The Peter Attia Drive - #407 ‒ Preventing cardiovascular and Alzheimer's disease： lowering LDL early, APOE4, and promising new therapies ｜ Michael Davidson, M.D.

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

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

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

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

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

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

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

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

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

00:00:57.480 head over to peteratiyamd.com forward slash subscribe.

00:01:03.940 My guest this week is Michael Davidson. Michael is a cardiologist, lipidologist,

00:01:08.860 and the founding CEO of New Amsterdam Pharma. He is a recognized leader in lipidology,

00:01:14.820 having coordinated more than a thousand clinical trials, published over 350 peer-reviewed papers,

00:01:20.740 and authored three books on lipid disorders and cardiovascular prevention. His work spans

00:01:26.920 statins, novel lipid-lowering therapies, omega-3 fatty acids, and cardiovascular drug development.

00:01:33.420 He also founded several biotech companies and clinical research organizations, including the

00:01:38.760 Chicago Center for Clinical Research, Omthera Pharmaceuticals, and Corvidia Therapeutics.

00:01:44.240 Michael also previously served as the president of the National Lipid Association.

00:01:49.380 Michael's career has spanned the intersection of lipid science, prevention, and drug development.

00:01:56.660 And with Obacetrapib, he is now at the center of one of the most important open questions in

00:02:01.700 cardiology, whether CTEP inhibition can finally deliver meaningful reductions in LDL, APOB,

00:02:08.860 LP little A, and ultimately cardiovascular risk. In this episode, we discuss how Michael's family

00:02:14.100 history shaped his career in lipidology and prevention, LDL as a causal driver of atherosclerosis

00:02:19.820 and why it should be treated more like blood pressure or smoking, the history of CTEP inhibitors,

00:02:25.740 why earlier drugs failed, and what makes obacetripib different, obacetripib's effect on LDL

00:02:30.840 cholesterol, ApoB, LDL particle member, LP little a, diabetes risk, and cardiovascular outcomes,

00:02:36.700 how obacetripib may fit alongside statins, ezetamide, PCSK9 inhibitors, and other lipid-lowering

00:02:42.540 therapies, the complicated biology of HDL, brain cholesterol metabolism, APOE4, and the potential

00:02:50.620 role for obocetripib in Alzheimer's prevention, and omega-3 fatty acids, DHA delivery to the brain,

00:02:58.800 AI in clinical trials, and the future of drug development for cardiovascular and neurodegenerative

00:03:03.420 diseases. So without further delay, please enjoy my conversation with Michael Davidson.

00:03:13.220 Michael, thank you so much for coming out. It is hard to believe that this is our first meeting

00:03:17.980 in person, given how many years we've worked together electronically and in other ways.

00:03:23.700 Right. Yes. Great to be here, Peter.

00:03:25.700 For folks who might not be familiar with your work, although we've certainly referenced it

00:03:29.400 a lot on the podcast, your work speaks for itself. So you certainly don't need us referencing it.

00:03:33.960 Maybe tell folks a little bit about yourself, starting with what you do clinically.

00:03:37.620 I'm a cardiologist, lipidologist. I run the Lipid Clinic at University of Chicago.

00:03:42.540 I'm a professor. So I see patients actually four full days a month, and that's a pretty loaded

00:03:48.400 prevention-focused practice. What's the path to that? Because lipidology is not necessarily a

00:03:54.840 subspecialty within cardiology, although, of course, people who listen to this podcast are

00:03:58.460 very familiar with it, and Tom Dayspring has been on a number of times. But what was your path

00:04:02.540 towards that from your training? Were you always interested in prevention? It's a passion because

00:04:07.700 my father died at age 47 of a heart attack. I was 16. I had abnormal lipids, ran into my family. So

00:04:16.120 I got very interested in lipids in medical school. So I did the original niacin trials that went back

00:04:21.880 into the 80s, late 70s actually. And then after training residency, I started doing research and

00:04:29.460 my fellowship on omega-3 fatty acids for the lipid effects. I was the first one to use fish oil

00:04:34.800 capsules to treat lipid disorders. And then from there, I got involved in all the statin trials.

00:04:40.120 It's been my passion is to be involved in clinical trials. And I still have cardiologists,

00:04:44.720 did the usual stuff that cardiologists do, and then focused primarily on prevention. So I opened

00:04:49.540 a prevention center out of my fellowship, along with a research company that did all the clinical

00:04:54.660 trials. And that's been my path ever since. And then the biotech startups started happening about

00:05:00.300 15 years ago. We have some overlap in some of those companies, one of which we'll talk about,

00:05:06.320 but maybe even before we get into some of the really exciting things going on in pharma,

00:05:11.520 maybe we can just speak a little broadly about prevention because there really isn't, at least

00:05:16.960 as I see it, a uniform consensus around how aggressively one should be trying to prevent

00:05:22.580 ASCVD. Sometimes I feel like people look at me as though I have multiple heads when I talk about

00:05:30.060 taking steps towards prevention in 30-year-olds, for example, because by any calculable metric,

00:05:35.800 their 10-year risk is very low and would not necessarily justify the steps we might take.

00:05:41.380 But you're more thoughtful. You're more astute when it comes to all these things. I'd like to

00:05:44.940 hear your point of view and how you think about primary prevention, and we can differentiate

00:05:48.940 that from secondary later on. It always is a hard sell to get a 30-year-old to start taking

00:05:54.620 a statin, for example. And so I explain the primordial prevention concept, which is basically

00:06:01.420 you want to stop plaque from forming before it's there. And it's a lot easier to stop it from

00:06:07.100 forming than it is to reversing it when it already exists. So we have this path and right now today,

00:06:13.020 which is the mainstream, is that you wait until someone has significant plaque buildup or even a

00:06:18.100 heart attack or a stroke, then you treat super aggressively to get the LDL down. But we know

00:06:24.300 from genomics that if you have a low LDL throughout your whole lifetime, it prevents

00:06:30.560 heart disease in a much greater degree. For example, I mean, if you lower LDL before you

00:06:36.260 have a heart attack, it's very effective. If you have a heart attack and you lower LDL, it still

00:06:40.420 has benefit. But once you have heart failure, there's no benefit to lowering LDL. And so the

00:06:45.720 earlier you start, the better. What I like is the eight gram rule, which is that eight grams of

00:06:52.060 cholesterol in your lifetime lead to heart disease. So that is 200 milligrams per deciliter times 40

00:06:57.420 years is eight grams, or 100 milligrams per deciliter times 80 years is eight grams, or 80

00:07:02.860 milligrams per deciliter times 100 years is eight grams. So you can think about it in that sense

00:07:07.900 that the data says that if you keep your LDL below 80 throughout your lifetime, you don't get heart

00:07:14.320 disease. So I think you said this as well. I mean, we have the knowledge to prevent heart disease

00:07:19.960 now. It's applying it earlier in life that really makes the difference. That's my pitch to the 30

00:07:27.660 year old. And there's a lot of pushback. And I try to then, if necessary, provide more information

00:07:33.300 to them, which is we can do genetic testing, we can do polygenic risk scores. If they're old

00:07:39.100 enough, we can do coronary calcium scanning or more advanced plaque analysis. We can look at

00:07:44.320 other risk factors like LPLA or CRP, things like that, which give us more information about who's

00:07:51.640 at higher risk. Of course, family history is so important. But believe it or not, I would think

00:07:56.700 family history would be one of the most powerful motivators, but it's not always the case.

00:08:01.720 It certainly was in you. I've shared my story about how rampant heart disease is in my family

00:08:06.940 and how it was sort of my foray into even thinking about all of these cardiometabolic diseases.

00:08:13.320 Your story, by the way, is just incredible in that it's identical to that of my father-in-law.

00:08:18.240 My father-in-law's father, whatever, my wife's grandfather, also died at 47 in the hands of

00:08:24.320 his 16-year-old son.

00:08:26.380 So my brother had bypassed at age 44.

00:08:29.480 And then here's the story that why I'm so passionate about early is that, so I was 16,

00:08:34.920 my brother was 14.

00:08:36.640 My father died.

00:08:37.780 My uncle, family doctor, checked our lipids.

00:08:40.040 They both were equally bad.

00:08:42.280 So I then went on to start medical school.

00:08:44.940 I took niacin in med school.

00:08:46.440 As soon as statins became available, I started taking a statin.

00:08:49.520 My brother, who's also a family doctor, went on to become a vegetarian, really strict diet.

00:08:55.260 Did not take statins until much later than at age 44 has a bypass surgery, a very significant

00:09:02.260 coronary disease.

00:09:03.680 It's an N of one, but it's a pretty good comparison.

00:09:06.900 You know, two brothers, one starts statin early, one starts statin later.

00:09:10.840 And so that decade, the 30 to 40 decade, and even maybe 20 to 30, such an important timeframe

00:09:17.440 when the plaque itself is starting to form relatively rapidly. And that's when you want

00:09:23.140 to have your most effective treatment there to prevent it. That's why I become an advocate for

00:09:28.420 earlier in life, the better when it comes to the standard theory. Now with women, we have to be 1.00

00:09:32.960 careful about the whole pregnancy issue and stuff like that. And we also still balance that as well, 0.98

00:09:37.360 but it's really upcoming, that type of information. I know people bring up

00:09:40.840 relative risk, absolute risk, 10-year risk. We're talking about more about preventing heart disease

00:09:45.760 when you're 80 or 90, not 10 years from now. Yeah. I guess the other point that for me was

00:09:51.520 a really wonderful way to think about it that I would credit to Alan Snyderman who wrote a paper,

00:09:58.420 I feel like it was about 12 years ago, but he, at least for me, was the first time somebody

00:10:03.600 repositioned the argument of rather than talk about treating five-year risk, 10-year risk,

00:10:08.760 15-year risk or trying to stretch that out, reframe it as we treat causal drivers of disease.

00:10:14.460 And if that's the case, then the discussion changes entirely. And people who listen to

00:10:18.540 this podcast have probably heard me use the analogy, but it's why we target smoking cessation

00:10:25.000 so aggressively. It's not that everyone who smokes gets cancer, and it's not that everyone

00:10:30.420 who gets cancer smokes. Those states are true. It is that smoking is causally related to cancer.

00:10:37.460 and therefore the time to quit smoking is before you start. And if you've already started,

00:10:42.760 the time to quit smoking is today, regardless of your risk going forward. So if a person has

00:10:48.520 been smoking for one year, a pack a day, they have a one pack year history of smoking. That

00:10:54.200 does not really increase your probability of getting cancer. You know, it's probably not

00:10:58.540 till you hit 15 or 20 pack years that your risk really steps up. But there's not a single doctor

00:11:04.560 on the planet that I would imagine saying, listen, Michael, I know you've been smoking

00:11:08.920 a pack a day for a year. Keep it up for another 10 years, 15, 20 at the most, but then we're

00:11:14.140 going to have to cut this out. Why? And that's an absurd example, but it gets to the point

00:11:19.360 of causality. And so in as much as LDL is causally related to ASCVD, which you pointed

00:11:26.900 out the Mendelian randomizations, let alone the clinical trials, this would be one of

00:11:30.880 the most assured parts of biology, it does become a little frustrating that it's not more widely

00:11:36.860 understood that you have to treat a causal risk factor regardless of the time horizon.

00:11:42.440 And the analogy goes, it's very good. Smoke is a really good one, but also blood pressure,

00:11:46.220 diabetes. We don't wait for stroke and heart failure to treat hypertension. We don't wait

00:11:50.900 for blindness and kidney disease to treat diabetes. So I don't know why LDL gets this

00:11:56.460 stepchild type of role when it comes to how we made it, but it is causal and we know that.

00:12:03.420 So why not treat it early as you can in life? And I think that's what we advocate for. We're

00:12:07.780 not always successful, but I think we get the message across. And like I said, we have the

00:12:11.840 tools now if we just apply them much more effectively. Why do you think that? I mean,

00:12:16.260 not that we should waste too much time speculating, but I don't have a great answer for it. So I don't

00:12:21.500 understand why we understand that nobody should be walking around with untreated hypertension,

00:12:28.060 even if they're young and even if they're asymptomatic, because most young people with

00:12:32.180 hypertension would be asymptomatic. And you could argue, well, is the anti-LDL issue really all

00:12:39.740 about drugs? Is it that people view it as sort of a pharma conspiracy or something like that? I mean,

00:12:44.940 what is your take on why this one issue around LDL has become so contentious?

00:12:51.060 Multiple reasons, but one actual reason I think is that people believe that LDL cholesterol can

00:12:57.540 be totally managed with lifestyle. If they ate one less egg or one less piece of meat or something

00:13:03.840 like that, their LDL would come miraculously down to normal. And I think what we're learning from

00:13:09.100 the genomic studies that we do a lot of genetic testing in the lipid clinic, we find that it's

00:13:13.700 almost always a genetic factor involved. And so when people realize it is genetics, there's nothing

00:13:20.340 they can do about it. They're already doing a great lifestyle and so forth. That becomes,

00:13:25.020 I think, part of the motivation to get them to do something about it. People don't like taking

00:13:30.180 drugs. I get it. It is about the statin, but we try to find other ways to get the LDL down if

00:13:35.840 necessary. But I wish there was more. Another problem we have, of course, is the primary

00:13:40.780 prevention guidelines, like American Heart, are not very effective. They don't really emphasize

00:13:46.020 as much as like you and I, if we were writing the guidelines, we'd have a very different

00:13:50.340 approach to how we manage elevated LDL. We would be advocating a much earlier intervention in life.

00:13:56.800 We just don't have those guidelines that are there to help get physicians on board as well.

00:14:01.740 It's interesting about the lifestyle point because ironically, lifestyle has a far larger

00:14:06.920 impact on managing high blood pressure, where weight loss and exercise individually and

00:14:13.020 collectively have a greater impact on hypertension, which still means there are a lot of people out

00:14:17.720 there, by the way, who don't respond, as you know, fully to those interventions and can be

00:14:21.260 exercising of normal weight and still have essential hypertension and require medication.

00:14:26.060 But I still don't see the reluctance or resistance to it the same way I do with managing dyslipidemia.

00:14:31.620 So I don't know. I don't have any great insight into it, but it does make me sad for people who

00:14:36.600 become ultimate victims of it. I see the clinical ramifications of it. It is sad because it's

00:14:42.580 preventable. We'll keep advocating. I think we're starting to see the trend happen. I mean,

00:14:47.960 how many more trials do you need to show that LDL, no matter how you lower it,

00:14:51.540 results in a clinical benefit? I think we'll talk about obocetrapin, but that might be the last

00:14:56.620 mechanism. Once that study prevails, the PREVAIL trial, we have a 10th different mechanism of

00:15:02.660 action of LDL lowering and benefit. And I think hopefully all the obstacles will go away and

00:15:08.120 people say, let's just start treating LDL as early in life as feasible. Let's talk a little

00:15:12.940 bit about Obacetrapib. Your colleague, John Kastrelin was a guest on this podcast, gosh,

00:15:18.180 I want to say three, maybe four years ago. Certainly one of the more enjoyable discussions

00:15:22.720 I've had because I tend to really enjoy talking about these things. But I think it would be a

00:15:27.580 fair assumption that many people listening to us now have either not heard that podcast and if they

00:15:31.760 did hear it, do not remember the ins and outs of CTEP inhibition. So let's assume that we're

00:15:39.360 starting from scratch. Let's talk about this totally new class of drugs. Let's also just

00:15:43.580 acknowledge as a disclosure, you are currently the CEO of a company called New Amsterdam

00:15:49.300 Pharmaceuticals, which is the manufacturer of a particular drug we're going to talk about in this

00:15:54.240 class. So take us back to, well, I don't know when the drug entered clinical trials, but I know when

00:16:02.020 it ended, which would have been about 2006. I think it was eight. Okay.

00:16:07.900 Torceptor PIV. Yeah. It could have been 2006. I feel like it was the fall of 06, but anyway.

00:16:11.780 You're right. It was definitely Christmas, December, got the call somewhere in that range.

00:16:16.540 All right. So first, this was Pfizer combining this with a Torvastatin in a clinical trial. Okay.

00:16:24.240 Go back from that a decade into the discovery of yet another mechanism, because you've already

00:16:28.640 alluded to this, which is lowering LDL is the goal, but there are many paths to get there.

00:16:34.440 Statins have taken one path, which is we're going to inhibit cholesterol synthesis.

00:16:38.500 The liver will overcompensate by trying to pull more LDL into the liver, and that will lower LDL.

