407 ‒ Preventing cardiovascular and Alzheimer’s disease: lowering LDL early, APOE4, and more
In a Nutshell
Michael Davidson argues that maintaining LDL below 80 mg/dL throughout life—ideally starting in your 30s—prevents heart disease far more effectively than lowering it after plaque forms, since 8 grams of cumulative LDL exposure leads to ASCVD. His company’s CETP inhibitor obicetrapib lowers LDL 45-50%, raises HDL 150%, cuts Lp(a) 50%, and appears to reduce Alzheimer’s biomarkers in APOE4 carriers by improving brain cholesterol metabolism. The key message is that both cardiovascular and Alzheimer’s prevention require treating causal lipid drivers decades earlier than current guidelines recommend.
These notes were generated by AI and may contain inaccuracies.
Peter Attia welcomes Michael Davidson, a cardiologist and lipidologist who runs the lipid clinic at University of Chicago. Davidson sees patients four full days per month in a prevention-focused practice. His path to lipidology was shaped by his father's death at age 47 from a heart attack when Davidson was 16, combined with a family history of abnormal lipids.
Davidson conducted early NASIN trials in the late 1970s and 1980s, researched omega-3 fatty acids during fellowship, and was the first to use fish oil capsules to treat lipid disorders. He participated in all the major statin trials and opened a prevention center after fellowship, along with a research company conducting clinical trials.
Davidson advocates for primordial prevention - stopping plaque from forming before it exists, which is easier than reversing existing plaque. The mainstream approach waits until significant plaque buildup, heart attack, or stroke occurs before treating aggressively. Genomic data shows that maintaining low LDL throughout lifetime prevents heart disease more effectively than lowering LDL after events occur.
"If you lower LDL before you have a heart attack, it's very effective. If you have a heart attack and you lower LDL, it still has benefit, but once you have heart failure, there's no benefit to lowering LDL."
Davidson presents the 8 gram rule: 8 grams of cholesterol in a lifetime leads to heart disease. This can be calculated as:
- 200 mg/dL × 40 years = 8 grams
- 100 mg/dL × 80 years = 8 grams
- 80 mg/dL × 100 years = 8 grams
Data indicates that keeping LDL below 80 throughout lifetime prevents heart disease. The knowledge to prevent heart disease exists; the challenge is applying it earlier in life.
For patients reluctant to start early intervention, Davidson offers additional information including:
- Genetic testing
- Polygenic risk scores
- Coronary calcium scanning (when age-appropriate)
- Advanced plaque analysis
- Additional risk factors like Lp(a) and CRP
- Family history assessment
Davidson shares his personal comparison with his brother. Both had equally bad lipids after their father's death. Davidson began taking niacin in medical school and started statins when they became available. His brother maintained a strict vegetarian diet and delayed statin therapy until age 44, when he required bypass surgery with significant coronary disease.
This decade from 30-40 years old represents a critical window when plaque forms relatively rapidly, making early intervention most effective.
Attia references Alan Snderman's paper from approximately 12 years ago that reframed the argument from treating 5-year, 10-year, or 15-year risk to treating causal drivers of disease. LDL is causally related to ASCVD based on Mendelian randomization studies and clinical trials.
The smoking cessation analogy illustrates this: doctors don't recommend continuing smoking for another 10-15 years before quitting, despite low short-term risk. The same principle applies to treating causal risk factors regardless of time horizon.
Davidson identifies several reasons for resistance to early LDL treatment:
- Belief that LDL can be managed entirely through lifestyle
- Genetic factors often underlie elevated LDL, as revealed through genetic testing in lipid clinics
- General reluctance to take medications
- Primary prevention guidelines from organizations like the American Heart Association that don't emphasize early intervention
Lifestyle interventions have greater impact on hypertension than on dyslipidemia, yet face less resistance, suggesting other factors beyond lifestyle concerns drive the controversy.
Davidson discloses his role as CEO of New Amsterdam Pharmaceuticals, manufacturer of a CETP inhibitor drug. The first CETP inhibitor trial (torcetrapib by Pfizer) combined with atorvastatin ended around December 2006.
