runs deep. Yes. Yes, absolutely. It's one

01:19:53.520 of the concerns with observational research. The other concern is that women with the most symptoms

01:20:01.080 are also more likely to start hormone therapy. And there seems to be an association between

01:20:06.860 more hot flashes, like the severity and frequency of hot flashes, especially at night, and amyloid beta

01:20:15.940 levels in plasma and white matter hyperintensities in the brain. So there's a few things that we need

01:20:21.460 to ascertain. And I think the way to do it is number one with education. And I think that a lot of my

01:20:28.640 colleagues and you and your colleagues are doing a fantastic job of making sure that women understand

01:20:35.060 the hormone therapy is on the table. And then they can come to us, the research sites, where they also

01:20:42.740 have access to the gynecology department, to the menopause clinics, where our clinicians are open

01:20:50.180 to the notion that if you do have the symptoms of menopause, and if you're willing, interested in

01:20:56.340 starting hormone therapy, that is perfectly doable. So I think by working with us, certainly there shouldn't

01:21:02.920 be much of a barrier to access hormone therapy, if indicated. We do follow professional guidelines.

01:21:10.100 Is there any woman who is in that perimenopausal transition who you think needs to be cautious

01:21:16.700 of hormone replacement therapy as far as her brain health is concerned?

01:21:20.660 I think the concern about starting hormone therapy before menopause is that hormone levels are

01:21:28.200 fluctuating. Using birth control, for instance, by blocking ovulation will make sure that once you do

01:21:36.060 go on birth control, you receive a standard dose of hormones that have been tested, and that can be

01:21:41.760 clearly monitored, right? You know exactly what dose you're given at any given time. Whereas with

01:21:48.080 hormone therapy for menopause, depending on what you're doing, you can't really do a blood draw every day

01:21:53.460 or every moment. So the concern is, once your level of estradiol are low enough, then we should be

01:22:02.040 fine. But if they're spiking, and you put more hormones in the systems, you may amplify the spike?

01:22:08.440 Our view is that once women are symptomatic, so we're not doing this based on blood, but women can be

01:22:15.380 symptomatic for a year, six months, or three years. There's just so much variability in this system.

01:22:21.680 But there's evidence that as soon as they're symptomatic, both from an estrogen and progesterone

01:22:27.800 standpoint, although the estrogen tends to be the symptoms that dominate, the benefits accrue

01:22:32.960 immediately with respect to vasomotor symptoms, bone health, cognitive performance in the short term,

01:22:39.460 sexual health, and sexual function. And that waiting until a woman has completely stopped,

01:22:46.320 her FSH is 40, and her estradiol is unmeasurable, at which point everybody would say,

01:22:51.680 yep, she's in menopause. You could spend five years getting to that point from the moment you

01:22:56.800 started having symptoms. And again, there's evidence that you've actually taken steps backwards with

01:23:01.840 respect to health. Now, what's the trade-off? The trade-off is you're going to have loops,

01:23:06.220 meaning you're going to have ovulations that force their way through the system,

01:23:09.960 and you're going to have all sorts of estradiol spikes. I can tell you that clinically,

01:23:15.440 most women are far less bothered by this than the reverse.

01:23:19.960 At the end, it's really about feeling better.

01:23:23.320 To your point, this requires nuance. This is not a set it and forget it policy.

01:23:28.560 This is something where, you know, and I had Rachel Rubin on the podcast.

01:23:31.840 This is where doctors like Rachel matter because they understand how to titrate the system.

01:23:38.220 They understand, oh, you know what? Even though normally you might put somebody on 200 milligrams

01:23:44.180 of micronized progesterone when they're fully in menopause, you might only need 50 milligrams today,

01:23:49.660 and you might only need 100 milligrams next year. So it's not an on-off switch.

01:23:55.160 It's more like a precision medicine type of approach.

01:23:57.980 It would be so good to have more research happening in parallel because I noticed that

01:24:04.380 there are many clinicians who are now open to working with their patients to address the needs

01:24:11.500 of the patients and kind of base what they're doing on their own experience and their relationship

01:24:16.920 with the patient. I think for them, and I know a lot of them, I'm friends with many of them,

01:24:21.440 what I think would be lovely to have is data that really works in parallel. So you have

01:24:28.320 maybe not clinical trials yet, but at least some information that can help guide the diagnostic

01:24:37.800 process. And the challenge is there's not a natural owner to doing that study. It's not as simple as

01:24:44.480 here's the latest version of a GLP-1 agonist, and we're going to go out and we're going to test

01:24:49.480 whether it's more or less efficacious for weight loss or type 2 diabetes, where

01:24:53.200 there's an obvious sponsor for that research. Here, we're talking about drugs that are cheap

01:24:59.940 and not protectable. One thing that is probably interesting is the selective estrogen receptor

01:25:06.340 modulators. Yeah. So let's talk about CIRMs a little bit. CIRMs are very interesting compounds

01:25:12.580 because like we were talking about before, there are different types of estrogen receptors,

01:25:18.860 at least three types that we know of, the estrogen receptor alpha, beta, and GPR. And they are

01:25:25.500 distributed differently in different parts of the body. For instance, within the brain, we have some

01:25:32.880 structures like the pituitary gland and the hypothalamus that contain similar amounts of estrogen

01:25:39.020 receptor alpha and beta, but are predominantly alpha because they're more reproductive tissues.