00:16:44.600 We're going to come and talk, I want to come back and talk about bile acid sequestrants.

00:16:47.820 We probably won't get to azetimibe. We might mention PCSK9 inhibitors.

00:16:51.280 C-type inhibitors are totally different, so just explain what they do.

00:16:54.240 It's good to start about what exactly CTP does.

00:16:58.000 It's called cholesterol ester transfer protein.

00:17:00.620 It transfers cholesterol from HDL into LDL.

00:17:04.520 So when you block it, HDL goes up, LDL goes down.

00:17:08.000 There's many other reasons.

00:17:09.360 It also improves LDL clearance from the liver, like a statin does.

00:17:13.520 We've proven that.

00:17:14.440 It does help clear LDL as well.

00:17:16.440 So animals that lack CTP, like the rat, the dog, they have very low LDLs,

00:17:23.600 very high HDLs. They don't get atherosclerosis unless you give the rat or the mouse CTP

00:17:29.680 transgenically. Then they get high LDL, low HDL, and they get atherosclerosis. Like the dog

00:17:34.800 don't get atherosclerosis. They're carnivores. They don't have any CTP. Monkeys have CTP.

00:17:40.720 Rabbits have CTP. Humans have CTP. We have, relatively speaking, high LDL, low HDL.

00:17:47.800 So in fact, one of the most prominent atherosclerotic animal models is to give the

00:17:51.900 mouse CTP. And that's what we use in a lot of our studies to show that you can induce

00:17:57.400 atherosclerosis and a different therapy can reduce atherosclerosis. So it's very much a

00:18:01.160 modulator of the LDL-HDL levels in mammals. Do you have a sense, given that everything

00:18:06.960 you just described is a mammal, which means evolutionarily, while we look different today,

00:18:11.460 we're coming from a very narrow part of the tree. Why do you think that difference exists?

00:18:15.860 And is there a reason you can speculate on? It's conservatory. In other words,

00:18:19.380 you didn't have chasing the woolly mammoth. You didn't have a lot of cholesterol available to you.

00:18:25.400 But why wouldn't dogs have the same issue then? 0.91

00:18:27.880 They have plenty of meat. All they eat is meat and all that kind of stuff. They had plenty of

00:18:31.400 cholesterol. So we think it was truly just a scarcity issue?

00:18:34.840 Yeah, scarcity. So animals that had more scarce nutrients would want to hyper-conserve cholesterol.

00:18:41.680 I mean, almost all, let's say vegetarian animals don't have CTP. So, you know, if you look at it

00:18:47.540 from that perspective, you know, we're meant to be a vegetarian maybe. So that's how do you think

00:18:52.440 about how CETP plays a role in lipid homeostasis. But it's a conservatory mechanism. If you can't

00:18:58.600 get a lot of cholesterol available from the diet, it's just a way of conserving cholesterol in the

00:19:03.160 body. So when you block CETP, you lower LDL, you raise HDL, you actually funnel more cholesterol

00:19:10.520 into the intestine. So you lose a lot of cholesterol in your body by blocking CETP.

00:19:16.440 So the story was, it was all about HDL raising.

00:19:19.320 At the time, this is 20 years ago, HDL raising was the holy grail.

00:19:23.960 We had data from the Framingham Heart Study that if your HDL was high, you're protected

00:19:28.560 from heart disease.

00:19:29.920 They used the ratio total cholesterol divided by HDL.

00:19:32.880 That was a real famous ratio that we looked at.

00:19:36.440 A lot of doctors still look at that as a risk predictor.

00:19:39.760 Raising HDL, we had a couple studies, the Helsinki Heart Study, the VA HIT trial using

00:19:45.060 a fibrate, lowered HDL, raised HDL a little bit, and it had some heart disease benefit. And so

00:19:51.440 everyone's about raising HDL. And where were we in our knowledge of niacin's impact on HDL and

00:19:57.420 events? We only knew niacin alone in the coronary drug project did lower events. We also knew niacin

00:20:06.020 as one of the drugs that was studied. It had a alone, that's the pre-statin era,

00:20:10.840 had a reduction in cardiovascular events in secondary prevention how significant relative

00:20:17.520 to what we see with statins not nearly the same as statins obviously niacin has a lot of side

00:20:21.980 effects i took niacin as before stands because that's what we had you get flushing and gi side

00:20:28.180 effects and liver issues and stuff like that but no we had niacin and niacin also raised hdl

00:20:34.360 cholesterol quite a bit like 20 30 that was the best that was also studied it didn't work out on

00:20:40.100 outcomes on top of statins. It was two studies, AIMHI and HPS Thrive trial. Both showed niacin

00:20:47.200 on top of statins did not lower LDL, so did not lower events. The Thrive HPS2 trial, there was,

00:20:54.700 it looked like, if you looked at it in another way to look at the analysis, it did lower events

00:20:59.660 by LDL lowering because it lowers LDL also. That was actually post-Toracetrapib. So,

00:21:04.800 So, tristetopid was the Pfizer CTP inhibitor. Everyone was rushing to get a CTP inhibitor

00:21:10.440 because it raised HDL 50%, 75%. We thought we'd have the cure for heart disease. Every 1% increase

00:21:18.380 in HDL in these other trials lowered risk by 3%. So, you can see the magnitude of the benefit

00:21:23.540 would be quite substantive if you can raise HDL. I personally had a low HDL in my family. So,

00:21:29.360 it was like, actually, I was the first one to patent CD inhibition back in the 90s. I was the

00:21:33.160 first one to file a patent on CETB inhibition. 1993 was the patent was issued. I had a lot of

00:21:39.120 hope that CETB inhibition would be the end of heart disease. So Tristetrapib was underway.

00:21:44.420 They rushed it, even though they knew, and I was involved in the first phase two trial,

00:21:48.380 that it raised blood pressure quite a bit. It wasn't clear how much, but we knew it raised

00:21:52.280 blood pressure three points, five points maybe. So it went into the big trials and unfortunately

00:21:59.060 it was stopped due to increased mortality, both cardiovascular and non-cardiovascular mortality.

00:22:05.040 So it was a real big disappointment. And do you think that that was attributed

00:22:07.980 solely to the increase in blood pressure? The blood pressure increase actually was

00:22:13.180 only the tip of the iceberg. The drug had a very prominent effect on increasing,

00:22:18.220 getting through the adrenal glands and would markedly increase aldosterone and steroid

00:22:23.580 production. And so it was just an off-target effect. We know that because in animal model,

00:22:29.780 like the rat, they give to our cetripib and the blood pressure would go up dramatically within an

00:22:35.880 hour without, and has no, they have no CTP. The rat has no CTP. So that said it's not, it's an 0.56

00:22:41.160 off-target effect. And since then, there's been a lot of CTP inhibitors that have, let's say four.

00:22:47.320 Yeah. So let's take them in order though, because the story is so interesting. So

00:22:50.080 This was a big blow to Pfizer. I don't remember, so I'm curious how it felt from the inside.

00:22:56.420 Was that viewed as a shot across the bow that this target might be the wrong target or was it,

00:23:02.940 no, we realized immediately this was an off-target effect. It was just a bad drug.

00:23:06.940 Merck was next, correct?

00:23:07.920 No, no, Roche, Dalsett.

00:23:09.080 Okay.

00:23:10.020 It was mostly off-target. Obviously, there were some that thought, okay, maybe the target's off,

00:23:14.360 but that came later. More validation of the target actually came later as well,

00:23:18.000 that it is a valid target. I can get into that. But the initial thinking was that it could be,

00:23:23.060 well, an off-target effect. And Fries did a lot of work on understanding that. Others that followed

00:23:27.420 can actually isolate the off-target effect to a certain composition in the molecule itself. I

00:23:32.980 mean, they knew that the chain around torcentropib that was the inducer of the aldosterone secretion.

00:23:39.720 You can do that in cell models. You can find what was the component that raised the aldosterone.

00:23:44.740 and they developed other CD inhibitors that lack that component of the molecule. And so those were

00:23:51.080 the next ones to follow. And then again, it's all about HDL. Keep in mind, it's all about HDL

00:23:54.800 raising. No one's thinking about it for LDL lowering. That's a key thing.

00:23:58.720 What was the thinking, Michael? So aside from the fact that epidemiologically we could observe

00:24:04.840 that people with high HDL, low triglycerides, and low LDL did better, was there any sense of

00:24:12.460 mechanistically, why would raising HDL do something? In other words, how was causality

00:24:18.640 established as opposed to association? That's a great question, but there's a lot of work that

00:24:24.840 went into the HDL function. You know, what does HDL do? So HDL is a lipoprotein that is also made

00:24:34.480 by the liver and the intestines, and it goes to different organs, and it basically picks up

00:24:40.960 cholesterol. It's effluxous cholesterol into the lipoprotein. It goes from an empty garbage truck

00:24:46.340 to a loaded garbage truck, an empty PETA pocket to a spherical, how you want to describe it.

00:24:51.480 It matures. It goes from this pancake up to a big ball filled with cholesterol. And there's

00:24:57.020 different enzymes along the way that kind of help maturate the particle. And once the spherical

00:25:01.960 particle is mature, it goes to the liver and gets cleared. And actually, the predominant

00:25:06.480 route, most people don't realize this, but the majority of bile that the liver makes comes from

00:25:11.180 HDL. It comes from HDL. That's the bile. It goes funnel more into the biliary system and then out

00:25:16.900 into the intestine. So HDL is the precursor, but it has a lot of functions that are really

00:25:23.380 interesting. It picks up a lot of garbage along the way. It picks up all these inflammatory

00:25:28.460 components, the cytokines, and it's like a dump truck. It's picking up stuff and it takes it back

00:25:34.020 the liver for clearance. That's what the purpose of it is. And part of that is cholesterol. And so

00:25:38.600 people thought it's going to the plaque, it's picking up cholesterol. We call it reverse

00:25:43.200 cholesterol transport. And it picks up the cholesterol and removes it from the plaque.

00:25:47.320 And that's why you get less heart disease with HDL. So it was all mechanistically based on that

00:25:52.700 finding. Even though the amount of cholesterol that it got from the plaque was actually minuscule

00:25:58.220 compared to what it does to other organs.

00:26:01.300 But it's the vehicle for cholesterol transport,

00:26:04.580 and it also has other functions.

00:26:06.280 It's very important in immune function

00:26:08.140 and trying to battle innate pathogens and things like that.

00:26:12.240 It has a lot of other important roles to play in human health.

00:26:15.820 But that was the reason why we thought HDL being high was good,

00:26:19.680 and then it functioned by removing cholesterol from the artery wall

00:26:22.660 and taking it back to liver for clearance.

00:26:24.660 Learned it's a lot more complicated than that,

00:26:26.200 but that was our thinking back in 2006 when torcetropib unfortunately kind of failed. And

00:26:32.640 just to be honest where I was, I mean, I was a big fan of torcetropib even though the blood

00:26:37.920 pressure went up because the HDL raising, the magnitude of that benefit would be greater than

00:26:42.140 a three-point increase in blood. So I was kind of thinking, okay, if any of it raises blood pressure

00:26:46.080 a little bit, it raises HDL 75%, we should get a net benefit. But it turns out it was much more

00:26:53.020 complicated than that as far as the side effect profile. Yeah. How much did HDL go up with

00:26:57.940 torcetropyl? About 75%. Wow. So despite that, that really tells you the off-target effects

00:27:03.360 were devastating. Right. Yes. Did it have an effect on LDL very much? I don't remember.

00:27:07.100 About 15%. At most, maybe 10 to 15%. It was really not even something people even paid

00:27:13.800 attention to, the LDL effect. Okay. So then the Roche compound comes along. What was the story

00:27:19.940 with that trial? So it's a weak CTP inhibitor. They even called it a CT modulator, a very safe

00:27:25.080 drug. And it raised HDL 30%, maybe 40 in some trials. So not much weaker HDL raising effect,

00:27:33.940 no LDL effect. So went into a large outcome study, very safe, no issues. It reduced the risk of

00:27:41.520 diabetes, which we'll come to later. That was one of the things about the HDL raising benefit that

00:27:46.160 we can talk about. Reduce the risk of diabetes, but had no MACE benefit. Did not cause any MACE

00:27:52.580 benefit. Major adverse cardiac events. Right, right. Heart attack or stroke reduction was

00:27:56.880 nothing. There was no benefit there. Not even a trend. It was just completely flat, but very safe.

00:28:03.180 And so that was the second failure, high-profile failure. Actually went to another trial later.

00:28:08.940 It's still neutral. No benefit on major adverse cardiac events. That's number two.

00:28:14.300 And then Merck?

00:28:15.120 Then Lilly.

00:28:16.140 Oh my gosh.

00:28:16.960 Then Lilly, which was evocetrapib.

00:28:19.680 That one's a good molecule.

00:28:21.200 Again, very safe and was studied.

00:28:24.660 Again, it was a time when Lilly did not have the resources it has now, obviously.

00:28:29.520 And so they were trying to go fast.

00:28:32.080 Everyone's trying to go fast because, again, this could be a huge opportunity to reduce

00:28:36.660 heart disease.

00:28:37.200 So they went into a study with acute coronary syndrome primarily, and they showed that

00:28:43.600 Again, LDL lowering about 15 to 20%, HDL raising again, 75% or so, a very good HDL raising effect.

00:28:53.680 They only went two years though, post starting the trial and they stopped for futility for no

00:28:59.820 benefit on major adverse cardiac events. In secondary prevention.

00:29:02.880 Secondary prevention. It was a high risk secondary prevention.

00:29:05.200 By the way, Michael, the first two studies were primary or secondary?

00:29:08.540 All secondary. These are all secondary prevention. Yeah.

00:29:11.200 What's important about that trial is azedamide, which is now a well-known, we use azedamide

00:29:16.380 all the time, ZETEA, for lowering LDL.

00:29:18.840 They also did an ACF, acute coronary syndrome trial, which was similar to what was done

00:29:22.580 in the ACCELERATE trial, which is the Lilly trial.

00:29:25.840 We know that two years is too short because in a similar trial with azedamide, which is

00:29:32.880 a well-proven LDL lowering drug, it has, everyone recognizes the benefit of azedamide, it took

00:29:39.660 more than two years for the lines to separate. After two years, their lines were, there's no

00:29:44.180 difference in major adverse cardiomyopathy, but after two years, they separated. And so the similar

00:29:48.120 would have happened, we believe, with evocetrapib in the accelerated trial.

00:29:53.000 Although with Repatha and Pryolent, we did see benefit at about two years.

00:29:57.420 This was only a 15% LDL lowering.

00:29:59.560 Yeah, yeah, yeah, yeah, yeah.

00:30:00.420 So it's also interesting, in the Odyssey outcome study, which is the Pryolent, which is the other

00:30:06.100 PCS-CAD inhibitor. If you look at populations that had LDLs below 100, which is what this

00:30:12.160 they also didn't separate until after two years. Same with Fourier. Which one was more heavily?

00:30:16.500 No, Fourier was for path. But which one was more heavily statinized? I can't remember.

00:30:20.720 They both are pretty, but Fourier was not acute coronary syndrome population.

00:30:23.620 It was a chronic stable. They send a separate, acute coronary syndrome population sometimes can

00:30:28.540 be pretty sick and they get a lot of events. And so you have to go longer to see the benefit.

00:30:33.680 Are you saying that your view is maybe the Lilly study was stopped too soon?

00:30:37.720 It was stopped for futility when it should have kept going.

00:30:40.760 That was for us a big issue for our drug, Obacetrapib.

00:30:44.520 That was the hardest one to explain.

00:30:46.400 And what were the features of the Lilly drug in terms of HDLC and LDLC kinetics?

00:30:52.860 The lower dialed at about 15% to 20%.

00:30:55.020 And raised HDLC.

00:30:55.960 ACL again, 75% or so.

00:30:58.640 It's a good, but the thing is it was safe.

00:31:01.100 it actually had a total mortality benefit that was significant, that it was not a major adverse

00:31:05.980 cardiac event benefit. So the drug would never have been, but that was not the primary endpoint.

00:31:10.200 That's interesting. It had an overall all-cause mortality benefit.

00:31:13.540 Right.

00:31:14.020 What other causes was that attributed to?

00:31:16.000 Well, it was largely cardiovascular, but it has some non-cardiovascular mortality. So when

00:31:20.520 cardiovascular mortality was almost significant, there was also some non-cardiovascular benefit

00:31:25.300 as well.