Cholesteryl ester transfer protein (CETP) transfers cholesterol from HDL into LDL. Blocking CETP raises HDL and lowers LDL while improving HDL clearance from the liver. Animals lacking CETP (rats, dogs) have very low LDL, very high HDL, and don't develop atherosclerosis unless given CETP transgenically.
Monkeys, rabbits, and humans have CETP and relatively high LDL with low HDL. CETP represents a conservation mechanism - when dietary cholesterol is scarce, CETP helps conserve cholesterol in the body.
Twenty years ago, HDL raising was considered the "holy grail" based on Framingham Heart Study data showing protection from heart disease with high HDL. The total cholesterol to HDL ratio was a famous risk predictor. The Helsinki Heart Study and VA-HIT trial using fibrates showed modest heart disease benefit with HDL raising.
Niacin alone in the Coronary Drug Project lowered events in secondary prevention (pre-statin era), though less significantly than statins. Niacin raises HDL 20-30% but has significant side effects including flushing, GI issues, and liver problems.
Two major trials showed niacin added to statins did not lower events:
- AIM-HIGH trial
- HPS2-THRIVE trial
HPS2-THRIVE analysis suggested LDL lowering (niacin also lowers LDL) might have contributed to event reduction.
Davidson filed the first patent on CETP inhibition in 1993. Torcetrapib raised HDL 50-75%, which based on epidemiological data suggested substantial risk reduction. However, the drug was stopped due to increased mortality (both cardiovascular and non-cardiovascular).
The blood pressure increase (3-5 points) was only the visible effect. Torcetrapib had prominent off-target effects on adrenal glands, markedly increasing aldosterone and steroid production. Animal studies confirmed this was off-target since rats lack CETP but still showed dramatic blood pressure increases within an hour of torcetrapib administration.
Four CETP inhibitors followed torcetrapib. The off-target aldosterone effect was isolated to a specific component in the torcetrapib molecule, allowing development of cleaner compounds. Importantly, early CETP inhibitor development focused entirely on HDL raising rather than LDL lowering.
Once the spherical particle matures, it goes to the liver and gets cleared. The predominant route for bile production comes from HDL, which is funneled into the biliary system and out into the intestine. HDL functions as a "dump truck," picking up inflammatory components, cytokines, and other garbage along the way, taking it back to the liver for clearance. While people initially thought HDL was picking up cholesterol from plaque in reverse cholesterol transport, the amount of cholesterol removed from plaque is actually minuscule compared to its effects on other organs.
HDL also plays important roles in immune function and battling innate pathogens. This understanding was the basis for why high HDL was considered beneficial in 2006, with the mechanism being removal of cholesterol from artery walls and transport back to the liver.
Torcetrapib
Torcetrapib raised HDL by 75% and lowered LDL by 10-15%. Despite the significant HDL increase, off-target effects proved devastating, including blood pressure elevation. The drug failed despite the magnitude of HDL benefit being thought to outweigh the 3-point blood pressure increase.
Dalcetrapib
This was a weak CETP inhibitor, sometimes called a CETP modulator. It raised HDL by 30-40% with no LDL effect and was very safe. It reduced diabetes risk but had no MACE (major adverse cardiac events) benefit—not even a trend.
Evacetrapib (ACCELERATE Trial)
This was a good molecule, very safe, studied in acute coronary syndrome (secondary prevention). It lowered LDL 15-20% and raised HDL 75%. The trial was stopped at 2 years for futility with no MACE benefit. However, the IMPROVE-IT trial with ezetimibe showed that 2 years is too short—lines separated after 2 years, suggesting evacetrapib might have shown benefit if continued. The drug had significant all-cause mortality benefit (largely cardiovascular) but not MACE benefit.
Anacetrapib (REVEAL Trial)
This trial saved the CETP inhibitor class. It was the largest outcome study with 30,000 patients over 4 years. Baseline LDL was 60 mg/dL, with 17% LDL lowering (11 mg/dL absolute). This produced 9% relative risk reduction, proving LDL lowering with CETP inhibition translates to cardiovascular benefit. The drug wasn't commercialized due to lipophilic properties causing deposition in fat tissue that wouldn't wash out for years, creating environmental concerns.
