Alex Harding

An internist in biotech

[Alex Harding @AlexHarding7]

@Biohazard3737 Same goes for biotech CEOs

[Alex Harding @AlexHarding7]

Hoping for October baseball for the @Orioles next year!

[Alex Harding @AlexHarding7]

@AppleHelix @ArvinasInc That’s an entirely separate question :)

[Alex Harding @AlexHarding7]

The @ArvinasInc LRRK2 CSF PD is damn impressive—a degrader crossing the BBB.

Makes me wonder how many good molecules have been cast aside because they didn’t conform to the Rule of 5…

[Margaret Manto @MargaretManto]

Matt Gline, Roivant CEO, on Anthropic's foray into biotech:

[Alex Harding @AlexHarding7]

@rtnarch 100% true, but injecting a “cocktail of chemicals” with unknown MoA in a Ph1 would be challenging for regulatory reasons. And, even if you convinced the HA to run the Ph1, you would still need a factorial design trial showing the contribution of each component.

[Neil Floch MD @NeilFlochMD]

If someone miraculously invented a drug that:
- successfully reversed obesity and most of its associated diseases
- improved longevity
- decreased the long term health costs

….and it was NOT covered by health insurance, ….

- then what would be the use of health insurance?

[Dave A Ricks @DaveARicks]

Adam the contrast was the point. Two different bad/fatal diseases yet very different response from the field to advances. We do not have a functional cure for many cancers but HR < .7 in any cancer is described as practice changing in oncology. Neurology seems underwhelmed by the same degree in disease change.

Of course cancer is not one thing and I could have been more precise- CLL is not good but now mostly not fatal, due to many advances from Lilly and our peers HR+ HER - breast cancer has long survival now yet GI cancers are mostly unsolved and survival curves are terrible.

You know what else is terrible? Almost every neuro-degenerative disease including AD and we (including the media) need to discuss it that way. Unfortunately I’ve witnessed end of life journeys up close in both diseases, while different horrors present themselves in each I’d say they are both pretty bad. Advancement 30%+ at a time should change practice and routinely be offered to patients and families. Oncology does this Neurology doesn’t.

[Alex Harding @AlexHarding7]

@Jasonmmast @adamfeuerstein $CRSP?

https://t.co/GWLxPDYvZx

[Professor Oak @Prof_Oak_]

Continue to be amazed &amp; confused by how long doomed biotechs manage to persist in translational purgatory, if for no other reason than the outsized drag of opportunity cost in an industry with decade-long cycle times

Anyway, goodbye $CRBU

[Sam Altman @sama]

for a long time i said that antisemitism, particularly on the american left, was not as bad as people claimed.

i'd like to just state that i was totally wrong.

i still don't understand it, really. or know what to do about it.

but it is so fucked.

[Alex Harding @AlexHarding7]

@AppleHelix I’m not anti-creatine, but that’s probably mostly water weight, not real muscle gain.

[Iron Direct Primary Care @ironDPC]

My questioning of lipids began in PA school in 2016, when a physician gave our class a lecture on evidence-based medicine. He showed us mortality data on statins. When you look at mortality alone, the number needed to treat can look underwhelming. He also explained relative risk reduction in a way that made it feel, to me at the time, like a Big Pharma manipulation of data.

That lecture planted a seed. I adopted the paradigm that LDL elevation was overblown. More importantly, I convinced myself that I was the exception: lean, fit, insulin-sensitive, weight training, eating “clean,” and therefore somehow protected from the overwhelming evidence linking ApoB-containing particles to cardiovascular disease.

After graduation, I kept studying lipids. In 2019, I became fascinated with arguments built around Eskimo, ancestral, and prehistoric populations. Looking back, I can see how selectively I interpreted that material. I focused on evolutionary narratives and ignored the obvious limitation: many of those populations died young, before modern atherosclerotic disease had decades to develop.

Around the same time, I discovered the Quest Diagnostics Cardio IQ panel. It was still early in its life cycle, and I was excited by advanced lipid testing. I became the primary of a patient who found me at the mom and pop urgent care and somehow knew that I was open to alternative thinking and who wanted to use the same test because, like me, he was lean, fit, and a weightlifter. Neither of us wanted to believe that someone who looked metabolically healthy could still be at risk for myocardial infarction, stroke, or vascular disease.

Then came 2020.

During the pandemic, my questioning of medical authority intensified. I studied the disease as it moved through Asia, then Europe, and by the time it reached the United States, I felt prepared. I worked in urgent care and treated thousands of patients. I wrote to local mayors and sheriffs, pleading with them not to lock us down. I became part of the anti-lockdown resistance, protesting on the streets and am proud of that. That period changed how many of us viewed institutions, public health messaging, and medical authority.