00:31:26.220 Did they pre-specify that they had to have an outcome at two years? Is that why they stopped it?

00:31:29.920 they had pre-specified that there would be an analysis at two years. There are also

00:31:33.940 corporate issues involved. It was a tough time for Lilly financially. It's very different than

00:31:38.520 where they are today. They had to make some quick decisions about prioritizing their pipeline.

00:31:43.740 As big corporations go, you have to make these decisions. So they had decided to cut the study

00:31:48.060 short. Again, HDL raising, you would think would have a benefit already. So it was the LDL

00:31:53.180 lowering wasn't really thought of as the main mechanism.

00:31:56.380 Did that study show any improvement in diabetes?

00:31:58.920 It did also, yes. All the CT inhibitors, even torcetripib, should have diabetes benefit,

00:32:03.740 all of them. So this is again why we're focusing on that with ovocetripib as well.

00:32:09.000 So the important study that was the Merck study, that actually saved the class,

00:32:13.200 even though the drug itself didn't make it to market for different reasons. It was,

00:32:18.660 to Merck's credit, I mean, they started the trial, again, all about HDL. But when they saw

00:32:23.680 what happened with the Lilly drug and Dalsetripib, they wanted then to make sure the study was

00:32:29.540 adequately powered to maybe pick up an LDL benefit. They had 30,000 patients. That's perhaps

00:32:35.860 the largest outcome study ever done. And they went for four years. So very large study. They

00:32:42.680 went for four years. The baseline LDL was 60. So we're talking about already low LDL. They got a

00:32:49.000 17% LDL lowering, 11 milligrams per deciliter, absolute LDL lowering. They got a 9% relative

00:32:56.540 risk reduction, which is what you'd expect, if not more so than what you'd expect. And so it

00:33:01.960 proved the LDL lowering benefit with CTV inhibition can translate into a cardiovascular benefit. And

00:33:09.920 that was the key study for us. Now, the reason why the drug didn't go forward was it had this

00:33:15.380 unfortunate deposition in fat tissue that wouldn't stop depositing. And so if you took it,

00:33:21.440 you would not wash the drug out for years, actually. So there's also environmental issues

00:33:26.240 that gets in the water supply. There was just a lot of issues that Merck would have to deal with

00:33:30.840 with a drug with that type of pharmacokinetic profile. And how did they figure that out only

00:33:35.800 in the large phase three study without knowing it in, say, a phase two? Like, what was it that

00:33:41.200 even brought that to their attention? As the studies went along and they started getting more

00:33:45.220 in the phase two studies, they didn't look at long-term clearance of the drug and it started.

00:33:50.840 They didn't do the PK stuff?

00:33:52.120 They did the PK. They knew it was a long half-life. They didn't realize how long. They

00:33:55.420 didn't realize how much it got deposited. I did some of them. They actually did fat biopsies to

00:33:59.980 see how much drug was in the fat tissue. And this was an oral drug?

00:34:04.000 Oral drug. Lipophilic, yes.

00:34:06.200 Wow.

00:34:06.940 Yeah. So it was unfortunate in that it was just not a viable commercial drug for that reason,

00:34:12.460 even though it had good safety profile, it lowered MACE, but also because the magnitude of the MACE

00:34:17.940 benefit was modest, people don't realize it's an issue for how you market the drug, but from a

00:34:23.100 proof of concept of proving the drug works to lower LDL and reduce events, it was a great study

00:34:28.380 to confirm that the LDL lowering with the CTP inhibitor can, in fact, result in a cardiovascular

00:34:34.460 benefit. You've referred to this a couple of times, so I just want to make sure people understand

00:34:39.200 why this is the case. And it actually goes back to sort of our argument earlier about causality.

00:34:45.340 When you line up all of the primary prevention trials, full stop, all of the secondary prevention

00:34:52.620 trials, full stop, all of the Mendelian randomizations and all of the epidemiology,

00:34:57.740 and you plot on the X-axis LDL-C and on the Y-axis event rate, they're all a bunch of lines

00:35:05.860 going down. So they're four distinct lines, different slopes based on different. And

00:35:11.220 therefore, given how tight those regressions are, it becomes very easy to predict what you're

00:35:18.800 getting at. I just want to make sure the listener understands why it is that you're saying if you

00:35:23.400 have a 17% reduction in LDL-C from 60 to 49, you would expect an event reduction on a relative

00:35:31.520 basis of 9%, et cetera. Yeah. The formula is easy. It's well-established. For a one millimole,

00:35:38.420 which is 38.8 milligrams per deciliter of LDL, you get a 22% melt of risk reduction.

00:35:46.520 And that's largely linear. It's linear, right? But absolute is the key, not percent. You got to

00:35:51.880 look at the absolute LDL lowering, and then you can calculate the benefit. And we're doing that

00:35:56.000 for our, obviously, our outcome study. We can have a predicted benefit based on our absolute LDL

00:36:01.120 lowering that we hope to achieve in our prevail trial. You power studies accordingly. You can see

00:36:05.780 what the benefits, as long as you go long enough too, that was the other key thing.

00:36:09.180 In the first year of treatment, you get half the benefit at best. It takes time for the lines to

00:36:15.120 separate. And that's what gets back to the other point you made earlier, which is all of these

00:36:20.220 regressions are independent of how you do it. The Mendelian randomization says we're going to lower

00:36:25.800 it genetically. It's the genetic lottery that determined your LDL versus mine. The primary and

00:36:31.720 secondary prevention trials are based on a potpourri of drugs. Same with the epidemiologic

00:36:36.760 stuff, which combines both genetic and drugs. So again, it speaks to this idea that reinforces the

00:36:44.420 causality of LDL because independent of how it's lowered, you're getting the same effects.

00:36:48.900 You'll often hear people say things like, well, it's hard to deny that these drugs improve

00:36:53.400 outcomes, but it must be something different than LDL lowering if you want to deny that LDL

00:36:58.880 plays a role. They'll talk about pleiotropic effects of these drugs or things like that,

00:37:02.280 but obviously those arguments become a little bit silly when you consider the totality of

00:37:06.260 the evidence. Right, right. I've never been a pleiotropic, in fact, that was the one always

00:37:10.060 debated that issue. I've always been, it's LDL, it's LDL. Now, I think the argument is stronger

00:37:15.440 that there's nothing special about statins other than they're very effective and well-tolerated,

00:37:20.880 but there's nothing special about the LDL-lowering about statins that result in a greater benefit

00:37:24.680 than any other LDL-lowering drug that can achieve the same levels in a patient.

00:37:29.600 So with all of that as background, now let's talk about Obacetrapib, the work that you and

00:37:36.640 John and the team, of course, but let's just talk about how you decided it was worth going

00:37:42.000 after this for a fifth time. Right. I had sold my previous company an interleukin-6 antibody

00:37:48.020 for heart disease, inflammatory. I don't know if you're familiar with IL-6, but it's the

00:37:52.520 key inflammatory cytokine. And that study is actually coming out this year. It's called

00:37:56.480 the Zeus trial. It's a big, big study coming up. I had sold Corvidia and John said, let's get

00:38:02.440 Obacetrapib back from Amgen. Amgen had put it on the shelf because Repatha had struggled on its

00:38:09.600 launch. They deprioritized it. So they kind of had it sitting there ready to go into phase two,

00:38:15.440 maybe even right into phase three potentially so we teamed up together and started new amsterdam

00:38:21.080 pharma how many years ago this is five almost this is our sixth year now that's so interesting

00:38:26.920 because rapatha launches in 2015 right they kept ob on the shelf for five years i think they bought

00:38:35.020 it in 2017 they bought it in 2017 interesting they actually did a lot of great work on the

00:38:40.860 manufacturing, the process. They worked on it, made it a much more cost-effective

00:38:44.860 manufacturer, but they didn't do any clinical trials. They just kept it on the shelf.

00:38:49.620 It already had an IND in the U.S. or this was all-

00:38:52.720 It already had phase two data.

00:38:53.820 In the U.S.?

00:38:54.500 In the U.S., yeah.

00:38:55.260 Okay.

00:38:55.660 It was ready to go. So we initially thought about statin tolerance and using it for those patients

00:39:02.240 as a quick route to the market, but we realized the FDA view on CD inhibition because of the

00:39:08.340 history, we really didn't have a path there. The FDA does not like statin intolerance as an

00:39:13.960 indication. They're very wary of it even really being a true thing, but believe it or not. I mean,

00:39:18.540 you and I see it, we all see it, but in the academic world and in the FDA, they don't really

00:39:24.580 believe it is a true phenomenon. Because if we look at the placebo-controlled trials, it's very

00:39:29.120 low rates. It's about 4.9%. Yeah, but even true statin tolerance is very hard to find. I mean,

00:39:33.460 in the trials. But anyway, it exists. It exists. That was our first path. However, once we met with

00:39:39.560 the FDA and realized we need to go into a full boat LDL lowering path, which is a much more

00:39:46.920 larger program. And also, we know we had an outcome study at the time of the launch of the

00:39:52.760 drug because of the history. But what were the phase two data that you had at the time of

00:39:56.580 acquisition? We had a TULIP study, which was the set of John had led in a predecessor company.

00:40:01.200 Yeah. Can you remind me and the listeners what that study looked at?

00:40:03.700 So it showed that 45% LDL lowering.

00:40:06.480 Okay. So what was different about OB?

00:40:08.860 It's a much more potent. So it's only, the other citium inhibitors are all in the

00:40:12.720 100 milligram plus range and they lowered LDL, as I mentioned.

00:40:18.100 And you're at 10 milligrams.

00:40:19.000 They were at 10 milligrams, even five milligrams we had initially too,

00:40:23.200 lowered LDL on that 40 to 45 to 50% range in the trials.

00:40:28.040 And the HDL raising?

00:40:29.260 150%. So it's a lot more effective on both lowering LDL and raising HDL. It was just a

00:40:36.180 much more potent CETP inhibitor. And the duration of the TULIP study was how long?

00:40:40.900 It was just a short-term study. I think it was eight to 12 weeks, something like that. It was

00:40:44.440 relatively short study. It was very effective, very well tolerated. It was a really nice LDL

00:40:50.060 benefit. So we wanted to look at for LDL lowering. Again, in the meantime, all the Mendelian

00:40:53.900 mandeliumization studies, besides the Merck data, at the same time, or roughly the same time,

00:40:59.740 all the mandelium mandeliumization studies confirmed that LDL lowering translates into

00:41:03.420 cardiovascular benefit. Weren't these the same studies, sorry to interrupt, Michael,

00:41:07.180 that suggested that the HDL raising had nothing to do with the benefit of CTEP inhibition?

00:41:11.980 Right, exactly. And other HDL raising genes saw no benefit. At least the benefit could not be

00:41:18.700 explained by the HDL effects. If you looked at HDL-raising genes that are associated with

00:41:24.880 lower risk, they also had other things like LDL-lowering or triglyceride-lowering. There was

00:41:29.720 no pure HDL-raising gene that was associated with protection.

00:41:34.700 This was a paper in Nature circa, what, 2012-ish?

00:41:39.520 Yeah, right, exactly. So that was the other important...

00:41:43.060 I feel like that doesn't get enough attention. To your point, you said, look, there are still

00:41:46.040 a lot of doctors out there, a lot of patients I hear saying, well, my, I hate the term. I hate

00:41:50.480 the term, but people know what I mean when I say this. My good cholesterol is high, so I'm okay.

00:41:56.580 So notwithstanding the nomenclature flub, even if they were to say my HDL cholesterol is high,

00:42:01.660 I'm okay. The answer is no. We've known for over 10 years that that is not true.

00:42:07.920 You notice I really worked hard to avoid saying the term good cholesterol. I'm avoiding using

00:42:12.540 that term because our mutual friend hates it, as we know, but it's the wrong way to phrase it.

00:42:18.000 Yeah.

00:42:18.200 It's the wrong way to phrase it.

00:42:19.320 But again, I think it's one thing to hear patients say that and they can be forgiven

00:42:23.520 because they don't live in this world. But I really struggle when I hear physicians say,

00:42:28.220 yeah, but his HDL cholesterol is high. So we're okay. I have a patient that comes to me and their

00:42:32.820 LDLC is through the roof and their HDLC is high. And they say, well, yeah, but my doctor told me

00:42:37.640 that because my HDL cholesterol is so high and my triglycerides are low, I'm fine, right?

00:42:43.480 So HDL is really interesting. We believe it could still have some cardiovascular benefits

00:42:47.500 if you raised it the right way. And also because HDLs, there are definitely genetic factors that

00:42:53.320 raise HDL that increase risk, like SRB1, for example, that's a known high HDL gene that does

00:42:59.980 increase risk. We know alcohol raises HDL and causes all kinds of higher mortality issues.

00:43:06.060 When you look at HDL, it's a complicated thing to have a high HDL. We could talk about a whole

00:43:11.400 podcast about how we evaluate a high HDL patient and how we think about how to advise them about

00:43:17.400 what it means because sometimes it could be quite good. I have a friend with very high HDL cholesterol,

00:43:22.300 about 100 milligrams per deciliter, and modest LDL cholesterol, so maybe also 100, 110 milligrams

00:43:29.680 per deciliter. And he had always assumed he was totally risk-free. Maybe his LDL was even lower

00:43:36.380 than that. It might've been 90 to a hundred with HDLC of a hundred, 110. And I was able to convince

00:43:43.600 him. I said, you know, look, you're right. Statistically speaking, your HDL cholesterol

00:43:47.360 is probably reflective of something good or protective, but I've read enough cases in the

00:43:52.500 literature to see that we could be looking at highly dysfunctional HDLC. So would you mind

00:43:58.060 getting a CT angiogram, which he did, and he had significant burden of disease.

00:44:02.680 Yeah, so he probably has the SRB1. That's Ashkenazi Jewish predominantly. It's where we 0.99

00:44:07.140 find it most commonly. It's still relatively rare. We see, I have a series on that in my

00:44:12.180 clinic that have the high HDL due to SRB1. There's anothelolipase. There's like three

00:44:17.920 HDL-raising genes that either don't protect or increase risk. And then you have alcohol. So

00:44:22.760 you throw that in the mix, you get a very complicated story about HDL being high,

00:44:27.220 being protective. The reasons to believe, we still have hope that Obocentropib's HDL-raising

00:44:32.040 benefit can result in a cardiovascular benefit, but we're also focusing primarily on what the

00:44:37.820 other HDL benefits could be, like we're still talking about later, like diabetes prevention

00:44:42.380 and Alzheimer's. That's what we hope is linked to the HDL-raising benefit. We're investigating

00:44:47.160 that now in all our studies. So when John was on the podcast, again, I think it was four years ago-ish,

00:44:53.380 but a lot has happened since then. So let's talk about where we are today. What is the state of

00:44:58.960 understanding of Obocetropib? And let's start with cardiovascular because what John predicted at the

00:45:04.480 time was this is going to have benefits in cardiovascular disease, but it's going to have

00:45:09.660 benefits in metabolic disease. It's going to have benefits in Alzheimer's disease. So let's take

00:45:15.980 them in that order. So let's start with what is known today based on the ROSE trial is the most

00:45:22.480 recent? Well, no, Broadway tandem. Yeah. So we've now done two more phase two trials and three very

00:45:29.060 large phase three trials, and they're all been completed and were actually filed in Europe

00:45:33.280 for approval. In the US, we're waiting to time it on the outcome study called Prevail. So we started

00:45:38.260 Prevail. So we had, I hope, the courage and the foresight to start our outcome study at the same

00:45:42.760 time as our phase three trials. As a small company, we were able to knock on wood, even though we'd

00:45:47.440 have the money to finish the trials, start the trials, get them completed, and then we raise

00:45:52.580 money along the way. Now we have plenty of cash to finish the trial, the Prevail trial. So we feel

00:45:58.680 it really worked out well for us because the success of the phase three trials allowed us to

00:46:02.860 fund the finishing the other trial. So it timed it extremely well for our efforts. We had rows

00:46:10.180 and rows two, both showed, again, the 45 to 50% monotherapy benefit. And then in combination with

00:46:17.120 azetamide, you know, in that 55% range, higher, better LDL lowering. We then did what was called

00:46:23.600 Brooklyn, which is a familial hypercholesterolemia trial. I know you've done multiple podcasts on,

00:46:28.040 it's an autosomal dominant genic disorder, very high LDL, showed a 40% LDL lowering over a year.