When primary prevention trials, secondary prevention trials, Mendelian randomization, and epidemiology are plotted with LDL-C on x-axis and event rate on y-axis, all lines go down. For every 1 mmol/L (38.8 mg/dL) LDL reduction, there's 22% relative risk reduction. This relationship is largely linear and independent of how LDL is lowered—whether by genetics, statins, or other drugs.
After selling Corvidia (IL-6 antibody for heart disease, ZEUS trial), the team acquired obicetrapib from Amgen, which had shelved it after Repatha launch struggles. New Amsterdam Pharma was formed approximately 5-6 years ago. Amgen had improved manufacturing cost-effectiveness but conducted no clinical trials.
Initial strategy focused on statin intolerance, but FDA does not accept this as an indication. The path shifted to full LDL-lowering development requiring outcome study at launch.
TULIP Study
Showed 45% LDL lowering with obicetrapib. The drug is much more potent than previous CETP inhibitors—dosed at 10 mg (or even 5 mg) versus 100+ mg for others. It achieves 40-50% LDL reduction and 150% HDL raising.
HDL vs LDL Mechanism Clarification
Mendelian randomization studies confirmed that CETP LDL lowering translates to cardiovascular benefit, while HDL raising alone does not explain protection. No pure HDL-raising gene was associated with protection—genes that raised HDL without LDL or triglyceride effects showed no benefit. Alcohol raises HDL but increases mortality. SR-B1 gene raises HDL but increases cardiovascular risk.
Phase 3 Trials Completed
- ROSE and ROSE 2: Showed 45-50% LDL lowering as monotherapy, 55%+ in combination with ezetimibe
- BROOKLYN: Heterozygous familial hypercholesterolemia trial
Ongoing Studies
- PREVAIL: Outcome study
- ALCHEMIST: Alzheimer's study started simultaneously with phase 3 trials
The company has filed in Europe and is timing US filing with outcome study results. Cash position allows completion of PREVAIL trial.
The phase 3 trials tested obicetrapib in patients with familial hypercholesterolemia who were already on maximal therapy. Nearly all patients were on high-intensity statins, and many were also on ezetimibe or PCSK9 inhibitors. Despite this, their average LDL remained around 100 mg/dL. Adding obicetrapib produced an additional 40% LDL reduction, bringing levels from approximately 100 mg/dL down to 60 mg/dL.
The Broadway trial enrolled patients with atherosclerotic cardiovascular disease, again on maximal statins with approximately 70% on 40-80 mg atorvastatin. Obicetrapib achieved roughly 35% LDL lowering at 12 weeks, with the benefit sustained through 52 weeks. This trial also demonstrated a 21% reduction in major adverse cardiac events within one year, though not statistically significant given the 200-patient sample size. A landmark analysis showed separation of event curves after the first six months.
The Tandem trial combined obicetrapib with ezetimibe and achieved approximately 50% LDL lowering. The combination pill is currently in development.
PREVAIL is an ongoing cardiovascular outcomes trial enrolling 9,500 patients with established atherosclerotic cardiovascular disease on maximally tolerated statins. The study requires a minimum follow-up of 2.5 years and is event-driven for four-point MACE (heart attack, stroke, cardiovascular death, and coronary revascularization). The trial includes over 400 sites globally across Europe, the United States, China, Japan, and other regions. The 2.5-year minimum follow-up ends at the end of the current year.
Europe does not require cardiovascular outcomes data for approval, as regulators accept LDL lowering as sufficient evidence of benefit. The United States requires outcomes data. Europe will therefore likely approve first, with a one-year window for payers to incorporate the outcomes data into reimbursement decisions. The company plans to align US launch with availability of the outcomes data to facilitate reimbursement at launch.
The Rose 2 trial showed discordance between LDL-P lowering measured by NMR and ApoB reduction. The NMR LDL-P assay used was the LipoScience assay (now owned by LabCorp), which is considered the most reliable NMR platform. LDL-P showed more substantial reduction than ApoB, with the latter demonstrating approximately 16% lowering.
Discordance between ApoB and LDL-P is driven primarily by insulin resistance and the presence of small LDL particles. When LDL-P is high but ApoB is not correspondingly elevated, small particles predominate. ApoB is a mass assay, whereas NMR measures particle number. The mass of ApoB per particle varies slightly with particle size due to glycosylation differences, which accounts for the observed discordance.