01:25:44.600 So the alpha receptor is more abundant or more expressed in reproductive tissues, whereas the

01:25:51.340 beta receptor, for instance, is more expressed in the cognitive parts of the brain. So what some

01:25:58.100 scientists and clinicians have been trying to do is to develop compounds that selectively attach

01:26:07.020 themselves to the beta version of the estrogen receptors. And for instance, I'll go back to

01:26:13.240 Robbie, Dr. Robbie Brinton. She developed what I believe to be the first neuroserm. So it's a

01:26:21.540 neurological selective estrogen receptor modulator that comes from plants, actually. She looked at all

01:26:28.560 different phytoestrogens and compounding them together into a formulation that's been shown to bind with

01:26:37.840 very high affinity to estrogen receptor beta specifically. And that means that, at least in

01:26:46.560 animals, she tested that very thoroughly in animals and she did a phase one clinical trial. We're now doing

01:26:53.180 a phase two phase two to be in women. But what she found is that this specific substance leaves your

01:27:02.000 reproductive organs alone, but goes up into the brain and binds to the estrogen receptor beta with high

01:27:09.360 affinity, therefore stimulating cognition in terms of memory, for instance, or executive function. And also

01:27:18.500 supports mitochondria activity because she's done a lot of work on mitochondria and seems to improve

01:27:24.460 neurogenesis as well. That's pre-clinically. So we're not looking at whether that formulation can

01:27:31.960 improve brain energy levels in women and hopefully memory performance. And also we are hoping reduce the risk of

01:27:42.420 developing Alzheimer's plaques. And we are working with women who are very early postmenopausal.

01:27:47.800 Is this pre-IND or is this in phase one yet? This is phase two, but it's considered a supplement.

01:27:55.600 So the FDA considers this a supplement. Because it comes from plants. Yeah. What's the name of it?

01:28:01.960 Phytocerm. Phytocerm. Yes. To be clear, it's a GRAS approved FDA supplement? Yes, it is approved by the

01:28:09.580 FDA. So the purpose of these studies is to be able to make claims. It's not regulatory. It's not

01:28:15.460 regulatory. They have done all the regulatory phase. We are now doing a clinical trial very

01:28:21.060 specifically to test whether it supports cognitive function and brain energy levels in women at risk

01:28:28.220 for Alzheimer's. And Rob is also looking at half flashes and visomotor symptoms in a separate clinical

01:28:36.280 trial. The reason I brought it up is that I think is a fairly unexplored avenue for support of brain

01:28:45.040 function and also for relief of menopausal symptoms in women. Because it makes sense to go for the source

01:28:53.700 of the symptoms that they have. Flashes, their eye sweats, their insomnia, mood symptoms,

01:29:00.000 cognitive symptoms, they start in the brain. So it would be wonderful to have a molecule that's never

01:29:07.120 been associated with an increased risk of cancers to any reproductive organs and just goes into the

01:29:13.520 brain. It does what it's supposed to be doing in the brain. But then we miss the activity in sexual

01:29:20.100 organs. We miss the bones. I mean, there's still so many benefits. Well, not the bones. I think that

01:29:25.480 estrogen receptor beta in the bones. But you're not going to get it with a CIRM.

01:29:29.980 Not with this specific CIRM. No. But there may be other CIRMs that are developed in the future.

01:29:35.440 But we have the perfect one. It's called estradiol. Why are we afraid of this? I mean,

01:29:39.560 I think this is the problem, right? We have to stop giving the fear-mongering people an excuse,

01:29:45.260 which is there is no evidence that estrogen causes breast cancer. This is a fallacy. This is a

01:29:51.380 complete fallacy. The Women's Health Initiative data by itself makes it very clear. Not a single

01:29:56.620 additional woman died of breast cancer as a result of taking even the conjugated equine estrogen.

01:30:02.300 If there was any increase in incidence, but not mortality of breast cancer, it was due to the MPA,

01:30:09.200 which again, how many women take MPA today?

01:30:11.740 Today, no.

01:30:12.540 Yeah, exactly.

01:30:13.000 It's been discontinued now.

01:30:13.880 So we have no women taking MPA today. Women are all taking bioidentical micronized progesterone.

01:30:19.140 And I want to be very careful that I never let someone have that out of saying,

01:30:25.340 but estrogen causes cancer. It doesn't. There's no evidence that it's causing cancer.

01:30:29.400 Yeah. I think the word cause has been misleading women for a really long time.

01:30:34.080 It has.

01:30:34.540 Because the idea is that you have no risk of cancer. You have no cancers already.

01:30:39.560 And somehow you take this molecule, this estrogen molecule, and boom, you get cancer. That's not

01:30:44.780 what's happening. No, but that's what, that's what women are being led to believe.

01:30:49.180 Yeah.

01:30:50.000 And again, the analogy that we should have women understand is the analogy between testosterone

01:30:55.940 and prostate cancer. Yeah, that's a good one.

01:30:58.580 So it was a very good analogy because it has been unequivocally demonstrated that testosterone,

01:31:04.780 either endogenously or given exogenously, does not drive prostate cancer. Does that mean

01:31:12.660 that when we have a man for whom we're treating him for prostate cancer, if he's not a surgical

01:31:18.100 candidate, that we don't do androgen deprivation therapy? No, of course not. We do androgen deprivation

01:31:23.220 therapy. But once a man has surgical therapy for his prostate cancer, guess what? We resume

01:31:29.940 testosterone replacement therapy. He had prostate cancer. We're giving him testosterone. But guess

01:31:34.980 what? Doesn't increase his risk. So here's why I believe that we're so brain damaged on this topic.

