Founder mode often entails sequencing and the administration of drugs that are bespoke, or not undergoing conventional drug development.

Of course, two things are possible:

1. Founders are curing their own cancer/ even though most doctors failed (Edison mode)

2. They are fooling themselves and possibly others (E. Holmes mode)

Here are 8 things I ask to help me make sense of these claims:

1. How was the cancer found? Was it symptomatic? Or was it picked up on a scan or test most people don't get?

If a 72 yo person presents with stomach pain and yellow eyes and was found to have stage IV pancreatic cancer, then a doctor can give them a prognosis: Most patients live between 4 and 20 months. If the founder takes something novel and ‘cures’ themself and is alive 7 years later— I am inclined to say: yea, they did it!

But if a cancer was found on a preventive whole body MRI, or on a CT scan accidentally found cancer after a car accident, and it’s, say, brain cancer, then the brain cancer survival statistics don’t really apply. We know people with some brain cancers live 18 months from when they present with crushing headaches and seizures, but we have no idea how long people live whose grade 2 astrocytoma is incidentally found on an MRI others don't get. Is it 20 years or 40?

When I hear about a founder with grade 2 astrocytoma who's doing well 3 years later I don't know if that's because of the novel treatment, the early diagnosis, or just what you'd expect if you scanned everyone's brain with MRIs.

2. Did you get surgery, remove all cancer, and the ‘cure’ is keeping the cancer away? Or did the cure shrink the cancer?

If you had a tumor fully resected, and it is indolent, it will take a long time to reappear no matter whether you eat turmeric, Xeloda, or take a novel bespoke gene therapy. If instead, your tumor was actively growing, visible on scans, and your cure made it shrink— that has to be taken seriously.

3. After your treatment, did you use a surgery to remove the last of the tumor? If this occured, the treatment is less impressive than if no surgery was needed.

4. Is a biological correlate distracting? If a treatment increases immune infiltrate at the tumor site- that doesn’t mean much. If the treatment increases T cells targetting.. you name it.. in the blood— that doesn’t point me any direction. Does the treatment shrink the cross sectional area of tumor?

5. Did your novel, targeted drug used a heavy, nonspecific payload? If you make a new drug but tether it to MMAE or a powerful radioisotope, note that the payload can be cleaved after killing one cell, or before, and the payload can do cancer killing all by itself. A high enough dose of an ADC with active payload will work in other places besides the target. Microtubule inhibitors work in a lot of cancer. It takes extra to persuade me that the novel ADC is critical to that.

6. If you use your novel drug with a tried and true backbone— new drug + chemo— it is impossible to know what <new drug> adds. If you have a doctor who says “I’ve never seen responses this good with chemo alone, ergo new drug must have helped” That doctor needs more education — or should be recruited as your kol depending on your goal. All sorts of responses happen stochastically with tried and true drugs, including some great outcomes.

7. If you find your tumor has a rare fusion protein that is consistently found in the cancer— DNAJB1-PRKACA— for e.g. and you make a small molecule inhibitor targeting the kinase activity (perhaps leveraging RevMed chem), then yes, I am a believer. But if you use a peptide vaccine instead… not as much. The history of the platform and class and success rate of the platform and class matters.

8. I love animals— I love having pets, and I want to end unnecessary animal research. Having said that, curing your pet— dog or cat — is a wonderful thing, but not as impressive as curing a human.

Some of the stories that I read about founders are genuinely impressive and warrant further, exploration and others are almost surely drawing the wrong conclusion. The founder is participating in unproven medical therapies that are not extending their life, but they do not see it.

I wish everyone who decides to pursue their own treatment for their own disease the very best, and I'm rooting for their success. I just hope they're careful to see the difference between Edison and Elizabeth Holmes. These rules of thumb may help.

On complex policy issues, however, it tends to hew close to the social, political, cultural and scientific center. It tends to be impartial…. For now.

It's worth remembering that we debate these issues, and use politics to settle them because we do not often have the gold standard answer. We (humans) are biased and limited in our own thinking. Naturally, for questions, we cannot fully comprehend, different perspectives will emerge, and groups of people will coalesce around those perspectives.

But as AI becomes smarter, it is possible that AI settles the question in a definitive way. How will we react?

What if AI says:

Medicare for all will provide care that will be considered suboptimal to most American consumers, leading to widespread resentment. It will do so at incredible societal cost, and face impossible questions about how far to extend the breadth of coverage. Decisions on what to cover will be end product of political lobbying and not the best available science, and rational arguments about evidence will fall on deaf ears. Ironically, a health care system that does not promise to treat everyone will result in less resentment and slow (but not avert) the de-stablizing spending curve we are on. As such, Medicare for all is a bad idea.

This specific medical product does not work. The historical data are in line with the range of expected outcomes. The mechanism of action is unsound. Subjective benefits are largely wishful thinking and the placebo effect. The harms and invasiveness are real. A rational person would conclude that no data presented has altered the skeptical pre-test probability. Expert sentiment is misleading and likely biased. Avoid this product.

Although your policy to tackle (insert controversial issue, eg. Climate change) feels good, it is unlikely to provide material benefit, and much more likely to lead to the misappropriation of funds. Upper bound projected impact on atmospheric CO2 is negligible. This occurs a tremendous cost.