CETP inhibition dramatically reduces small LDL particles (by approximately 90%), which contributes to greater LDL-P reduction relative to ApoB. The UK Biobank is being analyzed to determine where cardiovascular risk follows in discordant cases.
Non-HDL cholesterol (LDL plus VLDL cholesterol) correlates closely with ApoB. The MERC and REVEAL trials showed cardiovascular benefit was more strongly associated with non-HDL cholesterol than ApoB. However, a JAMA or New England Journal of Medicine paper published approximately six years ago found ApoB performed better than non-HDL cholesterol, though other studies have shown the opposite result. Non-HDL cholesterol represents a low-cost alternative to ApoB or LDL-P measurement, though the latter two may offer marginally greater precision.
Obicetrapib's ApoB lowering in the PREVAIL trial is expected to support the desired outcomes benefit, with even stronger performance when non-HDL cholesterol is used as the biomarker.
Obicetrapib reduces Lp(a) by approximately 50%, which exceeds the 15-20% Lp(a) reduction seen with PCSK9 inhibitors. The mechanism appears to involve reduced Lp(a) production, confirmed in a clearance study using labeled Lp(a) that demonstrated decreased synthesis. Other CETP inhibitors have shown Lp(a) lowering, though to a lesser degree consistent with their lower potency.
The Horizon trial is a phase 3 study of an antisense oligonucleotide targeting Lp(a) that is expected to read out this year. The trial was extended due to lower-than-expected event rates. Patients in the trial achieved LDL-C of 60 mg/dL, which likely contributed to the reduced event rate. This supports the approach of aggressively lowering LDL-C in patients with elevated Lp(a), even though statins raise Lp(a) slightly. The trial will provide causal evidence for Lp(a) as a therapeutic target.
Statins effectively lower LDL-C, reduce cardiovascular events, are well tolerated, generic, and easy to prescribe. However, they modestly increase diabetes risk, raise Lp(a), and primarily lower large rather than small LDL particles.
Obicetrapib lowers diabetes risk, reduces Lp(a) by approximately 50%, and lowers small LDL particles by 90%. These complementary effects make it an attractive add-on to statin therapy. The combination with ezetimibe achieves LDL reductions comparable to PCSK9 inhibitors or upcoming oral PCSK9 inhibitors.
The company estimates that obicetrapib alone or combined with ezetimibe will address 90% of patients' lipid treatment needs, providing primary care physicians with an oral option to add to statin therapy.
High-dose statins (80 mg atorvastatin, 40 mg rosuvastatin) produce minimal additional LDL-C reduction beyond moderate doses, as demonstrated by dose-response curves. The number needed to harm increases substantially at higher doses relative to incremental benefit. Many patients prefer to avoid high-dose statins.
Transaminase elevations (AST/ALT) occur more frequently in clinical practice than reported in literature when statins are combined with ezetimibe. Current guidance permits continuation if elevations remain below 3x the upper limit of normal, but this results in patients tolerating levels around 80 IU/L without clear evidence that this is harmless.
Drug pricing in the United States is driven by the need to generate rebates for pharmacy benefit managers to secure formulary placement. Pricing too low eliminates rebate margins and reduces the likelihood of formulary inclusion. Direct-to-consumer models such as Lilly Direct represent an emerging alternative that may improve patient access at reasonable out-of-pocket costs.
Repatha was initially priced at $16,000 per year, which severely limited access. Nexl (bempedoic acid) at $6,000 per year is considered overpriced relative to its modest efficacy and is largely reserved for statin-intolerant patients who decline injectable therapy.
Mendelian randomization evidence from the Bronx aging study of approximately 400 Ashkenazi Jews over age 100 identified CETP loss-of-function variants as protective against Alzheimer's disease. Individuals with both APOE4 and CETP loss-of-function had their elevated Alzheimer's risk essentially eliminated, reverting to the risk level of non-APOE4 carriers. These individuals also demonstrated favorable lipid profiles with high HDL-C, low small LDL particles, and high large LDL particles.
This genomic finding has been replicated in subsequent studies, including work from McGill University. Mouse models lacking CETP were engineered to express the CETP gene, which worsened amyloid pathology and cognitive performance in animals also expressing amyloid precursor protein. Administration of a CETP inhibitor reversed the cognitive impairment, as assessed by nesting behavior.