01:31:40.780 The urologist has a marked advantage over the clinician who treats breast cancer. And it comes

01:31:47.940 down to a very simple protein called PSA. It's the PSA that gives the urologist and the urologic

01:31:55.800 oncologist a marked advantage, which is we can always follow PSA. So when you have a man who has a

01:32:04.940 Gleason 3 plus 3, which is a cancer, and you're trying to decide, okay, he has prostate cancer,

01:32:11.800 but is it the kind that's going to kill him? Or is it the kind that's just going to stay in his gland

01:32:16.220 and stay localized? You watch and wait those men. We don't operate on a Gleason 3 plus 3,

01:32:22.380 even though it's cancer. But do we chemically castrate that man? Not a chance. If his testosterone

01:32:29.320 is 900, we rejoice. If his testosterone is 300 and he's feeling symptoms of hypogonadism,

01:32:37.980 do we give him testosterone? Absolutely we do. And we follow the PSA and we follow the MRI. And if we

01:32:45.760 need to do a biopsy, we do a biopsy. And if his cancer changes, we treat him. I think it's the fact

01:32:51.200 that we don't have the equivalent of the PSA for breast tissue. And in fairness, and in fairness to

01:32:57.420 those who have to make these decisions, we miss the blood biomarker that allows us to cheaply and

01:33:02.380 easily track the disease. But that doesn't change the underlying pathophysiology. I have nothing

01:33:08.020 against CIRMs. I didn't mean to get on my soapbox, but I don't want women to come away from this

01:33:12.760 discussion thinking estrogen is bad. Oh my God. They should be coming away with the opposite,

01:33:17.520 which is estrogen is very important for their brains. Estrogen is certainly on the table. And I

01:33:22.200 think that we have put ourselves in a difficult situation where now we need to re-educate not just

01:33:29.020 the patients, but also the entire medical and scientific community based on newer data.

01:33:36.920 I don't know how this really happened, but we have been stuck with the Women's Health Initiative

01:33:41.680 for decades. It doesn't happen in other fields of research, don't you think?

01:33:47.680 No, it did. This has happened, I think, in other fields of research. Yeah, sure. Think about the

01:33:52.580 literature or think about the phobia around dietary cholesterol.

01:33:55.960 Oh, okay.

01:33:57.080 Think about how much you can't eat the yolk of an egg, you can't eat shrimp.

01:34:02.680 Yes, I was thinking.

01:34:03.460 Dietary cholesterol raises cholesterol in the blood that causes heart disease. I mean...

01:34:08.000 But did it all come down to one trial? Because this is just one trial.

01:34:12.000 That's a fair point. You're right. In the case of dietary cholesterol,

01:34:15.400 it came down to a couple of epidemiologic studies, a couple of clinical trials,

01:34:23.580 none of which asked the question exactly. But the public was either misled or confused about

01:34:31.980 the difference between the chemical structure of dietary cholesterol, which is esterified versus

01:34:36.600 non-esterified endogenously produced cholesterol. And the story is much more complicated.

01:34:41.600 To your point, though, the Women's Health Initiative cost more than a billion dollars 25 years ago.

01:34:47.700 Yeah.

01:34:48.400 And therefore, it's not going to be replicated. So it was a study that had at least three or four

01:34:55.020 fatal flaws in the design, some of which I think are justifiable. The most fatal flaw in my mind is

01:35:01.260 they just used a garbage formulation of estrogen and progestin.

01:35:04.500 But that's what they had.

01:35:05.600 Exactly. And it's not the only thing they had, but they made the decision to use...

01:35:09.940 Yeah. They made the decision to use what doctors were prescribing most frequently.

01:35:14.240 I think that's a forgivable mistake. The real fault lies with the PIs and the media,

01:35:20.620 who I believe very nefariously promoted a false agenda. Again, when was the last time the NIH did a

01:35:28.240 press conference on a study like, really? Press conference? Post-game analysis? Really? That's

01:35:33.680 what we're going to do? That's strange.

01:35:35.140 Very strange. And I think there are certain members of the media, who I will refrain from

01:35:40.320 naming, who simply got it in their mind that this was the way it was, and they have been

01:35:46.820 completely immune to any form of logical intervention to show them otherwise. Even when

01:35:53.480 seven and even 19 years later, subsequent analyses find the same thing. 19 years later, we see the

01:36:02.720 same... And by the way, to my knowledge, that's the most recent publication. Maybe there's something

01:36:06.120 even newer. But 19 years post, we're seeing not one additional woman died of breast cancer

01:36:12.320 in the CEE plus MPE group relative to placebo. And that's even with the MPA. Where's the press

01:36:21.540 release on that?

01:36:22.460 I know. I know. There's a tendency, I think, to amplify the negative results as opposed to the

01:36:28.600 positive results. Like, even with the association with Alzheimer's. There are a few studies that

01:36:33.440 came out of Northern Europe showing that hormone therapy, and those are retrospective studies

01:36:41.440 that looked at women who have Alzheimer's today and were taking hormone therapy starting before

01:36:48.200 the Women's Health Initiative crash. And in those studies, there is an association between taking

01:36:54.460 hormone therapy at any age and an increased risk of dementia. There are two studies. They're all

01:37:00.500 over the news. All over the news. Hormone therapy causes Alzheimer's.

01:37:04.720 Right, but the 10 studies that show the opposite, it's crickets.

01:37:08.900 Yes. Yes. There are studies with like half a million women from the United States where, by the

01:37:14.400 way, the vast majority of studies shows a protective association that just never make the news.