But that skepticism came with a cost.

In 2021, I published a YouTube lecture on the Quest Cardio IQ test. At the time, very few clinicians were talking publicly about these advanced lipid panels. The lecture received thousands of views. In 2022, I found the Labcorp NMR panel and made another lecture. That one also reached thousands of people. I was not just studying this material anymore. I was teaching it.

From 2021 through 2024, I listened to nearly every countercultural medical authority I could find. It started with keto physicians like Gary Fettke and low-carb advocates on social media. Then came the carnivore movement, organ-meat influencers, alternative medicine conferences, and the online war between lipid denialists and lipid-lowering advocates. I even met Shawn Baker at an alternative medicine conference in 2023. The movement reached peak absurdity, in my mind, with the rise and fall of Paul Saladino and Liver King. I learned that they both sourced their supplement line from a manufacturer from a businessman in the supplement industry. I had a decision to make. Do I go full in on this and make millions selling supplements and become such an absurd online personality. I recognized for the first time the meaning of grift. The carnivore influencers had capitalized on a growing health conscious and skeptical market with a “natural” supplement. It was fancied as science perhaps with loose inspiration derived from the anthropological studies of a dentist Weston A Price who found less tooth decay in ancestral populations that at a traditional meat based diet.

I started Iron Direct Primary Care in 2021 as an alternative healthcare home for patients who wanted something outside the conventional system. For years, I cared for many patients who did not believe elevated lipids mattered. We ordered everything: MPO, homocysteine, hs-CRP, omega-3 testing, IL-6, Lp-PLA2, and thousands of NMR panels. Desperately searching for an answer to cardiovascular disease other than LDL and APO B burden.

The pattern was always the same.

Patients had elevated LDL particle number and elevated ApoB, and we often kicked the can down the road. We focused instead on insulin sensitivity, clean diet, seed-oil avoidance, exercise, nitric oxide supplements, methylated vitamins, inflammation markers, and metabolic optimization.

On Instagram, X and podcasts, the lipid battle raged. I listened to lipid skeptics like MD Tro and ex-carnivore MD Paul Saladino, some were compelling. Some were charismatic. Some wore white coats and spoke with absolute confidence. I thought some of them were brilliant. I even referred patients into that ecosystem.

Then reality started to break through.

In 2024 and 2025, I began seeing heart attacks and strokes in my own patient population. That floored me. These were not the stereotypical patients people imagine when they think of cardiovascular risk. These were patients who avoided seed oils, lifted weights, ate clean, took nitric oxide boosters, optimized methylation, and followed the advice of wellness influencers.

Yet they still had events.

There are carnivore FB groups that Carnivorecringe on instagram screenshots and posts revealing a toxic subculture. These self-help groups will gaslight patients who suffer from keto and carnivore with absurd claims like “they weren’t eating enough fats” or “they didn’t give carnivore long enough.”

One patient had a stroke and left my practice to pursue high-priced “natural cardiology” advice form Wolfson for $7,000. Another patient pursued an alternative cardiology path by surgeon Philip Ovadia for also $7,000 and later required triple bypass surgery. These cases forced me to confront something uncomfortable: if I continued to minimize ApoB and LDL particle burden, I might be helping patients feel reassured while their arteries continued accumulating risk.

So I re-read the data.

This time, I looked at the totality of evidence instead of the cherry-picked fragments that confirmed what I wanted to believe. I looked at Mendelian randomization, familial hypercholesterolemia, randomized trials, plaque regression imaging, PCSK9 biology, ezetimibe, statins, and the simple fact that lifetime exposure to ApoB-containing particles is causal in atherosclerosis.

That does not mean inflammation, insulin resistance, blood pressure, smoking, sleep, strength, or diet are irrelevant. They matter. But they do not make ApoB disappear. They do not make LDL particles magically harmless. A lean body and a good fasting insulin do not give the artery immunity from particle exposure.

I have now delisted my older YouTube lectures that were influenced by this lipid-denial ecosystem. I no longer want to contribute to the confusion.

One of the most common patterns I now see online is a medical influencer who has moved away from ordinary patient care, sells supplements, lives in another country (likely to avoid legal consequence) sells expensive diet consults, or gives generalized advice under the protection of “no doctor-patient relationship.” Many of these figures profit from the fear and distrust that exploded during the COVID era. That distrust is real. Some of it was earned. But bad institutional behavior does not make bad science true.

The people most affected by this misinformation are often conservatives. These are my people. They are skeptical of the system, suspicious of pharmaceutical companies, and deeply motivated to take responsibility for their health. I respect that. But skepticism without correction becomes a trap. A recent literature on the political polarization of health outcomes suggests that health and mortality are increasingly diverging along political and ideological lines, favoring better health in liberals.