Additional supporting evidence includes observational data linking higher HDL-C, particularly ApoA1, with reduced Alzheimer's risk, though findings have been mixed.
The company allocated $1 million from its $200 million Series A round specifically for Alzheimer's research, despite investor concerns about funding Alzheimer's programs given the high failure rate in this therapeutic area. The company went public in late 2022 after founding in early 2021.
A small study was conducted in Amsterdam with approximately 13 APOE4 patients who had mild cognitive impairment. The study used obeticholic acid and observed improvements in lipid metabolism.
APOE4 is fundamentally a lipid gene. Patients with APOE4 have high LDL and lower HDL levels. This makes obeticholic acid a potentially valuable add-on to statin therapy for the 25% of the population carrying this genotype.
The brain is a cholesterol-rich organ, with approximately 2% of its dry weight representing 50% of total body cholesterol. During childhood growth, neurons actively synthesize cholesterol. After adulthood, neurons stop making cholesterol, while astrocytes continue production to support neuronal repair and maintenance. Myelin is approximately 50% cholesterol.
The blood-brain barrier completely separates brain lipid metabolism from peripheral metabolism. There is no APOB in the brain - only APOE and HDL particles. There is no LDL in the brain, and no connection exists between plasma LDL and brain LDL. Liver transplantation demonstrates this independence: patients receiving an APOE3 liver retain APOE4 in the brain, with no cross-connection between liver and brain APOE genotypes.
- APOE33: average Alzheimer's risk (most common)
- APOE34: approximately 3x increased risk (20-25% of population)
- APOE44 homozygotes: approximately 10x increased risk (2-3% of population)
- APOE2: lower risk and associated with longevity
APOE4, produced by astrocytes, shows impaired cholesterol efflux and lipidation. This leads to inadequate neuronal nourishment and toxic cholesterol accumulation. The toxic cholesterol cannot be properly cleared, creating inflammatory conditions that promote amyloid and tau pathology. APOE4 patients develop Alzheimer's approximately 10 years earlier with one E4 allele, and 20 years earlier with two E4 alleles.
HDL is the primary lipoprotein that interfaces with the brain. Most brain APOA1 appears to originate from plasma HDL rather than local brain synthesis. HDL crosses the blood-brain barrier via the SRB1 receptor, which is heavily expressed in the choroid plexus. HDL delivers antioxidants to the brain, reduces inflammatory responses, and may facilitate amyloid removal. High HDL levels can potentially help remove excess cholesterol that APOE4 patients cannot metabolize effectively.
Statins do not appear to increase Alzheimer's risk based on observational data. Large statin trials examining Alzheimer's showed no benefit rather than harm. Secondary analyses suggest potential dementia risk reduction in APOE4 patients taking statins, though these findings require careful interpretation. The vascular dementia benefits of statins may be captured in overall dementia analyses.
In the pilot study of MCI patients, obeticholic acid reduced 24-hydroxy and 27-hydroxy cholesterol levels in cerebrospinal fluid - an effect not observed with statins. CSF antioxidant levels increased, and cognition remained stable with some anecdotal cognitive improvements. The Broadway study involving 2,500 heart disease patients (average age 65) demonstrated that obeticholic acid reduced p-tau217 elevation in the overall population, with greater effects in older patients and APOE4 carriers.
- p-tau217 and p-tau181: both correlate strongly with brain amyloid on PET imaging
- p-tau217 appears earlier in disease progression than p-tau181
- GFAP (glial fibrillary acidic protein): early disease biomarker
- Neurofilament light: later-stage inflammatory biomarker
- The p-tau217/amyloid-beta 42:40 ratio received FDA approval in the past year
In APOE4 homozygotes, obeticholic acid showed over 20% difference from placebo in p-tau217 reduction. All biomarkers improved including p-tau181, amyloid-beta 42:40, GFAP, and neurofilament light. The study demonstrated target engagement with potentially preventive effects against Alzheimer's disease.
p-tau217 cannot currently serve as a regulatory endpoint for Alzheimer's prevention approval, which requires demonstration of cognitive improvement. An additional study is planned to examine obeticholic acid's effects on Alzheimer's prevention more carefully. The APOE4 patient community is advocating for treatment access rather than waiting extended periods for definitive trial results.