01:37:20.960 And that, I find, is a pity because it leads to a very unbalanced conversation where so many women

01:37:26.840 today are like, oh, I need to go off the hormone therapy because of this observational study that

01:37:32.980 is a correlational study, right? Not even a trial. But I think it's so hard to disentangle scientific

01:37:39.060 information when it's a headline. You're hit by the headline. And it's very hard to really

01:37:46.040 understand the context and the nuances and just the fact that even in these studies that make the

01:37:52.340 headlines, those women were taking hormones before the Women's Health Initiative crashed.

01:37:57.500 You can see that they start with a certain number of women in the study and then the number just

01:38:02.560 plummets, right? So at the end, you're left with this subpopulation of women who just happen to be

01:38:09.600 still on hormones. I won't even pretend to disentangle that I could predict the biases

01:38:15.880 that are inherent in that type of a study. Observational research is hard.

01:38:21.320 That's like the worst example of observational epidemiology. I couldn't disentangle that.

01:38:26.180 Like I couldn't even tell you what corrections you need to make. I think your intuition a few

01:38:30.500 moments ago was the right one, which is the only way we're going to get better data on this.

01:38:34.480 It's by generating the right data.

01:38:36.340 We don't need more observational epi on this question. What we need are better and better

01:38:41.140 biomarkers that allow us to do more rigorous prospective randomized control trials. We need

01:38:47.240 RCTs and we're not going to get hard outcomes because it takes too long and it's going to cost

01:38:52.420 too much. But we could look at C2N. We could look at P-tau. We could look at any of the other brain

01:38:58.400 metabolomics. We could look at so many things. And so I guess my question for you is clinically,

01:39:05.440 I mean, most people listening to us aren't going to have access to an estrogen tracer. They have

01:39:11.200 to be in a clinical trial.

01:39:12.300 Or in a study suite.

01:39:14.240 Yeah. Yeah. So clinically, what do you think of C2N? What do you think of the commercially

01:39:20.340 available versions of P-tau and these other studies? Do you think that these are ready for

01:39:24.660 prime time? Do you think that-

01:39:26.340 For prevention or a diagnosis?

01:39:27.340 Yeah. Do you think that physicians and women could use

01:39:30.180 these as tools to track interventions that they're making?

01:39:33.940 Oh, to track interventions. Oh, yes. Yes, totally. This is what we're doing with care.

01:39:38.820 Okay.

01:39:39.020 So we're using these specific-

01:39:40.940 You're using commercially available assays.

01:39:43.040 Sometimes. We do have the machines by Dismosa. And so we are, our centers, our sites,

01:39:50.280 basically, they run their own assays, but it's the same machines that are used by commercial

01:39:56.920 entities. And we are using those as surrogate outcomes of Alzheimer's risk. We're using not

01:40:03.220 just the brain scans, but also the blood-based biomarkers, because they're much cheaper.

01:40:08.280 They're minimally invasive. They're easier to read.

01:40:11.920 We do need to have more long-term data to really establish their predictive value for each

01:40:21.120 individual. Right now, they're being used either diagnostically, which I think is very smart,

01:40:27.520 or for research to try and better understand if we can use those markers to really predict

01:40:34.360 who is at risk and who's not, and what is the positive predictive value or negative

01:40:39.440 predictive value for any given individual. And I think studies like care and other prospective

01:40:45.720 cohort studies and large-scale biorepositories, I think they're so important, like the UK Biobank,

01:40:53.680 thousands and thousands and thousands of blood samples that can be used and analyzed for these

01:40:59.460 purposes. So one way to maximize the potential of observational research in a three-year span

01:41:07.780 is to leverage data from all over the world. Because a lot of the information we have comes

01:41:14.680 pretty much always from the same studies, mainly North America, some European studies. And the

01:41:21.920 population in those studies tends to be quite homogeneous, predominantly white individuals

01:41:28.120 with a certain level of education, overall healthy. So what we're trying to do is to get data from all

01:41:35.400 over the world. So with care, we have access to female-specific data from six continents. We don't

01:41:43.520 have it from Antarctica, but we do have data coming from six continents. And all together,

01:41:50.360 with the other scientists involved in care, we are estimating receiving data from over 20 million women,

01:41:57.260 especially longitudinal data. So that's going to be a treasure box of data for scientists who are

01:42:05.620 interested in addressing these questions. And yes, that's what we're going to do. So that's the first

01:42:10.360 component of care, is to firmly establish neuroendocrine aging and really reproductive history

01:42:16.860 for women. All these different factors that seem to emerge already at puberty, and then are perhaps

01:42:24.520 even more unmasked during pregnancy and the postpartum period, and then tend to repeat themselves around

01:42:31.540 menopause. There seems to be a continuum, like you can kind of leverage a woman's reproductive history

01:42:37.780 as a potential stress test for future cognitive decline and Alzheimer's disease, or the opposite,

01:42:45.320 cognitive resilience. This is something that has never quite been done formally and in a standardized

01:42:50.980 fashion. So we are trying to do it now. For instance, high blood pressure is a risk factor for

01:42:56.480 Alzheimer's for both men and women. It may be even worse for women. Some studies suggest, there's a

01:43:04.120 suggestion, but effectively pre-eclampsia during gestation, right, during when a woman is pregnant,

01:43:11.660 is effectively a stress test on the body. It gives you a preview of whether or not you may have

01:43:18.040 chronic hypertension when you're older. And usually, if it starts during pregnancy, it may present itself

01:43:25.120 again during menopause, and then it may remain stable. Same for mood changes. If during puberty,

01:43:33.440 your neuroendocrine system is activated in such a way that you're more likely to suffer from anxiety or

01:43:39.360 from depressive episodes or from mood changes, then things may stabilize as you get older. But then they

01:43:47.440 actually come back during pregnancy and they could be unmasked again during menopause. It's so common

01:43:55.540 for women. And we know that midlife depression is a risk factor for Alzheimer's, more for women than

01:44:01.440 for men. So we're trying to put it all in context, not just what happens to you today, but what happened

01:44:07.540 to you in the past. I believe, and I'm sure you might agree, that hormonal history should be considered

01:44:14.640 a vital sign. That was nice. Yeah. So do we know if, let's take two examples. Let's take the blood

01:44:23.040 pressure example. Hypertension is a risk factor for Alzheimer's disease in men and women. And vascular.