00:46:35.060 So it's a one-year study. Tell me about those patients. They were already on?

00:46:39.000 Almost all were on high intensity stands. Many were on azetamide or PCS-GYN inhibitors already.

00:46:44.220 these are very well treated, but their LDLs were still too high. How high? Well, the average LDL

00:46:48.800 is roughly around 100, something like that. So in other words, the inclusion criteria is not

00:46:53.880 limited to exactly which drug you're on. It's just that you're on maximum therapy and your LDL is

00:46:58.720 still above a certain threshold. Right, right. So on the familial hypercholesterolemia patients,

00:47:03.420 when you show up at a LDL-C of 100 on whatever combination you're on, Obacetropib lowered it

00:47:10.300 by an additional 40 percent. So lowered it from 100 to 60 milligrams per deciliter on average.

00:47:15.660 Approximately. That was, I think, those are approximate numbers. About 40 percent was the

00:47:18.800 reduction at the end of 12 months versus placebo. And then the Broadway trial was, these are all

00:47:25.200 atherosclerotic cardiovascular disease patients. Again, all on maximal statins, majority on high

00:47:31.000 intensity statins, like 70 percent were on 20 or 40 of rosuvastatin or 40 or 80 of atorvacet.

00:47:36.760 they're all very well treated and we got roughly around a 35% LDL lowering benefit and then

00:47:44.480 adjusting for placebo a little less than that but it was around that percent change at 12 weeks and

00:47:50.540 then it continued for 52 weeks and so that was our two pivotal phase three trials and then in

00:47:57.140 combination with ezetimibe we did another trial called tandem where the LDL lowering was very

00:48:01.460 much 50% so we have a combo pill we're developing the two drugs together. What was important about

00:48:06.620 broad way, though, that really helped us a lot was we showed in that study a 21% reduction in

00:48:14.120 major adverse cardiac events in one year. It was not specifically significant because this is 2,500

00:48:19.740 patients, but it was the right direction and it was pretty powerful. If you look at kind of a

00:48:26.060 landmark analysis, we divide, take the first six months away, it starts to become significant.

00:48:31.300 Then remember, it takes time for these curves to separate. Everyone got excited. Again, we're

00:48:35.560 trying to obviously leverage that with Prevail because it's almost identical population in

00:48:40.340 Prevail, which is 9,500 patients. And it's an ASCVD, atherosclerotic cardiovascular disease

00:48:45.940 population. LDL, again, in the higher end, that's the key. You have to try to get higher LDL levels

00:48:52.000 but not being too high where it becomes, you get too much drop into therapies or you get on ethic

00:48:57.560 issues, ethical issues. So these are all on maxly tolerated stands. That's called Prevail. And that

00:49:02.840 study now is approaching, we're in the year three of the study now. And this study will read out

00:49:09.240 when? We have two factors to stop the trial. One is that we have a minimum follow-up of two and a

00:49:13.700 half years. Again, the learnings from the other trials. And the other is also event-driven. How

00:49:19.460 many events to power the study appropriately for what we call four-point MACE, which is the heart

00:49:24.440 attack, stroke, cardiovascular, coronary death, and then revascularization, coronary. So those

00:49:31.020 of the four-point mace. So we have that events need to be adequate to stop the trial. So that

00:49:37.100 may take us into a little bit longer than the two and a half year. The two and a half year mark ends

00:49:41.220 this year, at the end of this year, for the minimum follow-up. That will be a trial evaluated in both

00:49:47.900 Europe and the United States? All over the world, yeah. China, Japan, everywhere. It's 400 plus

00:49:52.520 sites and it's a global study. And the outcome of that study, should it be positive, would be

00:49:58.780 approval in Europe and North America for Obacetrapib in secondary prevention?

00:50:04.900 We actually don't need it for approval in Europe.

00:50:07.380 Why is that?

00:50:08.080 Well, it's a different system. They believe in LDL lowering is LDL lowering, so you don't need-

00:50:11.580 They don't need the outcome trial.

00:50:12.460 They don't need the outcome study for approval. The U.S. is a different story. Then they have a

00:50:16.360 year to set up their payer mix. So they have the year for the outcome study to kind of come into

00:50:21.400 the payer reimbursement. For U.S., it's different because you want to line up your reimbursement

00:50:26.540 at launch. And so we want to have the outcome study when we launch. So actually Europe will

00:50:29.940 be first. I went through the ROSE II trial the other day. And one of the things that I found

00:50:35.320 interesting was, and we'll link to the study so folks can see it, but the LDLC lowering,

00:50:41.140 there was also an NMR for LDLP lowering, and then there was ApoB. And there was quite a

00:50:47.180 discordance between those. The ApoB lowering, I don't remember, was maybe 16%, but it was much

00:50:53.340 more modest than the others. What do you think accounts for that? It's something we're looking

00:50:57.800 at. I know a lot of our good friends, Tom, Alan, they're all ApoB. I'm more of an LDLP person for 0.99

00:51:05.000 a couple of reasons. One, it's easier to get in our lab. Really? Yeah, easier actually. ApoB is

00:51:10.300 still pretty easy. I mean, I feel like ApoB is the easiest thing to order. Yeah, but LDLP,

00:51:14.500 but also LDLP to me is more commonly discordant than ApoB. You can see it more easily because

00:51:22.000 APO-B, LDL's 80, APO-B comes back 85. Is that really that kind of thing? Usually would typically

00:51:28.640 have with LDL-P of LDL 70, LDL particles 1500. It's very easy to see the discordance a lot easier.

00:51:37.500 Which assay did you use for NMR? This was the-

00:51:40.080 This is the liposcience.

00:51:41.660 Liposcience one, which is owned by LabCorp now?

00:51:43.580 It's owned by LabCorp, yes.

00:51:45.220 This is Jim Cromwell's old assay.

00:51:47.080 Actually, no, we use the liposcience, yes. That was the assay that we used.

00:51:50.260 Okay, so that is the most reliable one to my knowledge.

00:51:52.780 Right, exactly right.

00:51:53.580 And so we're doing a lot of work confirming that Tom actually published a paper,

00:51:58.320 when LDL-P and ApoB are discordant, what is driving that?

00:52:01.820 Which is not very common, but you do see it.

00:52:04.420 And it's insulin resistance and a lot of small particles.

00:52:08.340 You get discordance between ApoB and LDL-P.

00:52:11.100 And then there's a study done at the CDC.

00:52:13.080 Sorry, you said ApoB and LDL-P.

00:52:14.820 Did you mean ApoB and LDL-C?

00:52:16.480 No, LDL-P, when they're discordant.

00:52:18.200 In other words, when ApoB is not high and LDL-P is high, what explains that?

00:52:23.160 What is the reason why you have a normal ApoB and a very high LDL-P?

00:52:27.600 But I don't understand why small particles would explain that discordance because each

00:52:32.040 particle still has one ApoB on it.

00:52:34.220 It turns out the mass of ApoB is different by the size, slightly different from each

00:52:38.780 particle.

00:52:39.940 And that we believe is the explanation.

00:52:41.720 Why is that?

00:52:42.680 It's because when you have the smaller surface, the mass is not just the protein, there's

00:52:46.900 also other sugar molecules that attach to that that make the massive. When you measure them,

00:52:51.500 so the ApoB is a mass assay. The LDL-P is the NMR, the nuclear magnet is the resin. So CTP

00:52:57.520 inhibition is the one example where you get different results. LDL-P goes down very much

00:53:04.020 dramatically. Small particles go down by 90%. You wipe out small particles. ApoB doesn't go down as

00:53:11.820 much as you'd expect. And so we're trying to understand that better. Actually, we're doing

00:53:16.360 a lot of work on the UK biobank trying to see when that happens, where does risk follow?

00:53:22.300 That's actually going to be very interesting when you get your outcome data, assuming that

00:53:27.260 the discordance continues between those two biomarkers, because you'll be able to go back

00:53:32.480 and say, well, we would predict based on this LDL-C or LDL-P lowering what the event reduction

00:53:38.440 should have been, but that might be different than what's predicted by the ApoB reduction.

00:53:42.220 People have brought that up as an issue, but John and I both feel very confident that it's going to be as good. One thing I think that we look at is what's called non-HDL cholesterol, which is very close to ApoB. It's LDL plus VLDL cholesterol. We go back to the Merck trial, the REVEAL trial, that showed the benefit there was driven, it was really highly correlated with non-HDL cholesterol, more so than ApoB.

00:54:06.420 Although there was that paper in either JAMA or New England Journal of Medicine about six

00:54:09.840 years ago that compared APO-B to non-HDLC and APO-B performed better.

00:54:15.220 Others have shown the opposite though. It goes both ways. Listen, I love APO-B. I always have

00:54:20.320 like LDL. It's important to understand what discordance means. And discordance,

00:54:24.740 you can get discordance with using non-HDL, APO-B or LDL-P. I think by and large,

00:54:30.200 they tell you the same thing by and large. Yeah. This is a third order term.

00:54:33.800 Yeah, yeah. The cheap way is non-HDL. Then ApoB and LDLP are a little bit more precise,

00:54:39.940 but another blab test involved. We know that in our ApoB lowering with Prevail, we still do okay.

00:54:45.620 We do okay with the outcome benefit that we want to achieve. We even do better if we use non-HDL.

00:54:51.020 But of course, what Obacetrapib has that we're also excited about is the LPLA lowering benefit.

00:54:56.780 How much was that?

00:54:57.760 It's about 50%.

00:54:59.080 Which is more than a PCSK9 inhibitor.

00:55:01.280 Right, right. In the 50 to 150 range, you have to bracket it by the range of LPL-A.

00:55:06.780 What do you think explains that mechanism of action?

00:55:08.560 We don't know for sure because we've been trying again. We don't know why p-syscanine lower LPL-A.

00:55:12.880 We don't know the reasons, all the reasons why those drugs lower LPL-A about 15 to 20%. But we

00:55:18.920 were looking into this. We have done a clearance study. It does reduce production of LPL-A. We've

00:55:24.720 done an actual assay looking at a study labeling LPL-A and seeing that it does reduce synthesis

00:55:30.280 in that study. It was a small study, so we want to try it, but we don't necessarily have all the

00:55:36.300 reasons. But LpA lowering with the other CTV inhibitors as well, but not nearly as much.

00:55:40.000 It was the same thing. We have a much more potent CTV inhibitor.

00:55:44.020 And where are we right now with the antisense oligonucleotide that lowers LpA? That is still

00:55:50.440 in a phase three trial?

00:55:51.760 Yeah, it's coming out this year. It's called Horizon. It's a really important study.

00:55:56.140 Although it was extended.

00:55:57.600 It could have been stopped for futility, right?

00:55:59.920 It could have been stuff for futility, and it has a lower than expected event rate, which probably brings us to another important discussion about how do you treat a high LPLA today?

00:56:09.820 What is the mechanism?

00:56:10.820 So you want to bring the LDL down.

00:56:12.660 So they brought the LDL down to 60 in Horizon, and so they had a lower event rate than expected.

00:56:17.960 And so I think that proves what we're doing today.

00:56:21.280 I'm sure you hear it too, but when you have a patient who has high LPLA, they go, I don't want to be on a stat because it raises LPLA.

00:56:27.420 And so what do we do?

00:56:28.340 Well, the truth of the malady is you bring the LDL down, you really mitigate risk substantially.

00:56:33.680 So you think that the issue with that study was their LDLC was so low that you didn't

00:56:41.140 get enough events.

00:56:42.080 So abolishing LP little a with the drug, the ASO didn't matter.

00:56:48.020 No, no, I think it will matter.

00:56:49.300 Well, it didn't matter in the timeframe that they looked.

00:56:51.760 Didn't need to collect enough events.

00:56:53.020 I mean, it could well matter.

00:56:54.140 I mean, it's going to come out, we believe sometime this year has been reported.

00:56:57.500 So that'll hopefully convert the LPLA to our very opening discussion about causality.

00:57:03.160 That becomes then, we know it's an important risk marker.

00:57:06.500 We believe it's causal, a lot of reasons to believe it's causal.

00:57:09.400 And now we have a study that proves that lowering it has benefit.

00:57:12.560 It'll be a great achievement.

00:57:13.540 Now, this may not be the best LPLA-lowering therapy by far.

00:57:16.700 There's others that look a lot better that are moving into big trials as well.

00:57:22.040 And so the LPLA era is getting started. 0.58

00:57:25.300 I mean, we're getting to another, hopefully, very important causal factor that can be treated

00:57:30.120 aggressively with therapy.

00:57:31.540 And we hope Obacetrapib can be part of that treatment paradigm.

00:57:36.260 From a purely cardiovascular standpoint, how do you see Obacetrapib being used should

00:57:42.280 the outcome trial end in the fashion that you would expect?

00:57:46.660 Where do you think it fits into the current toolkit we have with respect to pharmacologically

00:57:51.340 lowering LDL?

00:57:52.520 So I want to say, again, your question is the right way to frame it. We have to have a positive

00:57:56.580 prevail trial and we have to show the benefits of the drug on top of statins. However, when you

00:58:02.140 think about what statins do well and what they don't do so well, so statins lower LDL and they're

00:58:09.380 well-tolerated and they reduce cardiovascular events and they're generic and they're very easy

00:58:13.820 to prescribe. That's all the good things. What does statins not do so well? It raises the risk

00:58:18.860 of diabetes slightly. It raises the LPLA, and it does not lower small particles very well. It

00:58:25.540 lowers mostly the non-small particles. But does ovocetrapib do well? It lowers the risk of

00:58:30.700 diabetes based on our Broadway trial data, as well as other CTV inhibitors. It lowers LPLA in

00:58:36.780 that 50% range, in that moderate band, and it lowers the risk of diabetes, lowers LPLA, and

00:58:41.820 lowers the small particles by 90%. And so it would be a great companion drug to statins.

00:58:49.040 We believe it becomes the go-to drug after statins for those reasons. It makes the statin

00:58:54.260 liability issues mostly go away. And it really is a good companion drug to a statin for that reason.

00:59:00.800 Now we have the combo pill with azetamide, which puts us in the same range as the

00:59:04.940 PCSKIN inhibitors or even the upcoming oral PCSKIN inhibitors.

00:59:08.620 There is another issue, John, with statins that I don't think gets enough attention because I

00:59:14.200 realize it hasn't demonstrated a clinical harm, but there's something that always, I guess we

00:59:21.380 just out of an abundance of caution, pay attention to it, which is it can have a pretty significant

00:59:25.860 impact on transaminases, especially when combined with ezetimibe. And we see that, honestly, I think

00:59:31.100 we see that far more than it's reported in the literature. So I understand what the guidance

00:59:36.200 says. The guidance says, if the AST and the ALT rise by less than 3X, you can ignore it. It's not

00:59:42.180 clinically significant. But that means you're going to be tolerating patients on Crestor and

00:59:48.000 Zetia walking around with AST and ALT of 80. And the reality of it is I just have a hard time

00:59:53.940 believing that that makes sense. What's your view on that? Do you think I'm being too much of a

00:59:57.380 Puritan? Not really. I kind of feel the same way. I'm a first do no harm kind of guy. I mean,

01:00:02.380 You bring up a very big point for me in my clinical practices. I don't like high-dose statins.

01:00:08.400 I don't like 80 milligrams of atorvastatin in particular. I don't like 40 milligrams of

01:00:12.060 resubastatin. Have you seen Tom Dayspring put out a beautiful figure? It's one of your classic Tom

01:00:17.160 figures that he pulled out of all the literature, and it shows the dose response of every statin.

01:00:22.760 And we now show it to our patients, especially those that come in on high-dose statins. And

01:00:26.960 I'm sure it demonstrates exactly what you're about to say, which is you're getting virtually

01:00:31.500 all the bang for your buck at that lowest dose. And that curve is so concave down that it's almost

01:00:38.800 unjustifiable to be on a statin above 50% of its max dose. Yeah. I mean, it's like the number

01:00:44.360 needed to harm versus the number needed to treat benefit is just not desirable. I try to cut the

01:00:49.440 statin. Patients, by the way, they don't want high dose statins either. The whole high dose statin

01:00:54.400 guideline paradigm is off. It doesn't work. And in defense of it, look, I want to be charitable

01:00:59.540 to everybody. It made sense 25 years ago because 25 years ago, you didn't have Repatha, you didn't

01:01:05.240 have Proluent, you didn't have Ezetimibe, you didn't have Bempadoic acid, you had nothing.