My mission now is to stop the preventable damage.

I went from lipid denialist to lipid-lowering advocate because my patients, my data, and the broader evidence forced me to change. I still believe in strength training. I still believe in metabolic health. I still believe in nutrition, sleep, sunlight, and personal responsibility. But I no longer believe those things cancel out ApoB.

The new message is simple:

ApoB matters. LDL particle exposure matters. Prevention matters. And changing your mind when the evidence demands it is not weakness. It is clinical maturity.

I was wrong before.

Now I am trying to make it right.

Further readings:

Ference et al., 2017 — “Low-density lipoproteins cause atherosclerotic cardiovascular disease”

This is probably the single best paper for the intellectual foundation of my reversal. It integrates epidemiology, Mendelian randomization, genetic disorders, and randomized trials. The authors conclude that the totality of evidence establishes LDL/ApoB-containing particles as causal in ASCVD, with risk depending heavily on magnitude × duration of exposure.

Read the full open-access paper

Borén et al., 2020 — the mechanistic follow-up EAS consensusThis is the one to read for how atherosclerosis happens: entry, retention, and accumulation of ApoB-containing lipoproteins in the arterial intima and the subsequent inflammatory response. It helps resolve the false choice of “cholesterol versus inflammation” both matter, but ApoB particle retention is upstream in the atherosclerotic https://t.co/bimO7oAXRy the full open-access paper

Cohen et al., NEJM 2006 — PCSK9 loss-of-function mutationsThis is one of the most persuasive natural experiments. Genetically lower LDL from PCSK9 variants produced dramatically lower coronary disease risk. In one population, roughly a 28% lifelong LDL reduction was associated with an 88% reduction in CHD; in another, a ~15% LDL reduction was associated with ~47% lower https://t.co/WMQG5QpMtz the NEJM paper

Ference et al., JACC 2012 — lifelong genetically lower LDLExtremely important because it explains why short drug trials can make LDL lowering look unimpressive compared with lifelong exposure. Genetic randomization allows us to ask what happens when LDL is lower beginning early in https://t.co/hK2bO0ursL the PubMed article and access the full text

CTT Collaboration, Lancet 2010 — 170,000 participants / 26 randomized trialsThis is an excellent paper to revisit for objection about statin statistics. Rather than focusing on one mortality NNT, it looks at outcomes per absolute LDL reduction across a huge randomized https://t.co/LXjIQetsDl the complete paper free on PMC

CTT Collaboration, Lancet 2012 — statins even in lower-risk populationsThis one is particularly relevant to “NNT ≈100” arguments. One important distinction: an NNT around that range can describe major vascular events in relatively low-risk people over a limited follow-up period, but it is not a universal “statin mortality NNT.” Absolute benefit necessarily looks smaller when baseline short-term risk is https://t.co/5FkaeGiYbC the paper / abstract

My original skepticism was not irrational—the mistake was extrapolating a modest short-term absolute benefit in low-risk people into the conclusion that lifetime ApoB exposure was unimportant.

FOURIER — Evolocumab cardiovascular outcomes, NEJM 201727,564 patients with ASCVD. PCSK9 inhibition lowered LDL dramatically and reduced cardiovascular events on top of statin therapy. This is important because it dismantles the argument that the benefit is some peculiar “pleiotropic statin effect.” Different mechanism, same direction: reduce ApoB/LDL exposure → fewer https://t.co/R0oxkml6cm FOURIER in NEJM

ODYSSEY OUTCOMES — Alirocumab, NEJM 2018Another PCSK9 inhibitor, another randomized trial, another reduction in ischemic events. Primary endpoint: 9.5% versus 11.1% after ACS.And yes. Events were lower in the alirocumab arm. Lipid denialists will say mortality wasn’t reduced enough. But these were patients already managed and in the system able to get emergency care to save from death. Primary composite (CHD death, nonfatal MI, ischemic stroke, or unstable angina hospitalization):903 patients (9.5%) vs 1,052 patients (11.1%).All-cause death:334 patients (3.5%) vs 392 patients (4.1%). That is 149 fewer first primary events and 58 fewer deaths over a median 2.8 years. The relative reduction on the primary endpoint was 15% and met the pre-specified significance https://t.co/Ra4AM6xZoe ODYSSEY OUTCOMES

IMPROVE-IT — Ezetimibe + statin, NEJM 2015Also philosophically important: lower LDL using a nonstatin mechanism, and cardiovascular outcomes improve further. Again, it is difficult to sustain a hypothesis that the benefit is merely a special property of https://t.co/lJ6YaF9Yv5 IMPROVE-IT