01:44:27.840 Yeah, vascular health in general. Men too. Do we know if when we see hypertension being a greater driver

01:44:36.060 of risk in women than in men, that we're not picking up the same underlying risk that is being driven by

01:44:42.680 something else, such as the neuroendocrine system? We do not know that. So we don't know if

01:44:46.920 independently these are all the case. Absolutely. There are very few studies that have looked at

01:44:54.580 neuroendocrine variables at all. Very few. In fact, if you look at the Lancet Commission,

01:45:01.600 for those interested in Alzheimer's disease, whenever it comes to Alzheimer's prevention,

01:45:06.620 we look at the recommendations of the Lancet Commission, which every few years produce an

01:45:13.340 update. And as of 2024, we have this risk model that accounts for approximately 45% of all Alzheimer's

01:45:22.700 cases. So modifiable risk. Modifiable risk. So there are 14 modifiable risk factors that have been

01:45:28.660 established to be meaningful and replicable by the Lancet Commission that all together account for

01:45:37.280 about 45% of Alzheimer's risk. And those are sex aggregated risk factors. They're valid for both

01:45:44.620 men and women. What is completely missing there is anything that is female-specific or male-specific.

01:45:52.640 And what they say, they have a section about menopause and hormone therapy. And they conclude

01:46:01.240 that hormone therapy may increase the risk of Alzheimer's, especially for women with an oophorectomy.

01:46:10.200 We were all like, what happened? One just has to be somewhat dismissive of these things. I mean,

01:46:14.920 I realize it's easy for you and I to be dismissive of them because we know- Oh, I'm not dismissive.

01:46:19.740 I'm really like, what can we do to change this? Because it's so important with women who do get

01:46:26.380 an oophorectomy to also be aware of the hormone therapy. No, no, no, no. But just to make sure

01:46:30.800 I understand. Yes. They're claiming that it's not the oophorectomy, that it's the hormones that

01:46:37.080 follow the oophorectomy? Well, we do know that undergoing an oophorectomy before menopause

01:46:43.280 increases associated with an increased risk for Alzheimer's. But what they say is that hormone therapy

01:46:49.180 can also increase the risk of Alzheimer's disease, especially for women- So what's the data for

01:46:55.100 that? It's a couple of studies. Yeah. My point is it's total nonsense. The studies that suggest

01:47:01.340 that if a 40-year-old or a 35-year-old woman undergoes an oophorectomy, that she's better off

01:47:06.480 without hormones than she is with hormones. I mean, I just don't believe those studies. I don't believe

01:47:11.360 those observational data. Yeah. I hear you. I mean, do you? I think observational research needs

01:47:17.020 to be cautious. But specifically those studies? Well, do I believe the studies? That's the study

01:47:22.660 that was done. Let me reframe it. If your 35-year-old sister was in a car accident or had an ovarian cyst

01:47:30.860 rupture or something like that and needed to undergo an oophorectomy, she's 35 years old. She is now

01:47:35.600 sitting in front of you in menopause. She's mechanically, chemically in menopause and she's

01:47:41.440 got hot flashes and she's got all the symptoms that a 35-year-old woman would absolutely have

01:47:46.180 in spades because of the abruptness of this. And she came to you and said, do you think I would be

01:47:50.520 better off with or without hormones? What would you suggest? With hormones? I think this is standard

01:47:56.120 of care at this point. But do you think you would be increasing her risk of Alzheimer's disease by

01:48:00.180 telling her that? No, that's why I'm so puzzled. That's my point. Okay, so that's my point.

01:48:03.860 We're in alignment. We're in alignment. I mean, whatever we know about hormone therapy is that

01:48:09.060 is especially beneficial for women who experience early menopause and especially when early menopause

01:48:16.220 is triggered by an oophorectomy. This is in professional guidelines. My only point here is that

01:48:21.900 these commissions sometimes cherry pick data to fit their agenda. That's the point. Why is that?

01:48:29.940 Why is that an agenda? Is it even an agenda? I don't know. What is concerning to me

01:48:34.480 is that neuroendocrine factors that are important for women, I think it's undeniable, are not part

01:48:40.880 of these recommendations for Alzheimer's prevention. And when they do identify...

01:48:46.080 Well, they are. They are, but they're in the wrong direction.

01:48:48.240 This one is. It's not positioned as a recommendation. It's more like we don't know enough.

01:48:53.880 I want to pivot and ask you about something else. If this is not something you think about,

01:48:57.200 that's fine. But it's something we are thinking about a lot.

01:48:59.600 Okay.

01:48:59.940 Which is, do you believe that independent of weight loss and insulin sensitivity, GLP-1 agonists

01:49:08.840 are going to have a protective effect in the brain?