01:01:10.180 This was your drug. We didn't have evidence of benefit either on top of these drugs.

01:01:14.020 Now we have many studies showing the benefits. The tide is turning. People are now advocating

01:01:19.400 guideline. Europe is already on board with more combo therapy early just to get combo therapy.

01:01:25.160 So I think Obacetrapid then becomes, in my view, based on our data, should become the

01:01:30.100 next go-to drug.

01:01:31.480 And if you need even more potent, you can add the ezetimibe combo pill.

01:01:35.840 So we believe that with OB alone and OB with ezetimibe, 90% of patients are done with their

01:01:41.580 lipid treatment.

01:01:43.180 And so that's how we think for primary care doctors, that's what they're looking for.

01:01:46.700 They're looking for something they can just add to a statin and then they can call it

01:01:50.880 a day.

01:01:51.600 Explain something to me.

01:01:52.920 So if you go and buy branded Zetia today, it's not that expensive. If you go and buy

01:01:59.580 branded Nexlitol, I mean, it's as expensive as an injectable monoclonal antibody.

01:02:06.300 Yes.

01:02:06.700 So do you, and maybe you don't want to talk about this, but I have to ask, I mean,

01:02:12.180 when you think about how Obisetrapib will be priced in the United States,

01:02:16.040 is it going to be priced more like Zetia or is it going to be priced more like Nexlitol?

01:02:21.240 Unfortunately, it's all about access, not about price.

01:02:24.020 That's the big difference.

01:02:25.740 So we haven't decided on price.

01:02:27.480 You don't do that until you launch.

01:02:28.880 But it's really, fortunately, how the system works.

01:02:31.440 If you price too low, the PBMs don't have any rebating profits to make.

01:02:37.600 So they don't put it on formulary.

01:02:39.260 And so you've got to be able to play the rebate game and price it accordingly.

01:02:44.580 We're starting to see that breakdown also with Lily Direct and all these.

01:02:48.360 we love to see that system get better as time goes on. And so people can just pay a reasonable 0.96

01:02:54.300 price out of pocket if they want to, and they get the drug. Look, you don't need advice from

01:02:59.420 a bonehead with a podcast, but I have to tell you, I really think it is a terrible mistake when these 0.97

01:03:05.060 drugs are priced outrageously. Well, we know C-berpatho was a huge example, right? Yeah. I 0.97

01:03:09.640 mean, what a blunder to price that drug at $16,000 per year. I mean, they could have run rough shot

01:03:17.720 over the market if they had just priced it within, yeah. So again, assuming that the results of

01:03:23.620 Prevail are as promising as they should be, you never know, but what they should be, it would do

01:03:30.440 such a disservice to patients for this drug to be priced, even at the level of Nexlatol. Again,

01:03:35.680 Nexlatol is an inaccessible drug. To me, it's a donkey ass drug. $6,000 a year for that drug, 1.00

01:03:41.620 it's just not worth it. That's why we have a hard time finding a place for it for that reason in

01:03:45.800 the clinic. It's really limited the stat intolerant patients that don't want to take

01:03:50.060 injections. And it stinks. It barely works. Like it's just a weak drug. Now, John predicted we are

01:03:56.900 going to see big effects on metabolic disease. Okay, that was an easy prediction to make based

01:04:03.020 on everything you'd seen before it. But the other prediction he made was even more intriguing to me,

01:04:08.500 which is it might be, and I'm paraphrasing, but something to the effect, it might be that the

01:04:13.100 most impressive thing about this drug is its effect on Alzheimer's disease. So tell me what

01:04:19.640 it is that made you and John feel that way four or five years ago, and have you been validated

01:04:26.180 in your predictions? Sure. It's very exciting for me to talk about it. When I came to New

01:04:30.960 Amsterdam with John, John said, Michael, I want to hear your story about Alzheimer's. And I said,

01:04:37.620 let me go through it with you because we had what we had at the time was a Mendelian minimization

01:04:43.760 we had the famous Bronx aging study where people who 400 Ashkenazi Jews over close to 100 or older

01:04:52.940 they found out what was the most common gene and that was CTP loss of function so they had a CTP

01:04:59.280 and they had less then later came the data showing that if you have APOE4 and you have a CTP loss of

01:05:05.360 function, your risk of Alzheimer's is significantly mitigated. So we had genomics validation.

01:05:10.840 What was the phenotype lipid-wise of those patients?

01:05:13.560 Well, they have high HDL.

01:05:15.560 These are APOE4 patients who phenotypically have very favorable lipid profiles, but until you do

01:05:22.820 the genetics, you wouldn't know it's due to...

01:05:25.320 And they have very low, small LDL, and they have very high, large LDL particles. I mean,

01:05:31.420 it's what CTP inhibition does. And so, Michael, what was the degree of risk reduction in those

01:05:37.360 APOE4 patients? Well, you look at the curves, it basically took the E4 risk away. Again,

01:05:42.640 this is one study. And so they basically became the risk of a non-E4 by having the CTP loss of

01:05:49.220 function. Wow. So that's a profound benefit. So that's one. And then it's been validated by other

01:05:54.720 genomic studies since then. We've done a lot of work ourselves on that. But also in Canada,

01:05:59.120 and mcgill a phd researcher had done some really great work on ctp so you remember the mouse does

01:06:07.440 not have ctp when you give the mouse amyloid precursor protein gene they get amyloid and

01:06:13.740 dementia and then when you give them the ctp gene with it it gets a lot worse it's a lot worse and

01:06:20.260 then if you give the mouse a ctp inhibitor you improve the dementia in that animal model so

01:06:27.620 they do nesting. Wait, wait, I'm a bit confused. You're saying you take a mouse that doesn't have

01:06:33.400 CTAP and you give it a CTAP. And you give it a CTAP inhibitor?

01:06:37.920 Right. Why would that change anything if they don't have it in the first place?

01:06:41.100 No, but you gave them the CTAP. Ah, got it. Okay.

01:06:42.680 You gave the mouse the CTAP gene. Yep. Then you block it.

01:06:45.760 Giving CTAP, first of all, made the dimension worse. Yes, understood.

01:06:47.920 And then you give it a CT inhibitor, it blocked it. So we had that data. So when I started with

01:06:53.360 john i said john can we study alzheimer's and we went to our investors and they said yes you can

01:06:57.980 have a slight amount of money to look it out unfortunately in the biotech world alzheimer's

01:07:04.300 does not it's a huge graveyard yeah and it's really hard to get funding for alzheimer's studies

01:07:10.560 but we had another obviously the ldl and the cardiovascular so we raised 200 million dollars

01:07:16.000 in our series a we got we got 1 million allocated for alzheimer's research at the beginning when

01:07:21.600 did you go public? 2022. So we started the company in 21, early 21. We started the funding and then

01:07:28.620 we went public in the end of 22. So that was the basis of our, the genomic data. We had the animal

01:07:34.700 data. We also had a lot of data on HDL being somewhat protective against Alzheimer's. We had

01:07:40.220 a lot of data on that, mixed data, not entirely convincing, but a number of studies showed HDL,

01:07:46.160 high HDL was protective against Alzheimer's, especially APOA1, which is the protein carrier

01:07:50.540 for HTL. That's where we started. And so then with our funding, we were able to do a small

01:07:55.980 study in Amsterdam where we looked at about 13 ApoE4 with impaired cognitive function,

01:08:03.100 mild cognitive impairment. We gave them ovocetrapib and we saw some improvements in lipid

01:08:08.120 metabolism. ApoE4, by the way, is a lipid gene. Let's back up and let's talk about what about

01:08:12.840 ApoE4 is detrimental. I mean, that's the key because you're talking about where does ovocetrapib

01:08:16.980 fit in. We have the LPLA, we have the diabetes benefit, the small particles, but also the fourth

01:08:22.340 leg of that is for APOE4, you also have high LDL as well and lower HDL. So it becomes another good

01:08:31.140 drug to consider. That's 25% of the population. So you can see where ovacetrapin becomes

01:08:37.540 a really good add-on to statin option if you have any of those characteristics.

01:08:43.060 And so let's talk about the homozygous E4 patient. What is happening in their brain as a result of

01:08:53.120 the homozygosity? Because again, it's a very subtle change in their ApoE protein due to that

01:08:59.980 gene. And how does it manifest itself in the brain? Cholesterol metabolism. Right, right.

01:09:05.100 Yeah, it's very complicated. And everything I'm saying is there's still a lot of gaps in our

01:09:09.940 knowledge about what exactly happens with AboA4. But first of all, the brain is a very cholesterol

01:09:15.920 rich organ. If you look at, don't hold me to exact numbers, but 2% of the dry weight of the brain

01:09:22.140 represents 50% of your body cholesterol. And so you have tremendous amount of cholesterol in the

01:09:27.560 brain. And then you have a lot of fat, mostly DHA. So fat and cholesterol make up a big part of the

01:09:34.240 brain mass. I mean, it's a big part of what the brain is all about. And cholesterol is synthesized

01:09:41.000 very avidly by the neurons during your childhood growth phase. And then after you become adults,

01:09:48.360 the neurons stop making cholesterol. And myelin, by the way, is 50% cholesterol. A lot of cholesterol

01:09:53.820 goes into making myelin. And then astrocytes are involved in making, they continue to make

01:10:00.420 cholesterol. They're there to help repair neurons. And also neurons are the wiring and the making

01:10:08.220 the thinking and all that stuff that brain function does. But the astrocytes are involved

01:10:12.120 in the nourishment and the garbage collection, all that stuff. They're involved in maintaining

01:10:16.660 normal brain homeostasis. Now, the complicating thing is the blood-brain barrier completely

01:10:22.180 separates off the brain lipid metabolism from the rest of the body. There's no ApoB in the brain.

01:10:28.480 there's no ApoB. It's all ApoE and HDL in the brain. There's no connection between LDL

01:10:35.740 and the plasma and LDL. There's no LDL in the brain. So the brain has its own cholesterol

01:10:40.600 metabolism homeostasis. And the particles, the lipoproteins in the brain are HDL-like

01:10:47.580 particles that have ApoE. And so ApoE, in fact, if you do a liver transplant, your ApoE4 before

01:10:55.640 the donor and you give somebody an ApoE3 liver, they still have ApoE4 in the brain. The brain

01:11:00.680 stays, there's no connection. I had no, I had never even thought of that, Michael.

01:11:05.540 Yeah. So that's the connection. Do you think there's any clinical

01:11:07.840 significance to that in organ transplant? So the question is with gene therapy now,

01:11:13.260 if you could convert someone from a E4 to an E3, can you do it in the liver and correct the risk?

01:11:19.200 We don't think so. We don't think so. We don't know. It's still a thought.

01:11:22.080 That's super interesting.

01:11:23.500 Yeah. So ApoE4, there's a 2, 3, and 4. Most people are 3, 3s. There's about 20, 25% are 3, 4s,

01:11:33.300 and then about 2%, so 3% are homozygote E4s. So E4s, your homozygote about 10 times risk of

01:11:40.160 Alzheimer's. The 3, 4 is about 3 times risk. 3, 3s are average. And if you're a 2, which is the

01:11:45.900 best, your risk is lower and you have longevity. It's one of the best longevity genes to be an

01:11:50.860 ApoE2. So that's the breakdown. So when you have ApoE4, what happens? So ApoE is made by the

01:11:57.300 astrocytes. It appears that they have impaired cholesterol efflux and lipidation when you have

01:12:04.420 ApoE4. And so the neurons don't get the same nourishment or they can't clear cholesterol,

01:12:09.720 it becomes toxic. And once cholesterol becomes toxic, there's something called 24-hydroxycholesterol,

01:12:15.160 which can go through the blood-brain barrier, 24-7 and other cholesterol. So the blood-brain

01:12:19.700 barrier protects all that. But there's one cholesterol that goes through that's 24-hydroxy

01:12:24.200 or 27-hydroxy. They can transverse the blood-brain barrier. The ApoE4 patients have impaired

01:12:31.180 clearance of cholesterol. And then that sets up a toxic situation where it keeps inflammatory.

01:12:38.380 And then amyloid and tau follow. Although Alzheimer's first described Alzheimer's,

01:12:44.000 It was amyloid plaques, tau tangles, and basically lipid deposits.

01:12:50.780 And so when you have ApoE4, you can't clear the lipid as well, and that sets up the whole

01:12:57.100 Alzheimer's cascade.

01:12:59.740 And then you get Alzheimer's roughly, 1-4 gives you Alzheimer's about 10 years earlier,

01:13:05.300 so 2-4 is 20 years earlier, that kind of thing.

01:13:08.320 So it's a very well-established risk having an ApoE4 genotype.

01:13:12.800 Right. So that's the background about ApoA4, and it's a lipid gene. And what is important

01:13:18.860 about Obocetripib is that we had that genomic data, we had the animal data, and then we know

01:13:24.480 that HDL is the one lipoprotein that does interface with the brain. We don't think ApoA1 is actually

01:13:30.800 made in the brain. It comes from the periphery, ApoA1 being the protein carrier of HDL.

01:13:34.840 Yeah, I thought it was made in the liver. You're saying if the brain's ApoA1 is made

01:13:38.480 in the brain or made in the liver? Most of it, we believe, comes from HDL from the plasma.

01:13:44.060 Not from the brain, yep. It may be made in the brain, but we don't think so. We think it's made

01:13:48.480 and comes across the blood-brain barrier. And so we know that for sure in the choroidal plexus,

01:13:53.800 which is the blood-brain barrier interface... And this is just mechanical. It's because HDLs

01:13:59.180 are so much smaller than LDLs. Right. Small HDLs. But SRB1, which is the receptor, we talked about

01:14:04.220 that is heavily in the blood-brain barrier and the coronal plexus. So the only thing that really

01:14:08.960 interfaces with the brain from the outside of the brain is HDL. And so having high HDL can help

01:14:15.360 remove the excess cholesterol that APOE4s can't metabolize effectively. And also we prove this,

01:14:23.000 it delivers antioxidants to the brain. They're very important to reduce the inflammatory

01:14:27.060 response. We know that, and also maybe DHA even can be delivered by HDL, but certainly it's acting

01:14:33.880 like it does. It's the removal system, amyloid even. Amyloid can be removed by HDL. So we believe 0.88

01:14:39.880 that the HDL going up and interfacing with the brain can help mitigate the APOE4-related

01:14:45.360 Alzheimer's risk. What about in the non-4s? We think it's the same true. I mean,

01:14:50.400 even obviously non-4s get Alzheimer's, but it's later. It's later in life. It's 1.00

01:14:54.820 late 80s or 90s instead of 70s, 60s or 70s even. We think of the same things are happening because

01:15:00.780 again, if you stop making cholesterol in the neurons, your neurons are dependent on

01:15:06.000 astrocyte cholesterol function and so forth. And as you get older, you know, these things

01:15:10.980 tend to get less effective. So we think that that part could also be mitigated by

01:15:15.420 raising H2O through Obacetrapib's mechanism of action.

01:15:19.960 What do you say to folks who say, aren't statins causing Alzheimer's disease? Because

01:15:24.140 if the brain's cholesterol depot is the most important cholesterol depot in the entire body,

01:15:29.920 and we give a person a statin, and we know that some statins can actually cross the blood-brain

01:15:34.140 barrier, shouldn't statins, even though the clinical trials would suggest the opposite,

01:15:38.680 just on an individual basis, is there some risk that statins would be driving Alzheimer's disease?

01:15:43.820 I mean, just the opposite, actually. I mean...

01:15:45.720 Yeah, I guess my question, sorry, should be, why do we not see that? Why do we see the opposite

01:15:49.960 in the clinical trials?

01:15:51.100 About not protecting against Alzheimer's?

01:15:52.720 Yeah, why do we see that statins protect against Alzheimer's disease? You could argue mechanistically,

01:15:57.240 but if they're lowering cholesterol

01:15:58.520 and they make it into the brain.

01:16:00.120 Yeah.

01:16:00.720 We're not sure.

01:16:01.540 I mean, the big statin trial

01:16:02.660 that looked at Alzheimer's

01:16:03.620 didn't show a benefit.

01:16:04.580 It did not work.

01:16:05.620 That's not necessarily the means

01:16:06.860 that statins can't protect against Alzheimer's,

01:16:09.760 which I think they potentially can.

01:16:11.540 But in the study that was done,

01:16:12.720 actually done, it didn't work.

01:16:14.120 Yeah, it was a wash.

01:16:14.780 It didn't work.