People were using PCSK9 inhibitors off-label for LP(a) reduction the moment it became clear that it was lowering LP(a) 20 to 30%. There are plenty of people using GLP-1 agonists today for neuroprotection even though that would still be considered off-label.
PTAL 217 with enough advocacy can become a regulatory endpoint. The flip side argument is that you are proving a biomarker but not treating the underlying disease. The actual tau in the brain leaks out into the plasma, so it is a measurable biomarker.
Anti-amyloid antibodies have modest benefit but not for APOE4 homozygotes. APOE4 homozygotes are the ones that 100% go on to get Alzheimer's if they live long enough. Rare diseases allow surrogates to get approval based on a surrogate that may or may not translate to clinical benefit.
HDL fooled everybody for decades. People were chasing the wrong thing and had all the epidemiology to back it up, but at the end of the day it didn't matter. If sitting here in 20 years talking about how the whole PTAL 217 thing took everyone down a rabbit hole, what would be the revisionist explanation?
In the last few months, supportive data came out. The GLP-1 trials (EVOKE and EVOKE Plus) showed no improvement in cognition and no effect on PTAL 217. On the other hand, the TRAILBLAZER anti-amyloid trials from Lilly showed a pretty high correlation between improvement in cognition and PTAL 217 going down.
A study with 300 patients that have pre-Alzheimer's evidence of higher tau levels based on their APOE status is planned. The goal is to get patients before MCI because MCI is like early heart failure where nerve degeneration has already occurred, the brain is already shrunk, inflammation is set in the cascades, and tauopathy is already well underway.
The Broadway data showed the most feedback was that the state is just too good to be true. It was only 30 patients but had homozygote E4 and it was so statistically significant it was accepted for publication. The Alzheimer's world is very fixated on amyloid.
The Alzheimer's world is very fixated on amyloid because it's such a devastating disease and it's all about hope. Anti-amyloid drugs have to be taken with worry about brain bleeding and they don't work on homozygote E4 either because of ARIA. ARIA is a contraindication to use in E4 homozygotes.
Anti-amyloid drugs could be beneficial but they have to be started 20 years sooner. The question is whether you can really convince somebody who's 40 years old to take an infusion to prevent amyloid, especially if the side effects are potentially catastrophic.
The target population could be age 60. From Broadway data, the rapid progressors are identified by who has tau go up the most, and E4 is older, even metabolic, even diabetes that makes the amyloid. The next study will hopefully set the stage for a phase three trial looking at conversion of normal to MCI by preventing that conversion.
The cardiovascular part involves omega-3s for heart disease. The STRENGTH trial was developed with 4 grams of DHA and EPA, mostly EPA with some DHA. It didn't work overall but there was a hint of benefit in the secondary prevention arm.
There are two reasons omega-3s can reduce risk: triglyceride lowering and anti-thrombotic effects. EPA antagonizes the arachidonic acid pathway and has anti-thrombotic and anti-inflammatory effects. STRENGTH was set up as a triglyceride lowering trial.
The REDUCE-IT trial showed benefit with EPA only. The mineral oil placebo may have increased inflammatory signaling and affected absorption of statins because LDL levels went up about 10%. Even taking all that away, EPA still worked, though it may not have been the 25% that people think it was.
The placebo in STRENGTH was corn oil, which is inert. The hypothesis was that mineral oil itself probably increased events slightly, which made the EPA look better alone than it might have been. The PROMINENT study with pemafibrate showed no benefit in a similar population to STRENGTH.
EPA works on reducing events but not through triglyceride lowering - it's more for the anti-thrombotic, anti-inflammatory effect. DHA has health value especially for the brain because the brain is predominantly DHA. DHA supplementation trials on cognition didn't really show much benefit.
APOE4 people don't get as much DHA to the brain - they have a true DHA deficiency in the brain. The MFSD2A transporter is a transporter across the blood-brain barrier for DHA. DHA prefers the lysoPC form of DHA, which is the preferred substrate for DHA to cross the blood-brain barrier.
LysoPC DHA is coming and studies will be done showing it getting across the blood-brain barrier completely independent of the CLIA work. DHA has to come from the periphery across the blood - the brain cannot make that much DHA from other fatty acids.