01:49:11.480 I think it's really interesting and it makes sense. See, that's the thing. Biological plausibility

01:49:19.100 needs to be present for any study to be done. And sorry, I just want to complete my train of thought.

01:49:26.540 When you do observational research, which is also relative to the GLP ones, you need to have a

01:49:32.700 hypothesis that is based on something. You can't just do a fishing expedition. You need to have

01:49:39.440 preclinical work showing, for instance, that once you have a noophorectomy, estrogen is beneficial.

01:49:46.380 If you start right after the surgery, you keep taking it until the natural age of menopause.

01:49:51.280 That we know from preclinical research. That is your biological plausibility. So then you do translational research.

01:49:57.280 You power your observational study to assess a hypothesis that is based on preclinical work.

01:50:05.040 That is still observational, but it's solid. If instead your study is showing the opposite,

01:50:13.360 there's some issue there. So that is my concern when it comes to some kind of research that gets

01:50:19.000 published. It gets published because that's the data that you have. And regardless of whether or not

01:50:23.860 your conclusions are in alignment with preclinical research, you publish it anyway. That's a question

01:50:29.580 mark for me. And everybody falls back on the Women's Health Initiative. Once they find an increased

01:50:35.340 risk, it's like, well, the Women's Health Initiative found so-and-so, even though it goes against

01:50:40.600 biological plausibility. So for the GLP ones, I think there's a lot of potential. There's actually

01:50:45.740 someone I know who's developing ligands or tracers for GLP ones in the brain. And I think that's going

01:50:53.100 to be really interesting. Do you have a sense of what the mechanism of action would be beyond

01:50:58.700 two things that are important, which are metabolic? We completely, it's totally logical that even if

01:51:05.200 you have some insulin resistance absent diabetes, making that better is going to make it better.

01:51:10.120 We appreciate that obesity is accompanied, even if it doesn't come with diabetes, it's still

01:51:15.060 accompanied by inflammation. And as we reduce inflammation, yes, all those things. But the real

01:51:20.460 question I'm asking is if I took someone just like you, and I'm assuming I know I can see your

01:51:27.020 health and you're in perfect health, but you are at high risk for Alzheimer's disease.

01:51:31.560 Were they take a GLP one?

01:51:32.800 Yes. If we put you on a micro dose of trizepatide, low enough dose that I'm not going to take weight

01:51:39.340 off you because I don't want to take any weight off you. I don't want you to suffer the consequences

01:51:43.440 of sarcopenia or anything like that. Is that going to provide protection for your brain? This is the

01:51:49.300 question I want to know. This is the study I want to see done. And of course, you can only

01:51:53.520 study this prospectively with very good brain biomarkers. We're not going to be able to do this

01:51:58.440 study for 20 years and follow a bunch of women. But if we have, as I think we are, getting to the

01:52:05.060 place where we have really good biomarkers, can we start to ask this question scientifically?

01:52:10.180 We can start. The question is never interest. I think so many scientists are interested

01:52:16.360 in testing it. The problem is always funding. It's really expensive to do this kind of research and

01:52:23.400 do it well. And who is going to sponsor it? Hopefully the NIH. I mean, historically, the NIH has been

01:52:30.980 the biggest source of research fundings in the United States. This is clearly a national level

01:52:38.980 problem. It's a bipartisan problem. Hopefully this is happening. It's not my field necessarily,

01:52:46.000 but even just the fact that people are developing ligands and tracers for GLP-1s everywhere in the

01:52:51.880 world, including the brain, I think is a strong indication that there is a lot of interest.

01:52:56.120 As of today, I don't know how to answer. I really don't know how to handicap that.

01:53:02.820 The data look somewhat promising though. I have seen some unpublished data, which as you pointed out

01:53:09.380 a few minutes ago, I mean, you have to take that with caution. So there's lots of things that you

01:53:13.740 have to be mindful of, but I have seen some unpublished data that suggest that very low 2.5 milligrams

01:53:20.440 of trisepatide, as an example, is meaningfully reducing blood-based and CSF-based markers of

01:53:29.500 neuroinflammation and protein aggregation. And again, we're talking very small numbers here.

01:53:35.920 You know, we're talking like N of 20. Yeah, but you have to start somewhere.

01:53:39.440 But it's pilot. It's pilot stuff. Yeah, you're pilot.

01:53:40.660 And that's really interesting. So I wondered if you had a more informed point of view,

01:53:44.960 but it sounds like you're equally interested, but yeah. Well, Lisa, this has been a very

01:53:49.680 interesting discussion. It's really the discussion I wanted to have was kind of this intersection of

01:53:54.420 the brain and Alzheimer's disease. Unfortunately, I feel like we are still a little bit in the dark

01:53:59.480 because while it appears that the neuroendocrine differences between men and women probably

01:54:05.720 account for the majority of these differences, there may still be other things, right? We still

01:54:11.000 don't understand if women are more susceptible to hypertension, dyslipidemia, insulin resistance,

01:54:15.880 sleep disturbances, all of the other risk factors. We don't know if they're disproportionately

01:54:20.160 affecting women. And so I guess we're left with the following question. Ultimately, if you're

01:54:24.960 listening to this podcast, you might take interest in all the nuances we've talked about, but if you're

01:54:29.940 a woman, you want to know what should I be doing different? What should I be doing different than I'm

01:54:36.060 already doing? And so what is the answer to that question? If you're already paying great attention

01:54:42.080 to your sleep, if you're already paying great attention to your nutrition, if you're insulin

01:54:46.680 sensitive, if you're managing all of these things that we've talked about, is the big takeaway from

01:54:52.400 this discussion that if you're on the fence about hormone replacement therapy, it's probably something

01:54:58.660 that should be in your purview and you should probably think about it more seriously. What would you

01:55:02.680 say to a woman who's either about to go through menopause, in menopause, or just out of menopause,

01:55:08.220 who's coming at this purely through the lens of her brain? I don't know that many women who really

01:55:14.560 check all the boxes. I think whenever patients come to the clinic, to the Alzheimer's Prevention

01:55:20.520 Clinic at Walker Nut Medicine, which Richard Isaacson launched, I think it was 2013, most of our patients

01:55:28.160 come to us because they're really scared that they may be experiencing early onset dementia.