01:16:15.780 But in a recent study analysis,

01:16:17.060 they showed that if you did have APOE4

01:16:19.380 and took a statin,

01:16:20.060 your dementia risk was lower.

01:16:21.560 But that's an observational study.

01:16:23.180 And that's complicated.

01:16:23.880 Yeah, I think that's fraught with issues.

01:16:25.540 I prefer these secondary analyses

01:16:27.120 on randomized controlled trials.

01:16:28.660 Yeah, yes.

01:16:29.400 When people ask about this all the time,

01:16:31.280 and you can say at least,

01:16:32.620 based on observational data,

01:16:33.860 there's no increased risk of Alzheimer's.

01:16:35.920 First of all, I just want to get them over the hump

01:16:37.300 that's not going to increase the risk of dementia,

01:16:39.000 because you hear that.

01:16:39.880 We really focus on that protection issue.

01:16:43.360 It's a hard sell to say it actually prevents

01:16:45.240 against Alzheimer's because we don't think

01:16:47.260 it gets to the core reason.

01:16:49.240 It can't affect the lipid metabolism in the brain

01:16:51.280 the way obocentropid potentially can.

01:16:53.860 I think what we see in the literature,

01:16:55.260 at least the way I look at it, Michael,

01:16:56.360 is that when you look at all forms of dementia, you do typically see in secondary analysis a risk

01:17:02.640 reduction with statin trials. And I suspect that that is picking up the vascular dementia

01:17:07.220 improvement. And because these studies are not powered to look at Lewy body dementia versus

01:17:12.860 vascular dementia versus Alzheimer's disease, it's a bucket of dementia. And so you're seeing

01:17:17.280 this reduction in vascular dementia, but it's just being counted as overall dementia,

01:17:21.880 but not pathology specific. That's absolutely true. I mean,

01:17:25.000 LDL is a dementia risk factor. But it might not be an Alzheimer's

01:17:28.740 risk factor through this mechanism. It's a stroke benefit. So we think that

01:17:32.040 obviously the elbow centropib having LDL lowering can benefit stroke,

01:17:35.960 but also we do believe or hope that we can show an Alzheimer's

01:17:39.860 prevention benefit. Talk about what we showed already, but we're playing another study

01:17:44.180 to look at it more carefully this year to look at the benefit of

01:17:48.360 with cetropib on prevention of Alzheimer's. So let's talk a little bit about what you saw

01:17:53.340 in your biomarkers to kind of validate all of these hypotheses. The first study that was a

01:17:59.220 pilot study showed that we did in fact reduce 24-hydroxy and 24-7, the sterols in the CSF went

01:18:07.180 down. You're not seeing that with statin, so it was a difference. And sorry, how did you, did you?

01:18:11.360 By CSF, right. We measured CSF in these patients and we saw that the 24 and 27-hydroxy went down.

01:18:18.360 We saw stabilization of biomarkers, and we saw cognition stable.

01:18:24.160 We also had a few anecdotal cases of patients really having quite a beneficial effect on

01:18:29.440 cognition.

01:18:30.520 It was a pilot study.

01:18:31.680 It was not powered, open label.

01:18:33.680 But we saw target engagement with 24 and 27 hydroxy going down.

01:18:38.280 We also, in another analysis, showed antioxidants in the CSF going up.

01:18:42.860 We had target engagement, that we are, in fact, removing these potentially toxic sterols

01:18:48.220 and improving antioxidant levels in the CSF.

01:18:52.240 So that's a proof-of-concept study that we did.

01:18:55.340 Tell me about those patients.

01:18:56.260 Did you say those had MCI?

01:18:57.480 MCI, all had MCI, yeah.

01:18:58.680 And just explain to folks clinically

01:19:00.420 how patients with MCI would be behaving.

01:19:03.080 They do cognitive testing.

01:19:05.060 There's all different types of scoring systems,

01:19:07.220 but when you do the memory testing

01:19:08.900 and functional testing,

01:19:11.020 they score showing they have impairment.

01:19:13.360 But what's important about Alzheimer's

01:19:15.160 and why we're focusing on it

01:19:16.860 is that we're learning now the abnormality starts 20 years before there's even mci there's a long

01:19:25.100 lag period between when we start seeing the pathology the amyloid and the tau and all the

01:19:31.600 biomarkers going up and actual mci and mci of course then has different levels until you get

01:19:37.580 to the functional impairment where you're now this is actually subjective cognitive impairment

01:19:42.180 where people think they have impairment when they do the testing they're still pretty okay and then

01:19:46.100 you have those that are truly cognitively impaired. Yes. So just so folks understand what we're

01:19:50.820 talking about here, which is Alzheimer's is not a disease of old age. It's a disease of middle age

01:19:56.240 that presents in old age. And that's saying exactly what you just said in a different way.

01:20:00.700 I mean, that's in some ways a frightening thing to think about, but it's just the reality of the

01:20:04.200 disease. And it also speaks to the lens through which we want to consider prevention. I would

01:20:08.680 argue the same is true of cardiovascular disease. It's a disease of middle age. It presents in old

01:20:13.040 age for most people. Unfortunately, for someone like your father, it presented in middle age,

01:20:16.800 but it starts effectively the moment you've got these LDL particles circulating. And so

01:20:22.100 what would you say is the canary in the coal mine that is measurable? Would you say that it is the

01:20:28.060 presence of P-tau? Yeah. That's what's exciting about the field right now. Like I said, I came 0.96

01:20:33.320 from the LDL world. Your LDL became a great causal. When I was involved, they weren't even

01:20:38.740 sure LDL was really treatable when I started. I mean, they weren't even sure that you could

01:20:42.980 reduce LDL and prevent heart disease in the 80s. Yeah, I mean, so I mentioned already,

01:20:47.940 once you already have heart failure, lowering LDL doesn't seem to matter anymore.

01:20:52.060 With Alzheimer's, I think we're seeing the same thing. Once you have MCI,

01:20:55.820 and once you already have functional impairment, it's already too late. The brain is already

01:20:59.700 shrunk and you got neurodegeneration. And so I think we're way too late in how we treat

01:21:05.040 Alzheimer's. So we can find a prevention therapy. And do you think that's why your statement earlier

01:21:10.800 about how much this is a graveyard for pharma is basically the case, which is on the one hand,

01:21:18.480 it's very difficult to develop drugs for primary prevention because the trials would take too long.

01:21:24.780 So you have to treat an active disease. But in this particular disease, if you treat the disease

01:21:30.740 as dementia, you're hosed. It's almost like had cardiologists had the misfortune of trying to

01:21:37.280 use lipid-lowering therapy to treat heart failure, we may never have developed lipid-lowering drugs.

01:21:43.340 That's right. They would start with heart failure, it would all have been failures. We did heart

01:21:46.060 failure trials and it didn't work. So it stands. So that's kind of the interesting thing, isn't

01:21:50.300 it, right? Which is with Alzheimer's disease, we don't have the equivalent of the MI,

01:21:54.460 the survivable MI, where if you intervene from a secondary prevention standpoint,

01:21:59.480 we can develop those things because all of those tools that were developed for secondary prevention

01:22:03.520 can now be deployed for primary prevention and have the maximum impact on saving lives.

01:22:09.320 So it's a very unfortunate consequence of the pathology, though an understandable mistake on

01:22:15.000 the part of pharma because that's the hand you're dealt. You have to do clinical trials.

01:22:19.040 Right, right, right. I hope that PTAL 217, which is what we found in our, can ultimately become

01:22:24.340 a biomarker for regulatory approval. It's just we're not there yet.

01:22:27.840 How does it differ from PTAO-181? And let's talk about them through the lens of a little bit what

01:22:33.900 they are biochemically, but also what do you think is the difference clinically between when you're

01:22:38.720 measuring PTAO versus AB4240 and some of the other biomarkers that are becoming more common now?

01:22:45.320 Sure, sure. It's a little bit of a background on these biomarkers. So they're all, when we did our

01:22:49.840 first pilot, PTAO-217 wasn't available. We had 181, we had other ones. We convened a big advisory

01:22:57.340 board of Alzheimer's experts. I'm not an Alzheimer's expert. I'm trying to become one

01:23:01.880 now, of course. The panel said, this is a new biomarker called PTAL-217. If you can show over

01:23:07.800 a year with our existing study within Broadway that you're reducing PTAL, that would be a great

01:23:13.180 validation that what you saw in your pilot study is truly an effect. That's what we did.

01:23:18.560 So PTAL-181 and 217 are both TAL measurements, but they correlate with amyloid in the brain.

01:23:24.940 they're very strong predictors of amyloid on PET. And so if you look at centeloids of amyloid,

01:23:32.120 which is the gooky stuff in the brain, the protein aggregates and so forth, the misfoldings, 0.89

01:23:38.620 the amyloid tau 217 and 181 predict that very well. In fact, in some ways, I think the PET

01:23:44.400 could be archaic relatively soon based on these biomarkers.

01:23:49.040 That's a big deal given the radiation and the cost associated with those tests.

01:23:52.300 Right. From what I look at the literature, I mean, they're as good as the PET almost

01:23:56.100 on area under the curve. So they're very good at predicting amyloid on the PET. They're

01:24:01.340 very good at predicting progression of disease from normal to MCI. They're very good at looking

01:24:06.100 at MCI to full-blown stage 2 or 3 Alzheimer's. And even progression of Alzheimer's is also

01:24:13.300 predicted by PTAL-217. So PTAL-217 occurs a little bit earlier than 181. So it's a little

01:24:20.380 but earlier in the process. So it picks up earlier. There's also a new one called brain-derived

01:24:24.380 P-tau, which is maybe even better, but that's still early. So that's a tau. It's amyloid

01:24:30.200 first and then tau. And there's this conversion of amyloid to tauopathy, which is a critical

01:24:37.500 transition in the disease process where the amyloid is that kooky stuff. And then tau itself 0.97

01:24:45.520 is the one that really drives the neurodegeneration that occurs later. So it's first amyloid,

01:24:51.840 then a lot of tau, and then you're basically, unfortunately, it's too late.

01:24:56.500 P-tau-217 is not a CLIA-approved assay, is it?

01:24:59.980 Well, this 217, yes, it is CLIA.

01:25:02.640 It is.

01:25:02.920 But it's divided by amyloid 40, beta-40. That got approved this past year.

01:25:09.140 Sorry, you mean the ratio of-

01:25:11.300 P-tau to amyloid is approved.

01:25:14.380 It is approved.

01:25:15.180 Yeah. Okay. So AB 4240 is CLIA approved by itself? Right.

01:25:20.260 PTAL 181 is CLIA approved by itself? Yeah. If it's not approved already,

01:25:24.140 it'll soon be approved. It's being used a lot. 217, not by itself, but in ratio is an approved.

01:25:30.420 Right. Right. Okay. And do you think that the ratio of PTAL 217 to AB 4240 offers a benefit

01:25:38.040 over PTAL by itself? Well, the data is actually kind of mixed. There's actually PTAL 217 alone

01:25:44.700 looks awfully good. Whether the ratio is better, I've never...

01:25:47.740 So you guys did your own assay for that?

01:25:49.580 We do standard assay for 2-TAL-217. It's called, it's Quantarix, which is one of the standard ones.

01:25:55.300 There was something called Precivity, which does the P-TAL-217 with A-beta 4240. But we

01:26:00.700 looked at all the biomarkers. We looked at P-TAL-217, P-TAL-181. We looked at GFAP, which is,

01:26:08.940 I call it glial factor activation protein. It's not the right name. It's glial

01:26:13.240 fibrillary acidic protein. It's a glial activating biomarker. Also very good at early disease

01:26:20.580 protection. And then we had the ratio, 42-40 ratio, the no filament light, which is a later stage

01:26:27.120 inflammatory biomarker. I mean, you see a lot in injuries, but as well as multiple sclerosis or

01:26:33.820 ALS, those kinds of things. There's disease in Alzheimer's too, but later in Alzheimer's.

01:26:38.600 We had our whole study, Broadway, 2,500 patients, and we had a sample that we could do.

01:26:44.240 It was pre-specified analysis.

01:26:46.220 We did analysis at the beginning and then 12 months at the end of all these biomarkers,

01:26:51.280 with PTAL-217 being our critical one to look at.

01:26:55.200 We found that overall, there was a reduction in PTAL-217 going up in the overall population.

01:27:02.560 But, which I think is most exciting, is that if you look at what we were worried about

01:27:06.680 the most was that we had a population of heart disease patients that were not that old. The

01:27:11.220 average age was 65. Are we going to really see anything in a year? Because if you know the data

01:27:16.520 on these biomarkers, it takes many years for these things to transition into something really

01:27:21.980 advanced. And so that's the key. But even low levels of these biomarkers are really very

01:27:27.220 predictive of future Alzheimer's risk. And so what we found was overall it worked.

01:27:32.360 If you had people that were older, it worked even better. If you had people that had E4,

01:27:36.500 it worked even better. And if they were 4-4, homozygotes, we saw this profound benefit on not

01:27:42.220 just PTAL 217, over 20% difference from placebo, but all the biomarkers improved. 181, amyloid 4240,

01:27:51.220 GFAP, neurofilaminolite, all got better with Obacetrapib. And that's what we published in

01:27:57.000 our paper. So we felt that this data really confirmed that Obacetrapib was in fact active

01:28:04.120 in potentially preventing Alzheimer's disease. Now, how do we know how to interpret whether

01:28:10.760 those will end up being, maybe the answer is we don't, but do we have a sense of if this could

01:28:15.980 translate to a clinically meaningful benefit, to a reversal of disease, a halting of disease,

01:28:21.360 something of that effect? Or is it your belief that that could probably only happen if we

01:28:25.860 establish causality through trials like this, and then take the same playbook that we took

01:28:30.800 with cardiovascular disease, which is move up 20 years, start treating before any of these things

01:28:36.180 are even remotely present, and just hope that that alone prevents even the development of the

01:28:42.360 pathology. The answer is I don't know. The field is evolving, and the regulators have to help out

01:28:47.720 here because right now PTAL 217 cannot be an endpoint for regulatory approval. It does require

01:28:54.000 improvement in cognition, which might be an impossible thing to show.

01:28:58.600 Unless you took for decades.

01:28:59.740 What we can say is that for us, we're going to go ahead with another study. We do know that

01:29:05.220 the APOE4 community, there's a new alliance of patients out there that are advocating for

01:29:11.360 treatment. It's like any genetic disorder. They don't want to wait 10 years. They want something

01:29:15.720 now for treatment.

01:29:17.340 Don't you think it'll just be used off-label for that reason anyway?

01:29:20.320 Well, we can't unfortunately...

01:29:22.020 you have nothing to do with that, but I'm just saying like, look, people were using PCSK9

01:29:25.860 inhibitors off-label for LP little A reduction the moment it became clear that it was lowering

01:29:30.160 LP little A 20 to 30%. Now we have no idea if lowering it 20 to 30% mattered, but it seems

01:29:36.340 that it's not unheard of for people to take off-target use. There are plenty of people

01:29:40.480 using GLP-1 agonists today for neuroprotection, even though that would still be considered sort

01:29:45.020 of off-label as well. Yeah. Unfortunately, I can't talk about that as a CEO of a biotech

01:29:48.960 company about off-label use. But I think the key is that I think PTAL 217 with enough advocacy

01:29:56.000 can become a regulatory endpoint for this reason. I mean, it's so good.

01:30:01.620 Let's take the flip side of that. The Feynman argument that in science, the easiest person to

01:30:05.760 fool is yourself, or the goal of science is not to fool yourself and you're the easiest person to

01:30:09.780 fool. So what would be the flip side argument? What would be the case for PTAL ultimately does 0.96

01:30:16.280 not translate to clinical benefit. What would be true? If we had a crystal ball that showed us

01:30:20.900 that, what would you sit here and say today as to why? What would the explanation be?

01:30:24.920 You're proving a biomarker, but you're not treating the underlying disease.

01:30:28.240 But it is the disease. I mean, that's the actual tau that's in the brain that leaks out into the

01:30:31.980 plasma. So it is a measure. If we're using these anti-amyloid antibodies, or now they're working

01:30:36.300 on anti-taal antibodies, and they've still got a ways to go, obviously those anti-amyloid antibodies

01:30:41.440 have some very modest benefit, but not for APOE4 homozygotes. So I do feel that the homozygotes

01:30:48.820 in particular, which are the ones that we know 100% go on to get Alzheimer's if they live long

01:30:53.320 enough, like a lot of other rare diseases allow these surrogates to get approval based on a

01:30:59.580 surrogate that which may or may not translate to a clinical benefit. Yeah, I'm thinking less of it

01:31:04.200 from a regulatory question and more just from a biologic question of it's easy to get fooled.