If an individual is not consuming much fatty fish and measures their red blood cell membrane concentration of EPA and DHA at 4%, supplementing with high quality over-the-counter EPA and DHA to get to 10% would probably lower cardiovascular disease risk but probably not have an impact on Alzheimer's disease risk given the inability to get the DHA in the brain.
You have to make the lysoPC form which you can do with either a fish-based oil or a krill-based oil. There is no version of lysoPC DHA on the market today but one is already available krill oil based and there's one with fish oil based coming. Studies will be done on the science behind that as a dietary supplement.
AI could speed up the time it takes to do a clinical trial by a log and reduce the cost by a log. Taking drug development from $4 billion to $40 million or even $400 million, and from 10 years to one year would change the course of human history.
Regulatory reform is needed before AI can meaningfully impact clinical trials. Everyone has tried to make clinical trials more streamlined with less data collection. Clinical trials could be looked at based on using AI to assess who took the drug the right way and how well it worked, instead of being diluted by everyone who stopped the drug or wasn't compliant.
Making efficacy based on actual compliance rather than intention to treat would make a huge difference. In Broadway or ROSE 2 trials, the question is what percentage of patients enrolled were nearly 100% compliant. The benchmark is 80% compliance. When doing per protocol analysis, anyone below 80% compliant gets taken out, but the FDA will never accept per protocol analysis.
LDL trials with different mechanisms of action are getting harder and harder to do because of the ubiquity of drugs on the market already and the amount of LDL lowering that's already in the baseline. You can't ethically withhold treatment to patients in a trial.
All LDL lowering drugs including PCSK9 inhibitors have a little bit of a diabetes signal except for obicetrapib. It's not just patients going on to get type 2 diabetes - patients who don't progress all the way show an increase in insulin resistance. The signal is dose related, age related, and weight related.
Bile acid sequestrants which lower LDL show an improvement in glycemic control. When you have bile acid removal, you have to make a new one, which creates energy requirements. Blocking bile acid from being reabsorbed can ultimately reduce blood sugar.
Some patients who can't tolerate any other drug for hypercholesterolemia are still on bile acid sequestrants. They provide about 10 to 15% LDL lowering. With ezetimibe now available, bile acid sequestrants aren't used as much because ezetimibe is a lot easier to take.
Ezetimibe doesn't appear to have an increase in insulin resistance associated with it clinically, though genomically there is a slight signal. PCSK9 inhibitors genetically have that diabetes signal as well.
Recent data suggests changes in the type of bile acids could be a mechanism, along with beta cell preservation issues. The glucose issue with statins hasn't been fully figured out mechanistically.
The asymmetry is enormous - the risk reduction from lipid lowering on cardiac events is far bigger magnitude than the increase in risk of type 2 diabetes. The net benefit is still very much in favor of taking the statin despite the diabetes discussion that comes up frequently with patients.
There are no other CETP inhibitors in the pipeline. If successful, there will be a lot more, but it's very hard to beat obicetrapib which has low dose, is extremely well tolerated, and has favorable pharmacokinetics. The molecule came from Amgen through Mitsubishi, a Japanese company that first discovered it then sold it to Amgen.
Mitsubishi wanted to make a better CETP inhibitor - one with no fat uptake, much lower dose, more potency, more bioavailable. The approximate dollar amount to develop this drug from acquisition through clinical trials to approval is well over a billion dollars, not counting what was spent before acquisition to get it into phase two.
The average drug right now costs about three to four billion dollars to approval. This accounts for the failures too, but it might be even more if you count for the failures. It's a lot of money. It's a high-risk, high-reward investment.
In the current investment environment where AI is the most interesting and exciting thing to invest in, with capital flying into AI, biotech companies face significant difficulty raising money. The question is where we are in the biotech life cycle and why biotech is forever going to be complicated because of the time horizon.
Big pharma companies have big sales forces and companies, but they have pipelines with big gaps and patent cliffs coming for all of them. They can't grow - in order to grow they have to get new product. In the way the system's set up, everything ultimately goes generic. If companies even want to grow 3 or 4% per year, which is pretty minuscule growth, they have to add a lot of new products to their pipeline.