01:55:36.500 And we go through a series of tests and more often than not, these are midlife women. What they're

01:55:43.380 describing could be attributed to midlife changes, including menopause. In some cases, clearly, there's

01:55:50.620 an increased risk of Alzheimer's that needs to be mitigated. At this point, what we do for Alzheimer's

01:55:56.780 prevention is predominantly behavioral. So the ABCs, if you will, of Alzheimer's prevention are

01:56:03.500 lifestyle-based and then include managing medical conditions like high blood pressure, insulin

01:56:11.740 resistance, diabetes, obesity, all the cardiovascular risk factors. That can be done with a combination

01:56:18.640 of tools that leverage diet, exercise, stress reduction, sleep hygiene, and whatnot, and also medical,

01:56:30.240 pharmaceutical routes when appropriate. For women, specifically midlife, I do think that having a serious

01:56:39.360 menopause conversation is important, not just for the short term, not just for the symptoms of menopause,

01:56:48.620 that one may be experiencing today, but because the research is moving so fast. And I do feel, as a midlife

01:56:57.000 woman, I do pay attention to all the things you mentioned. I am extremely conscious about my lifestyle and so

01:57:05.580 disciplined. It's almost ridiculous. Like this morning, I couldn't find anything that I wanted to eat for

01:57:11.220 breakfast or just skipped it because I'm not going to have a bagel first thing in the morning. I want to have

01:57:18.340 all those answers, but we need to also wait for the research to get done so that we can then give women

01:57:27.120 the right information. It's the first principle of medicine, right? First of all, do no harm.

01:57:32.860 And I think as long as lifestyle is concerned, it is important to be very consistent. And I think a lot of people,

01:57:41.940 especially in this country, try a lot of different things. And then maybe those things don't work out. They feel

01:57:47.840 like they're not working out and they switch from a keto diet to veganism and then go back to something else. And I think

01:57:54.340 consistency is important. We do know that some very specific patterns are conducive to brain health.

01:58:02.560 And I think it's important to embrace them and stick with them for long enough time. Like exercise,

01:58:10.120 yes, but there's so many different ways to exercise. That at least we have information that is specific to

01:58:16.460 women, including women in midlife, which is the association between intensity and gains, where

01:58:23.280 gains is not building muscle mass, but more like for health, overall health, follows almost like an

01:58:29.740 inverted U shape, where moderate intensity exercise, if performed frequently enough, is conducive to the

01:58:38.900 greatest gains, which I think is the zone two. It is, but it turns out to be a little bit more complicated

01:58:45.220 in that intensity matters and duration matters. And it depends on how much time an individual

01:58:52.940 has. And so the less time an individual has, the more they have to prioritize intensity.

01:58:59.040 That makes sense.

01:58:59.580 The more time they have, the distribution curve actually skews to lower intensity. But if you're

01:59:06.440 in sort of a sweet spot in the middle, then you're probably going to get the most bang for your buck

01:59:10.780 at a modest level of intensity, which might actually be even north of zone two. I'm overdue for a

01:59:16.860 discussion on this because I feel like sometimes I talk about this and I ironically create confusion

01:59:23.160 by not being nuanced enough, which is a rare accusation for me. So I think I need to provide

01:59:29.400 a little more nuance on that. But you know what I take from what you're saying, Lisa, which I agree

01:59:33.280 with, by the way, that is the single most important thing to do is maximize the known lifestyle levers

01:59:39.140 because it buys us more time. And I am more optimistic today than I was five years ago about

01:59:46.260 treatments for dementia. Yes.

01:59:49.480 Five years ago, I was in a state of despair. I thought this is a disease that will never be

01:59:55.360 treatable. I really believe that once the proteins started folding in the brain, like there was just

02:00:00.740 nothing that could be done. And now when I look at treatments like Clotho and just full disclosure,

02:00:07.340 I'm a co-founder of a company that is trying to develop a Clotho injection. And I look at the

02:00:13.440 potential around GLP-1 agonists. And I look at the ways that we're getting better understanding of

02:00:18.060 hormones. And I look at other molecules, exercise mimetic proteins. I'm starting to think we just

02:00:24.720 need to hold on long enough. Like if you can delay the onset of something by 10 years through all of

02:00:32.040 these modifications, that could be the difference between a normal cognitive life and a cognitive

02:00:38.440 life that is cut short. And unfortunately, that's not the answer to the person who's 75 today who has

02:00:45.380 dementia. And that is tragic. And I wish I could say, don't worry, tomorrow we're going to have something

02:00:51.160 that's going to reverse this. I don't believe that personally. Maybe that makes me a pessimist.

02:00:55.740 But I absolutely believe that in a decade, we're going to have things that are going to make a real

02:01:01.780 difference. And therefore, if you're sitting here and you're 60 years old, and you can keep checking

02:01:07.180 all those boxes, as you said, that could be the difference between you being the candidate who gets

02:01:13.180 the rescue before you're fully in the throes of the disease versus not.