GLAGOV — Evolocumab and actual coronary plaque regressionThis was one is Anti Aging Medicine. We can reverse plaque? That is regenerative medicine!. LDL reached about 36.6 mg/dL; percent atheroma volume fell, and 64.3% of evolocumab patients demonstrated regression versus 47.3% with https://t.co/9s9CT8pTVA GLAGOV in JAMA

PACMAN-AMI — Alirocumab, JAMA 2022This is even more visually compelling because multimodality intracoronary imaging showed not only greater plaque regression but changes consistent with plaque stabilization: reduced lipid burden and thicker fibrous caps. Again, this is Anti Aging and Regenerative https://t.co/PYbYLoN0Ns PACMAN-AMI in JAMA

ApoB vs LDL-C discordance — JAMA Cardiology 2024It demonstrates how widely ApoB can vary among people with the same LDL-C or non-HDL-C and quantifies clinically important https://t.co/8HMzQX7v1U the JAMA Cardiology paper

There is also a 2025 systematic review of 15 discordance studies involving 593,354 people. ApoB outperformed LDL-C in all 9 studies directly comparing the two in the review. That one is particularly relevant to explaining why I moved away from obsessing over LDL subfractions/NMR toward ApoB. Read the ApoB systematic review on PubMed

And REACT is especially important

The REACT study was just published in NEJM on August 29, 2026. It examined almost 17,000 ostensibly healthy adults aged 18–70 in Denmark and Spain. Atherosclerosis was already detectable in the 18–29-year-old group—8.7% of men and 6.7% of women—and plaque prevalence and volume climbed dramatically with age.

Read the new REACT paper in NEJM

“I am young, lean and fit, therefore these lipid numbers probably do not apply to me.” REACT shows why that reasoning is dangerous: subclinical disease can already exist decades before the person looks or feels like a cardiovascular patient.

The conservative-vs-liberal mortality paper

Elder & O’Brian, Nature Human Behaviour, May 14, 2026 — “The political polarization of health outcomes in the USA.” They found that conservatives in their cohort experienced worsening health during the 2010s and higher mortality in the early 2020s, particularly from internal causes such as heart disease, cancer, and stroke.

Read the full Nature Human Behaviour article

The authors found a mortality divergence associated with political identity; they also found that changing socioeconomic composition of the coalitions, including education, income, and insurance, explained part of the difference

[Andrew Dunn @AndrewE_Dunn]

ADDITIONAL NEWS: Anthropic says it is "drawing the line at Phase 1" for its own pipeline of rare/neglected disease programs.

Kauderer-Abrams tells me he draws that line as before Phase 1 studies begin, in an effort to not compete w/ pharma customers:
https://t.co/0uecsF1cS7

[Alex Harding @AlexHarding7]

@at_fome @AnthropicAI Yeah I agree with that

[Alex Harding @AlexHarding7]

The @AnthropicAI sequence mining announcement this week reminded me how serendipitous it was that SpCas9 was one of the first CRISPR-Cas systems described. Most other variants of Cas can't match it on its combination of potency, selectivity, and versatility.

Many companies have worked on better versions of Cas (@metagenomi, @arbortx, @ProfluentBio, @Basecamp_Res, @editasmed, etc). There are a few specific use cases (eg, different PAM sequence) where these variants can be useful, but SpCas9 is still the GOAT.

For the most part (again, there are some exceptions), those other companies are working on a solution without a problem.

[Tom Ellis @ProfTomEllis]

@AlexHarding7 @AnthropicAI This is what I keep telling my students. The real superpower of Cas9 is not the cutting DNA, it’s the ability to fuse it to hundreds of different proteins and express it in hundreds of different organism and still get it to work. Very rare for any protein

[Patrick Collison @patrickc]

Some data we recently assembled on entrepreneurship/compute in Europe: https://t.co/x8pbpHLun7.

We hope that one of the useful roles that Stripe can play is in collecting and publishing empirical data pertaining to entrepreneurship and industry in Europe. There's growing appetite to get Europe on a better footing, and cross-sectional comparisons can often shine light on where opportunities lie. If you're interested in this kind of thing, we publish more at https://t.co/YoZDuYbaBi.

[Alex Harding @AlexHarding7]

Verification makes discovery matter | Science https://t.co/TmUdkth0Yf

[Alex Harding @AlexHarding7]

@john_evans3 @daphnezohar This is the CEO of a $2T dollar company announcing it as if it’s important—I think a similar finding has probably already been done by AI 1000 times at other cos. Also same person who said AI will cure most major diseases in 5-10 years…