Big companies have become pretty dysfunctional on innovation about developing new drugs. Small companies can be a lot more nimble, a lot more flexible, a lot more risk-taking, and don't have to worry about corporate hierarchy to make decisions. Biotech is right now in a really good position.
AI is going to help a lot on coming up with the right targets and making clinical trials more cost effective and lower cost. Even in drug discovery, China has become a huge investment - for every chemist in the United States there's probably 10,000 in China. They have the capabilities to really surpass anything that can be done from a drug development perspective just by the person power they have.
This sets up an opportunity. There is still the science and the innovation and the creativity that could lead the world in coming up with new ideas, but China is going to be able to take it to the next level with how to get a drug produced and made to human data as quickly as possible.
Cloth is a good example. The hardest part about cloth before getting involved was that no one could manufacture it effectively. After going to China, specifically Wooi, they figured out a process and now it's a very robust process and it's going very well. Human clinical trials are hoped to begin pretty soon.
From a global perspective, if everyone could work well together, there are the makings of a biotech revolution, but there has to continue to be a high benefit to risk ratio.
There's excitement about stringing together podcasts over a series of years to get updates from one exciting idea to another. There was incredible excitement about obeticholic acid when first discussed with John, and everything updated makes the drug look very promising for getting into the hands of people - hopefully not just for cardiovascular disease but perhaps even more excitingly for Alzheimer's disease, especially in E4 patients.
The Chinese at Wooi made a great protein. The mechanism of action (MOA) part is still unclear. The genomic validation showed that if you have the cloth gene, you're protected against Alzheimer's, particularly E4. There was great animal data consistently showing that if you inject cloth peripherally, you improve cognition. It went to a primate study and showed great data in the primates, published in Nature journal.
There was a well-validated target and drug, but the missing part was the MOA - no one knows how it works.
Only about 3% of FDA-approved drugs do not have a clear mechanism of action. Examples include metformin and Tylenol. When embarking on getting this to humans, the decision was made to give up on trying to figure out the MOA because it was necessary to find people willing to fund the studies up until human data and prove that it does improve cognition in a human trial.
The latest research indicates that the full-length cloth doesn't likely cross the blood-brain barrier, but there's a new fragment that may be cleaved that does cross the blood-brain barrier and that could be the mechanism of action. Peripherally giving it does improve cognition.
Getting closer to having the supply in place and starting the trials to get ready for the IND filing. The goal is to have human data in roughly a year from now.
It's very high risk, but many investors in Jocasta come at this through the lens that this is partially a financial investment, but many also think of it as philanthropy without the tax break. The motivation isn't just putting money into this company to make money - what would be greater is if this drug actually works.
A lot of people take that approach in biotech, especially when it comes to a disease like Alzheimer's where people are suffering so much right now with nothing to offer them.
A double first cousin was just lost at age 72 to Alzheimer's. She was an E4. Having seen patients struggle with it, there's a wish for better advice about what to do.
The question of what to do if you're an ApoE4 carrier - some say don't even bother checking it because what can you do about it. That's considered the wrong approach. Knowing you have it, you can obviously do better lifestyle, exercise, taking some DHA even though you may not get enough into the brain, keeping no diabetes, normal blood pressure, not smoking. There's a lot you can do.
It's frustrating when people suggest that knowing your status couldn't provide the motivation to do the things that are hard to do when you're 30 and 40 years old, long before this disease takes hold. There's respect for a person who says "I don't want to know" - that's totally fine and would never force that on anybody.
There's a tendency to go too far, looking too much at biomarkers that in retrospect probably weren't important enough. When looking so hard for things, you start to make signal out of noise. This can create too much anxiety. There are examples where patients suffered more anxiety than they needed to over something that wouldn't be taken to the bank as an actionable and treatable and modifiable biomarker.
Pulling for cloth and obeticholic acid, and if they're successful, what that's going to do is open up the door for more things. What is loved about the two of them is that they're potentially working by different mechanisms. It's good to go after complicated diseases from multiple vantage points. It's clear that multiple therapies are going to be needed.
Obeticholic acid alone could be a preventive strategy, but more will be needed - something else as well to make a difference.
Keep Peter Attia MD in your library
Save the videos and channels worth coming back to, and find them again in one place.