02:01:18.740 I completely agree. And whenever I mentioned lifestyle, and people are always like, oh, yeah,

02:01:25.540 sure. But really, it's so important. And not that many people are as consistent throughout the years.

02:01:35.160 And when we talk about brain health, I think it's important to understand that the brain is not

02:01:40.100 the same as the rest of the body. And people are so used to seeing results quickly. Within a matter of

02:01:46.240 weeks, you can lose a couple of inches, or you can grow muscle. The brain is built for stability,

02:01:53.200 whereas the rest of the body is built for change. If you want to make an impact on your brain cells,

02:01:59.960 you need to hit them frequently enough and long enough that that change is going to be recorded

02:02:06.180 as an epigenetic mutation, or as an epigenetic change, or as something structurally permanent.

02:02:12.500 And that, thank goodness, takes a really long time.

02:02:17.040 Thank goodness it's a long time on the bad end.

02:02:19.620 Yes.

02:02:20.020 But that's the price you pay on the good end.

02:02:22.460 Yes.

02:02:22.820 So the good news is it takes a while to cause damage. The bad news is it takes a while to

02:02:27.800 create resilience.

02:02:29.080 Yes, but you can create resilience, which I think needs to be emphasized that this is not

02:02:35.180 like a woo-woo wishy-washy thing. It really can give your brain cognitive resilience and brain

02:02:42.040 reserve, which is what you want in the end. You want your neurons to be strong. And you do that

02:02:48.160 by having your body, for instance, move. We know the movements produces BDNF in the brain, produce

02:02:55.340 iridin in the brain that support the health and growth of your dendritic, your synaptic extensions,

02:03:02.820 for instance. We know that if you reduce inflammation, if you reduce oxidative stress,

02:03:07.500 which is something you need to do from the outside, your brain will do better, will age

02:03:13.080 less. So these are all things that are very realistic that every one of us can do on a daily

02:03:19.580 basis. We can all make good choices that support the health of our brains in the long term or not.

02:03:26.020 And then it will show. It may not show today, but it may show 10 years from now. So this is really the

02:03:32.500 time, I think, to invest in brain health because at the end of the day, we all want our cognitive

02:03:38.700 lifespan to match our lifespan. This is what we're trying to do. And then we do research as fast as

02:03:44.600 we can. I was so happy to have the opportunity to launch care because it's very high speed research

02:03:52.820 that should deliver in three years. I was joking with my daughter that by the time she goes through

02:03:59.480 puberty, I may be going through menopause and I want an answer by then. So hopefully.

02:04:05.780 I'm really grateful for the grant you have. And I think, frankly, I did the math on this once. I

02:04:12.260 believe that the entire Manhattan Project in today's dollars was about three, four billion

02:04:18.780 dollars, which, by the way, that's a paltry sum of money for what it accomplished, right? I mean,

02:04:23.400 it changed the course of history for all its good and bad. But three to four billion dollars is a

02:04:29.660 trivial investment for the United States government. I would love to see an investment of that size

02:04:35.140 to tackle this question, because if you think about what you're going to be able to accomplish in

02:04:39.180 three years with 50 million dollars, can you imagine what a, I hate to use the term because

02:04:44.120 it's so overdone, but what a moonshot would look like here? And look, it might have to just come

02:04:49.400 from the private sector, as it has in your case, through the Wellcome Trust. But nevertheless,

02:04:53.320 I'm sure there's somebody out there listening who's thinking about what this type of a moonshot could

02:04:56.660 look like. That would be wonderful. And scientists from all over the world are ready to do this kind of

02:05:03.440 work and really, they really hope they would be able to do more. Well, thank you for taking time

02:05:08.520 away from both your family and your work to come out and visit and share your insights with us.

02:05:12.800 Thank you for your time. I appreciate it. Thank you for listening to this week's episode of The Drive.

02:05:18.600 Head over to peteratiamd.com forward slash show notes. If you want to dig deeper into this episode,

02:05:26.420 you can also find me on YouTube, Instagram, and Twitter, all with the handle peteratiamd. You can also

02:05:32.800 leave us review on Apple podcasts or whatever podcast player you use. This podcast is for general

02:05:39.340 informational purposes only and does not constitute the practice of medicine, nursing, or other

02:05:44.020 professional healthcare services, including the giving of medical advice. No doctor patient

02:05:48.860 relationship is formed. The use of this information and the materials linked to this podcast is at the

02:05:55.140 user's own risk. The content on this podcast is not intended to be a substitute for professional

02:06:00.180 medical advice, diagnosis, or treatment. Users should not disregard or delay in obtaining medical

02:06:05.800 advice from any medical condition they have, and they should seek the assistance of their healthcare

02:06:10.680 professionals for any such conditions. Finally, I take all conflicts of interest very seriously.

02:06:16.680 For all of my disclosures and the companies I invest in or advise, please visit peteratiamd.com

02:06:23.160 forward slash about where I keep an up-to-date and active list of all disclosures.

The Peter Attia Drive - January 26, 2026

#381 ‒ Alzheimer's disease in women： how hormonal transitions impact the female brain, the role of HRT, genetics, and lifestyle on risk, and emerging diagnostics and therapies ｜ Lisa Mosconi, Ph.D.

Episode Stats

Length

Words per Minute

Word Count

Sentence Count

Misogynist Sentences

Hate Speech Sentences

Summary

Transcript