01:31:11.340 We can all be fooled. I mean, HDL fooled everybody for decades. For decades, people were chasing the

01:31:18.380 wrong thing and they had all the epidemiology to back it up. But at the end of the day,

01:31:22.980 it didn't matter. I'm not saying this is the case. I'm not even suggesting that

01:31:27.440 it looks likely that that's the case. And I certainly hope it's not the case,

01:31:31.700 But I always find it interesting to imagine if we're sitting here in 20 years having a

01:31:36.080 podcast talking about, oh, that whole P-tau thing that took us down a rabbit hole, what 0.97

01:31:40.980 a waste of time and money that was. 0.95

01:31:42.420 We didn't save a single life.

01:31:44.120 I wonder what the story would be.

01:31:46.060 What would be the revisionist explanation of that?

01:31:48.420 Here's what I'm also saying.

01:31:49.400 In the last few months, there's been some supportive data that came out that's helpful.

01:31:53.060 One is the GLP-1 trials, Evoke and Evoke Plus, and no improvement in cognition, no effect

01:31:59.480 on P-tau 217.

01:32:00.660 On the other hand, the Trailblazer anti-amyloid trials literally showed a pretty high correlation

01:32:05.880 between improvement in cognition and P-TAL-217 going down.

01:32:10.220 We're getting more data and we're going to do our own study.

01:32:12.900 We are going to do a study with 300 patients that have pre-Alzheimer's, you want to call

01:32:18.360 it.

01:32:18.700 They have evidence of higher TAL levels based on their APOE.

01:32:22.940 With MCI?

01:32:23.780 No, not yet.

01:32:24.540 We want to get it before.

01:32:25.360 Like I said, we think MCI is too late.

01:32:27.280 And so we got to get it before MCI.

01:32:29.660 MCI is like early heart failure.

01:32:31.120 Right, right.

01:32:31.720 We know the brain has already shrunk and all that stuff.

01:32:34.520 Inflammation has set in.

01:32:35.660 The cascades, the talopathy is already well underway.

01:32:39.460 I mean, so we can't wait that long and hope.

01:32:41.900 But we can hopefully show that with our next study, we confirm all this.

01:32:47.580 Even though it was not a prospective trial, it was a pre-specified analysis.

01:32:51.620 We confirm all this, especially in the APOE4 homozygote patients, the profound benefit.

01:32:56.620 What we got most feedback from was,

01:32:59.420 it's just this data is too good to be true.

01:33:01.180 How can this be true?

01:33:02.580 It wasn't only 30 patients.

01:33:04.000 It had homozygote E4.

01:33:06.080 But it was so statistically significant.

01:33:08.060 It was accepted for publication and all that kind of stuff.

01:33:10.960 So the question is, what do we need to do to convince people?

01:33:13.300 But right now, the world, Alzheimer's world is very fixated on amyloid.

01:33:17.260 I am surprised to hear that, Michael,

01:33:19.320 given the number of bullet holes that have been fired through so much of that, right?

01:33:25.200 I mean, including some of the fraudulent work that was done in the last decade on this front.

01:33:30.060 Why do you think that they're still clinging so hard to that story?

01:33:33.900 It's all about hope because it's such a devastating disease.

01:33:37.100 I don't want to comment on the anti-amyloid drugs because it looks like some of the data

01:33:41.900 might be encouraging, but it's a tough sell.

01:33:44.640 Even you got to take it.

01:33:45.680 You got to worry about brain bleeding and all that kind of stuff.

01:33:47.820 It doesn't work on homozygote E4 either.

01:33:50.180 Because of ARIA or just doesn't?

01:33:51.700 ARIA is what I'm talking about.

01:33:52.860 It doesn't work as well on the cognition improvement.

01:33:54.780 And you have the higher risk of ARRIs. They don't give it. There's supposed to be a contraindication

01:33:58.860 to use it in E4 homozygotes. That could be the same question. Is it that anti-amyloid drugs

01:34:04.220 could be beneficial, but they have to be started 20 years sooner? And that's what they're trying

01:34:07.800 to do. I mean, they're trying to go earlier. I don't know if it's early enough is the question.

01:34:11.600 Can you really convince somebody who's 40 years old, whatever it is, to take an infusion to

01:34:16.580 prevent amyloid? Especially if the side effects are potentially catastrophic. Yeah, right, right.

01:34:20.780 so that's going to be a tough sell. So we're trying to find over six, we're trying to pick

01:34:24.600 maybe age 60, because we know from our Broadway data, who are the more rapid progressors? Who

01:34:30.240 does TAL go up the most in? And it is obviously E4 is the older, even the diabetes that makes

01:34:35.940 the amyloid. So I think that we're going to do another study and then hopefully that'll

01:34:39.680 set the stage for a phase three trial where we could look at conversion of normal MCI preventing

01:34:45.020 that conversion. That'll be a long-term study that we hope to do pending on the results of

01:34:50.360 the next trial. Well, this is incredibly exciting. So again, the hope here would be that we see

01:34:56.100 approval in the United States following the European approval by a period of a year or two

01:35:01.220 for cardiovascular disease. And in parallel, we see the right clinical trials being done

01:35:06.900 on AD prevention. And this would make it a first. There would be no approved drug

01:35:12.520 out there for AD prevention. Yeah, there's so much more we could talk about on this,

01:35:17.220 but I do want to spend time on a topic that we talk about from time to time, but I have a feeling

01:35:22.980 you're going to add a lot of value to my somewhat limited knowledge, which is the most up-to-date

01:35:28.240 thinking on the role of fish oils, DHA and EPA. But you want to talk about it through cardiovascular

01:35:33.740 disease first or talk about it through Alzheimer's or both? I think the cardiovascular part,

01:35:39.960 I'll give you a quick answer to that because I was involved in developing a omega-3 for heart

01:35:44.580 disease called Epinova. We went to an outcome study, and it didn't work, but it was DHA and

01:35:50.180 EPA, but mostly EPA had some DHA. And that was four grams? How much?

01:35:56.100 That was four grams, yeah. It was called the strength trial. I think what I learned,

01:35:59.400 because I've always been a big advocate for omega-3s for triglyceride lowering, and there's

01:36:04.040 really two reasons why omega-3s can reduce risk. One is triglyceride lowering, and the other is

01:36:07.960 by antithrombotic effects. EPA antagonizes the arachidonic acid pathway and has like a

01:36:14.580 antithrombotic effect, anti-inflammatory effect. And so there was really two reasons to think why

01:36:19.200 omega-3s could lower risk. And one was through the EPA antithrombotic effect, the other is

01:36:25.240 triglyceride lowering. And so what Strength was set up to do was really a triglyceride lowering

01:36:30.180 trial, and it didn't work overall. There was a hint of benefit in the secondary prevention arm,

01:36:35.320 but it was 50-50 primary. And so if I had to do it all over again, I would have made it all

01:36:39.120 secondary, just like mostly the REDUCE-IT trial was, which is the EPA-only trial. So the REDUCE-IT

01:36:45.580 trial showed a benefit, and it could have been, likely had some effects of the mineral oil placebo

01:36:51.460 tilting it a little bit. I've heard this explanation before, but say a little bit more

01:36:55.900 about why mineral oil as a placebo could have been harmful. But you'd actually look at the

01:37:00.000 inflammatory signaling and increased inflammation. And so it had some inflammatory, Paul Ritker

01:37:05.400 helped now analyze that from the Reduce It study. It also may have affected absorption of statins

01:37:11.240 because LDL levels went up. LDL's level went up about 10%, so 9% to 10%. I think it showed that

01:37:17.720 EPA works. EPA works. It may not have been the 25% that people think it was, but it still worked.

01:37:24.380 You take all that away, it still worked. EPA only worked. And then the strength trial didn't work

01:37:31.180 overall. It may have worked on secondary prevention in a, not post hoc, but in that

01:37:34.840 population. And the placebo in strength was what? It was corn oil. I see. Just so folks understand

01:37:39.920 what we're talking about here. In the EPA only trial, the placebo was mineral oil, but the

01:37:44.860 hypothesis was that the mineral oil itself probably increased events slightly, which made the EPA look

01:37:50.580 better alone than it might've been. But in the combined trial, the DHA EPA trial, the placebo

01:37:57.780 was corn oil inert and you're doing apples to apples. Yeah, corn oil might have its own.

01:38:03.040 We had a hard time coming up with a placebo because it's not easy to find an oil placebo.

01:38:06.620 Mineral oil, it isn't truly inert, but it does have that intestinal effects.

01:38:10.880 I think the mineral oil controversy is, I still think reduces, shows a benefit for pure EPA.

01:38:16.420 Now then we had the prominent study, which is a triglyceride lowering trial

01:38:19.660 with a fibrate, PMO fibrate showed no benefit. It's very similar to the strength trial

01:38:25.440 population. What that says is that EPA does work on reducing events.

01:38:31.300 But not through triglyceride lower?

01:38:32.520 Not through triglyceride lower. It's more for the antithrombotic, anti-inflammatory effect,

01:38:36.720 which brings me to DHA. So DHA, I still think has a lot of health values that we should

01:38:41.860 talk about, especially for the brain, because the brain is predominantly DHA. That's the fatty

01:38:46.780 acid in the brain. Now, the problem is with DHA, we tried DHA, supplementations on cognition

01:38:53.120 didn't really show much. You've had the same you've seen here, and he's talked a lot about,

01:38:57.820 but 8.4 people, one of the other things is they don't get as much DHA to the brain. They have a

01:39:02.600 true DHA deficiency in the brain to some degree. A new discovery, which is very exciting, is the

01:39:08.280 MFSDA2 transporter. It's a transporter across the blood-brain barrier for DHA. It turns out that DHA

01:39:19.600 prefers the lyso-PC form, the lyso-phosphatidylcholine form of DHA. That's the preferred

01:39:26.760 substrate for DHA to cross the blood-brain barrier. And so now for the first time,

01:39:32.320 you can now get lyso-PC-DHA. It's coming. It's on the horizon. And so we're actually,

01:39:38.240 as part of Jocasta, we're going to be doing studies with lyso-PC-DHA on showing getting

01:39:44.200 across the blood-brain barrier. Completely independent of the cloth at work.

01:39:47.300 Right, exactly. Yeah. That's coming. And I've been on omega-3. One of the things that really,

01:39:52.560 until this receptor was identified, we couldn't understand why DHA doesn't get into the blood

01:39:57.740 brain barrier like you think. Most of your DHA has to come from the blood. It has to get there.

01:40:02.540 It has to get there. Your brain cannot make that much DHA from other fatty acids, if at all. I mean,

01:40:09.060 I think it can make some, but you have to get DHA from the periphery.

01:40:12.940 So your belief today would be that if an individual is not consuming much fatty fish,

01:40:19.720 you measure their red blood cell membrane concentration of EPA and DHA, and it's very

01:40:25.220 low, 4%. And they were to supplement with a very high quality over-the-counter EPA and DHA,

01:40:31.720 and they were to get that to 10%, your belief is they are probably lowering cardiovascular

01:40:36.460 disease risk, but probably not having an impact on Alzheimer's disease risk,

01:40:40.900 given the inability to get the DHA in the brain?

01:40:43.980 You have to get the Lysopc form to get to the brain.

01:40:46.760 You have very limited capacity to convert that into the Lysopc form.

01:40:50.120 Yeah, absent another...

01:40:51.460 Yeah, so you have to make this Lysopc form,

01:40:53.640 which you can do with either a fish-based oil or a krill-based oil.

01:40:58.540 You can make Lysopc DHA.

01:41:01.100 So that's the next hopeful excitement around the DHA and brain health

01:41:06.200 that I think needs to be further evaluated.

01:41:09.000 There is no version of that on the market today.

01:41:11.220 It's coming, it's coming, yeah.

01:41:12.760 But will it be a drug or will it be a supplement?

01:41:14.780 No, there'll be a dietary supplement, yeah.

01:41:16.780 Do you know who's making that?

01:41:17.860 One is already available for krill oil base,

01:41:19.640 but there's one with fish oil base coming.

01:41:21.780 In Jocasta, we're gonna do the science behind that as well.

01:41:24.480 For a drug or for a supplement?

01:41:26.120 For a supplement, yeah.

01:41:27.040 Okay.

01:41:27.720 You've been involved in so many clinical trials

01:41:29.560 over the course of your life,

01:41:31.160 and there's so much discussion about AI and medicine

01:41:35.480 and the impacts it can have,

01:41:37.800 But I'd have a hard time thinking of a more important impact that AI could have on medicine

01:41:43.140 than if it could speed up the time it takes to do a clinical trial by a log and if it

01:41:48.760 could reduce the cost by a log.

01:41:50.080 So if you could take it from $4 billion a drug to $40 million, or frankly, if you could

01:41:57.040 take it to $400 million, right?

01:41:58.500 And if you could take it from 10 years to one year, you could change the course of human

01:42:02.760 history. Where do you see AI factoring into what appears to be the long pole in the tent,

01:42:08.620 which is doing clinical trials? It's a hard thing until there's

01:42:12.800 regulatory reform. I hate to say it. I mean, I know that's a tough thing to advocate for because

01:42:18.400 first of all, I have to figure out how do you do the data? It's still like everyone's tried to say,

01:42:23.160 can we make clinical trials a lot more streamlined, less data collection? But also

01:42:27.960 to me, I wrote a paper about this. Why can't we look at clinical trials based on using AI who,

01:42:35.580 if you took the drug the right way, how well did it work? Instead of, we have to get diluted by

01:42:41.420 everyone who stopped the drug, who wasn't compliant. Yeah. So in other words, make the

01:42:45.520 efficacy not based on intention to treat, but on actual compliance. Right. And that would be a huge

01:42:51.820 difference already. Your effects are diluted so much by, especially for long studies,

01:42:57.780 you have to drop in. So you can use AI modeling even to show that for those who took the drug,

01:43:04.120 took it the right way, the benefit worked well. When you look at a trial like either Broadway

01:43:10.440 or Rose 2, what percentage of the patients enrolled in those trials do you believe were

01:43:15.040 nearly 100% compliant? We actually, this is what's called on-treatment analysis or

01:43:19.800 kind of looking at PK, you know, who has the right, taking the right drug. It always works

01:43:23.680 so much better. But in a drug like a once a day pill, which is as easy as it gets for the most

01:43:29.700 part, I guess maybe an injectable every few weeks would be similarly compliant. What is expected

01:43:35.480 compliance? You want to see 80%. That's kind of the benchmark. Yeah. In fact, when you do your

01:43:40.640 protocol analysis, you say anyone below 80% compliant. But the FDA will never accept her

01:43:46.660 protocol analysis. Yeah. Well, they should. I mean, that's the thing. When I put it in this 0.51

01:43:50.560 paper, I said, if we could find a way to streamline drug development, that's one thing. But AI could

01:43:55.720 help a lot even more. I mean, we could do a study, instead of being 10,000, it could be 2,000. It

01:44:00.080 just took the people that, like, you get to your log reduction. That's just one example. I mean,

01:44:04.700 I don't know how much longer we can keep doing LDL trials with different MOAs. It's getting

01:44:09.520 harder and harder to do these studies. Because of the ubiquity of drugs on the market already,

01:44:14.260 and then the amount of LDL lowering

01:44:16.160 that's already in the baseline?

01:44:17.380 Right. You have to be ethical.

01:44:18.680 You can't withhold treatment to patients.

01:44:20.800 We have a lot of patients that can access to other drugs

01:44:24.080 and you just can't deny that to patients in a trial.

01:44:26.880 It's going to be harder and harder to do these studies.

01:44:28.940 So let's come back to statins for a moment.

01:44:31.000 What do you think is the best explanation

01:44:33.120 for the increase in the incidence of type 2 diabetes

01:44:35.720 that we see in statins?

01:44:36.700 Again, it's not small, but it's a real signal.

01:44:38.980 I'm sorry, it's not large, it is small.

01:44:40.560 It's a real and undeniable signal.

01:44:42.220 And by the way, I would say that

01:44:43.240 it's not just patients going on to get type 2 diabetes, it's insulin resistance.

01:44:47.300 Patients who don't progress all the way to type 2 diabetes, we do see an increase in

01:44:49.900 insulin resistance. What do you think is explaining that?

01:44:52.040 Well, first of all, it looks like all LDL-lowering drugs, even PCSK9s, have a little bit of a

01:44:57.660 diabetes signal, except for obocetrapin. I mean, just the opposite.

01:45:00.760 Which is the opposite.

01:45:01.580 Yeah. It is about LDL.

01:45:03.640 But that's not entirely true. Don't bile acid sequestrants, which lower LDL,

01:45:07.180 albeit somewhat weakly, show an improvement in glycemic control?

01:45:10.920 Right. I published a paper on that too. I mean, so, but I'm talking about the Mendelian

01:45:14.900 minimization work. We look at diabetes and LDL, you see a consistent, except for the

01:45:20.060 bile acid sequestrant part. I'm not sure what the drug matching for a gene would be,

01:45:24.840 but it's not that often. If you can't, the bile acid sequestrant gene matching is not

01:45:28.340 an easy one, but you're right. But bile acids are very energetic. They take a lot of steps

01:45:33.740 to make a bile acid. So you look at when you have a bile acid removal, you got to make

01:45:39.100 a new one. So that creates energy requirements. So you can see how blocking a bile acid from

01:45:46.200 being reabsorbed, which is why you reabsorb it so you don't have to worry about making it again,

01:45:50.080 can ultimately reduce potentially blood sugar because it is lining up energy that you need to

01:45:56.200 replace. That's my simple thinking about bile acids. Are there any patients that are in your

01:46:01.620 clinic that are still on a bile acid sequestrant? Yes, there's some, yes. These are patients that

01:46:05.880 can't tolerate any other drug or can't afford it?

01:46:08.300 Or there are any familial hypercholesterolemia and they need, we have some, yeah.

01:46:10.500 And how much LDL lowering do you get?

01:46:12.160 About 10 to 15%. It's not great.

01:46:14.640 And the side effects?

01:46:15.800 The GI side effects. There's some that are better. We use them a lot. We don't use them

01:46:18.880 that much anymore once azetamide came into play because it's pretty much the same as azetamide,

01:46:22.980 but azetamide is a lot easier to take.

01:46:24.980 And does azetamide have an increase in insulin resistance associated with it?

01:46:29.260 The data suggests clinically not, but I think genomically there is a slight signal there.

01:46:34.300 the MPC-101, the knockout types. But certainly PSK9 inhibitors genetically have that.

01:46:40.560 So what do you think is the mechanism for that?

01:46:42.100 I honestly don't know. I saw this recent data, which I think you talked about,

01:46:47.180 where the change in the type of bile acids that you see could be a mechanism. But there's been

01:46:52.220 other things too. I mean, the beta cell preservation and so forth. I don't know.

01:46:57.600 It's been an enigma about why. I don't think anyone knows for sure. But we do know a couple

01:47:02.080 things that I think are helpful. One, it is dose-related. It is age-related. It's weight-related.

01:47:07.820 So if you have those high-dose, older, obese, you know, more common. To get our point earlier about

01:47:14.540 not being on high-intensity statins, why is it worth the six, whatever it is, four to six percent?

01:47:20.540 Why not go at a lower dose and lower risk? What's more about how do you mitigate the risk? No,

01:47:25.440 the MA, it is, we haven't figured it out. I mean, this might be a reasonable explanation,

01:47:29.540 But up until now, we haven't really figured out why the glucose issue is a problem.

01:47:35.040 Yeah, although it's important, as we noted in our piece, that the asymmetry is still

01:47:39.960 enormous.

01:47:40.940 So the risk reduction from the lipid lowering on cardiac events is a far bigger magnitude

01:47:48.220 than the increase in the risk of type 2 diabetes along with its expected manifestations of

01:47:55.040 that.

01:47:55.300 But the hope, of course, is that as more and more physicians become aware of that, understand it,

01:48:01.200 move to combination therapy to lower statin dosing, I think it becomes less and less of an issue.

01:48:07.620 That's the point. The net benefit is still so great. Unfortunately, it's a big discussion with

01:48:11.800 patients very frequently. I don't want diabetes. Why are you giving me this? Your explanation is

01:48:17.240 what we try to talk to them about. It's the overall net benefit is still very much in favor of taking

01:48:22.000 the statin. How many other C-tap inhibitors are in the pipeline at the moment? To our knowledge,

01:48:26.800 no others. We're it. If we're successful, there are going to be a lot more, but it's very hard

01:48:30.800 to beat ovocetripib. It's got a low dose, extremely well-tolerated in our clinical trials.

01:48:37.440 Remind me, the molecule that when you guys bought it came from?

01:48:40.920 Well, Amgen through Mitsubishi, a Japanese company, first discovered it. It was sold to

01:48:46.300 Amgen. What were they doing? What was Mitsubishi doing when they were discovering it? Were they

01:48:50.420 doing it specifically for the purpose of following the earlier C-DEP inhibitors and then they sort

01:48:55.040 of abandoned it. Well, they wanted to make a better CTP inhibitor. So one that had no fat

01:48:59.240 uptake, much lower dose, more potency, more bioavailable. And so they gave us, we got a

01:49:04.800 great drug. We got a great drug to work with. It's going to be very hard to match the profile

01:49:09.320 that Obacetrapib has from a pharmacokinetic perspective. And to give folks again, just a

01:49:14.580 sense of what does it take to develop a drug? So if you look at the money you paid to acquire the

01:49:19.560 drug, what you had to raise through the clinical trials and what it would take to get to approval.

01:49:23.860 What's the approximate dollar amount to do that? It's well over a billion, but this started

01:49:28.340 counting what it was cost up until we got it. I mean. Yeah. It's a billion of New Amsterdam plus

01:49:33.660 all the money that came before it to get it into phase two. Yeah. So the average drug right now

01:49:38.600 is about three to $4 billion to approval. That accounts for the failures too, but it might be

01:49:44.600 even more if you count for the failures. It's a lot. It's a lot of money. Like I said, it's a

01:49:49.460 high risk, high reward investment. So been able to get to this point and we're hoping to have the

01:49:56.120 drug to patients as soon as possible. What do you think overall of the landscape of biotech

01:50:00.920 investing? I mean, we live in this world now where AI is clearly the most interesting and

01:50:05.940 exciting thing to invest in. Capital is flying into AI. It doesn't really matter that lots of

01:50:13.200 people are saying, hey, look, this is a bubble. We need to be a little more cautious with how we

01:50:16.980 do this. This feels like 1999 for the internet, et cetera, et cetera. But it's so interesting

01:50:22.480 when you look at biotech companies that, I mean, it's very, very difficult to raise money.

01:50:27.920 Where do you think we are in the biotech life cycle? And do you think biotech is just

01:50:32.180 forever going to be complicated because of the time horizon?

01:50:36.080 The latter is true, of course, but what's really important is the macro issues is that we have

01:50:41.220 all these big pharma companies that have big sales forces and companies, and they have pipelines that

01:50:48.660 are the big, big gaps and the patent lists are coming. All of them, in order to grow, they have

01:50:53.660 to get new product. I think in the way the system is set up, you know, everything ultimately goes

01:50:57.960 generic. And so if you see where that is, if companies even want to grow three or 4% per year,

01:51:04.460 which is pretty minuscule growth, they got to add a lot of new products to their pipeline.

01:51:10.160 The way it's worked out is the big companies also have become pretty dysfunctional on innovation

01:51:15.560 about developing new drugs. The small companies can be a lot more nimble, a lot more flexible,

01:51:20.860 take more risk, and don't have to worry about corporate kind of hierarchy to make a decision.

01:51:26.360 We feel that biotech is right now in a really good position, and I think AI is going to help a lot.

01:51:31.440 I hope we can figure out a way to make clinical trials a lot more cost-effective and lower cost,

01:51:36.180 but even drug discovery. China, for example, has just become a huge investment in, for every

01:51:44.400 chemist in the United States, just like 10,000 in China. They have the capabilities to really

01:51:49.280 surpass anything that we can do from a drug development perspective, just by the person

01:51:55.460 power that they have. But that sets up an opportunity. I mean, I still think we have

01:51:59.520 the science and the innovation, and we have the creativity that could lead the world and coming

01:52:04.900 up with the new ideas. But China is going to be able to have us be able to take it to the next

01:52:09.140 level to how they get a drug produced, make it to the human data as quickly as possible.

01:52:15.160 Clotho is a good example. I mean, the hardest part about Clotho before we got involved was

01:52:20.140 no one could manufacture it. No one could make it effectively. Then we went to China, Wuxi,

01:52:25.880 and they figured out a process. And now it's a very robust process and it's going very well.

01:52:29.960 So we hope to be in human clinical trials pretty soon.

01:52:33.440 If you look at it from a global perspective, if everyone could work well together, we have

01:52:37.200 the makings of a biotech revolution, but we continue to have a high benefit to risk ratio

01:52:43.900 too.

01:52:44.360 I mean, so we have to see how that plays out.

01:52:46.680 Well, Michael, this is really exciting.

01:52:48.180 I love when we can sort of string together podcasts over a series of years and get kind

01:52:53.180 of an update from one very exciting idea to another.

01:52:56.280 I was incredibly excited about Obacetrapib when we spoke about it the first time with

01:53:01.300 John and everything you've updated us on today only makes that, look, we just can't wait

01:53:05.940 to get this drug into the hands of people and hopefully not just for cardiovascular

01:53:11.060 disease, but perhaps even more excitingly for Alzheimer's disease, especially in those

01:53:15.460 E4 patients.

01:53:16.800 Anything else you want to talk about on the clotho front?

01:53:18.660 I know we didn't spend too much time on it.

01:53:20.660 You and I are both involved, which is why I'm a little bit coy to talk about it.

01:53:24.100 I don't want to, I don't like to talk about things that I'm directly involved in as well.

01:53:27.640 Right, right. It's moving along well. Like I said, the Chinese, the wuxi made a great

01:53:31.460 protein. What we decided, just to kind of step back where we were a couple years ago, we had this

01:53:37.360 genomic validation that if you have the clotho gene, you're protected against Alzheimer's E4

01:53:43.300 in particular. We had this great animal data consistently showing that if you inject clotho

01:53:49.420 peripherally, you improve cognition. It went to a primate study and showed great data, and the

01:53:54.960 primates got published in Nature, you know, a journal. We believe we had well-validated target

01:54:01.240 and drug, but the missing part was the MOA. No one knows how it works. And by the way, that's a

01:54:05.840 painful point, right? I mean, there's only about 3% of FDA-approved drugs do not have a clear

01:54:11.460 mechanism of action. Yeah, metformin doesn't, isn't it? Right. Tylenol. Yeah, right. We have

01:54:15.940 other examples of where the MOA is not known. So when you and I embarked on, okay, how are we going

01:54:22.120 to get this to humans? What we had to decide to do was we basically, we have to sort of give up

01:54:27.120 on trying to figure out the MOA exactly because we just have to find people willing to fund the

01:54:31.900 studies up until human data and prove that it does improve cognition and a human trial. Then I think

01:54:38.400 it's going to be hopefully a solid going forward, not just hopefully Alzheimer's as a treatment for

01:54:44.460 Alzheimer's disease, hopefully even a prevention too, because it does have all those criteria.

01:54:49.080 We still know the MOA. We do think that the latest is that the full length, yeah, the full

01:54:53.500 length cloth doesn't likely cross the blood-brain barrier, but there is a new fragment that may be

01:54:58.620 cleaved that does cross the blood-brain barrier. That could be the mechanism of actually, we know

01:55:02.500 that peripherally giving it does improve cognition. We're getting closer. Now the supply is in place

01:55:08.720 and we're starting the trials to get ready for the IND filing that, you know, the official

01:55:13.640 we hope to have human data in roughly a year from now. So we can come back and talk about that with

01:55:18.020 maybe Dina together. We can talk about benefits of cloth though.

01:55:22.040 Yeah, it is. It's very high risk, but I know so many of obviously the investors in Jocasta and

01:55:27.600 many of them really come at this through the lens of this is partially a financial investment,

01:55:32.000 but many of them also think of it as philanthropy without the tax break, which is I'm not just

01:55:37.580 putting my money into this company because I want to make money. That would be great.

01:55:40.740 but what would be greater is if this drug actually works and i feel like a lot of people take that

01:55:46.260 approach in biotech especially when it comes to a disease like alzheimer's disease where

01:55:50.560 we're just watching people suffer so much right now right nothing to offer them we have to keep

01:55:56.020 working at it i just lost a double first cousin my mother's brother married my father's sister 0.78

01:56:01.120 so we have all the same grandparents she just died at age 72 of alzheimer's he was an e4 i've

01:56:06.500 seen it and i've seen of course a lot of patients struggle with it and i wish we had better advice

01:56:10.580 about what to do. But we do know, I do think maybe we should end with is, what do you do?

01:56:14.560 What if you're an ApoE4? There are some that say, don't even bother checking it because what can

01:56:18.860 you do about it? I think that's a wrong approach. I mean, knowing you have it, you can obviously

01:56:23.600 better lifestyle, exercise, taking some DHA, even though you may not get enough into the brain is

01:56:29.620 still, and then keeping your obviously no diabetes. Normal blood pressure, not smoking,

01:56:34.460 managing lipids. Yeah, there's a lot you can do. I agree with you. I find it very frustrating when

01:56:38.840 people suggest, and by the way, I totally respect a person who says, I don't want to know. Totally

01:56:43.240 fine. Would never force that on anybody. But to suggest that knowing that couldn't provide the

01:56:48.120 motivation to do the things that are hard to do when you're 30 and 40 years old, long before this

01:56:54.020 disease takes hold, I think that's probably the wrong side of that bet for me. But at the same

01:56:58.360 time, I do see a lot of people going too far. And I would almost put myself in that camp of where I

01:57:03.180 was maybe four or five years ago, trying to look too much at biomarkers that I think in retrospect

01:57:08.980 probably weren't important enough. I think when you're looking so hard for things, you start to

01:57:13.720 make signal out of noise. And I also think that that can create too much anxiety. And quite frankly,

01:57:18.620 I can think of at least two examples where a patient suffered more anxiety than they needed

01:57:24.020 to over something that I just don't know that I would take to the bank as an actionable and

01:57:29.520 treatable and modifiable biomarkers. So I'm pulling for Clotho, I'm pulling for Obocetrapib.

01:57:35.080 And then of course, if they're successful, what that's going to do is open up the door for more

01:57:38.600 things. But what I love about the two of them is potentially they're working by a different

01:57:42.540 mechanism. So I love when you can go after complicated diseases from multiple vantage

01:57:46.940 points. And it's clear we're going to need multiple therapies. That's the thing. Obocetrapib

01:57:50.960 alone could be a preventive strategy. You're going to need more. You need something else as

01:57:55.720 well to make a difference. Well, Michael, thank you for the amazing work you're doing. And thanks

01:58:00.080 for taking the time to come out here today. It was so much fun talking to you. Yeah, so much fun,

01:58:03.100 Peter. Thank you. Thank you for listening to this week's episode of The Drive. Head over to

01:58:08.820 peteratiyamd.com forward slash show notes. If you want to dig deeper into this episode,

01:58:15.640 you can also find me on YouTube, Instagram, and Twitter, all with the handle peteratiyamd.

01:58:20.920 You can also leave us review on Apple podcasts or whatever podcast player you use. This podcast is

01:58:27.880 for general informational purposes only, and does not constitute the practice of medicine,

01:58:32.200 nursing, or other professional healthcare services, including the giving of medical advice.

01:58:37.240 No doctor patient relationship is formed. The use of this information and the materials linked to

01:58:42.880 this podcast is at the user's own risk. The content on this podcast is not intended to

01:58:48.240 be a substitute for professional medical advice, diagnosis, or treatment. Users should not disregard

01:58:53.560 or delay in obtaining medical advice from any medical condition they have, and they should

01:58:57.840 seek the assistance of their healthcare professionals for any such conditions. Finally,

01:59:02.760 I take all conflicts of interest very seriously. For all of my disclosures and the companies I

01:59:08.000 invest in or advise, please visit peteratiamd.com forward slash about where I keep an up-to-date

01:59:15.320 an active list of all disclosures.

The Peter Attia Drive - September 14, 2026

#407 ‒ Preventing cardiovascular and Alzheimer's disease： lowering LDL early, APOE4, and promising new therapies ｜ Michael Davidson, M.D.

Episode Stats

People

Transcript