Actually, on this topic, Robert f. Kennedy is correct, and considerable uncertainty remains about the long-term use, year over year, of SSRIs in the youth.

In short : what happens when AI weighs in on political issues.

You may be thinking this has already happened. AI models are heavily biased towards the liberal epicenters that have developed them, and favor liberal answers.

I agree this has been the net bias of AI to date but those models were not truly intelligent. AI was pantomiming its creator and not thinking.

At least some of the mathematical solutions AI has now offered surpass human beings and are also clever, innovative, and show flare. At least some of these proofs are correct. For some very specific question with a right answer, AI has reached that right answer in a clever way— often by bridging disparate domains of mathematics.

As model complexity grows, it is very likely that AI will do even better. Does my drug offer a favorable risk benefit profile? We treat that like a subjective question, and largely it is, but part of the reason is our minds are limited, and our understanding of biology is fragmented and incomplete. There is an objective right answer. What happens if AI reaches it indisputably in its 'mind'.

Is Medicare for all a good policy choice? In the near term, I envision AI weighing in on this question. Perhaps different systems will converge on the same answer. How will people respond when we start to believe that AI may no longer be biased, but have a point?

When I was in high school, I had massive overlapping classes with my colleagues. As the years went on, you would see Jimmy answering questions correctly across many classes. Jimmy was writing great essays. Jimmy was aceing genetics tests. Jimmy was a great student, and over time we all got to see it. So when Jimmy said something provocative, something we didn't initially agree with, we decided to give it a second thought because it was Jimmy and he was no dummy.

This happens less in our modern online worlds as we don't really know each other's thinking well.

As AI solves tangible real world problems it will gain the credibility of Jimmy. If it answers political questions definitively, and in a heterodox manner, betraying no bias to one side or the other across questions, It may fundamentally alter core assumptions about governance.

In an increasingly tribal world, it may destabilize our political platforms.

Here is my review and much more about Elizabeth Holmes and Theranos. (Spoilers to follow)

1. Elizabeth Holmes is psychotic— a sociopath. She is a master manipulator, with intense interpersonal charisma. She lacks remorse, understanding, and at times a sense of humor. She should have been a politician— she would still be in Washington.

2. She defrauded investors and put patients at risk by repeatedly claiming her testing platform could do things it could not.

3. If I make a robotic box that promises to make a cheeseburger in 2 minutes, but can’t— you can easily verify that it cannot. If I make a box and tell you that keeping on your nightstand for decades will help you live longer and better— that is harder to verify. Elizabeth Holmes made the first box— her claims were proximal and immediate and easy to measure. Many in biotech make the second box. Drugs and devices and biologics that promise a longer and better life.

4. Elizabeth Holmes is somewhere on a spectrum of startup founders lying. She overstated claims. So do many others. She is probably only in the 80th percentile of lying to be honest among those who claim to offer revolutionary services. She would barely stand out at a SF founder party.

5. Elizabeth Holmes is somewhere on a spectrum of selling health care services that don’t work. If she made a gene therapy or drug for neurologic disorder, and made similarly false and harmful claims, she may yet still have a multi-billion dollar company. It takes years— if at all— for the false promise to be detected. She could just change her endpoints mid-trial, give her drug in combination and without a control arm, or claim that you can’t randomize at all and use historical controls. So many ways biotech Theranos could succeed! Theranos is far from the only fraud, i.e. product that claims to make you better off, but doesn’t.

6. Elizabeth Holmes was knocked down for several reasons 1. she is psychotic 2. her product doesn’t work, but also 3. she flew too high— too much attention (people love to see such successes fail). It makes us all feel better to know the Harvard president was a plagiarist— if you learned the rector at Ghent was a plagiarist, you would shrug. It’s Ghent after all.

7. Her prison sentence of 10 years exceeds even armed robbery sentences and seems a bit harsh. It feels ‘scapegoaty’. — oh dear, have I also been conned?

8. There is no interest in medicine in separating real innovation from pseudoinnovation. Singling out Theranos in the latter category, but not seeing it in countless analogous situations reflects that.

9. The journalists who cover biotech lack medical expertise and can’t see that Theranos is exactly the same as…. the company they are plugging in the next article promising future breakthroughs.

10. In contrast with most fraudsters, EH did not sell stock, she did not buy nice houses or fancy cars. Her fraud was so deep she believed it too.

Fielder’s movie is dazzling, and one cannot help but feel that he too becomes seduced by the charms of Elizabeth Holmes. His entire film, which many will view as indicting her, is perhaps favorable given the extent of the evidence against her.

At times, the viewer wants to know much more than Fielder probes. How did she put Billy under her spell so quickly in that first conversation? What happened afterwards? In the days and weeks to follow. Why did he not interview others at Theranos?

The viewer, like Amanda Seyfreid, wonders about the Fielder’s objectivity, as he alters the dialog he remembers from the prison visit— because, as he notes— he didn’t feel what she said was that incriminating.

Fielder portrays Billy as controlling, possessive, but I suspect the reality is that EH has orchestrated this. Just as the viewer wonders whether Holmes orchestrated the payment deal for the film, and the inner workings of Theranos itself.

Finally, we should not forget that Holmes chose Fielder— specifically recruited him, a comedian, to make the film. While many will think that decision backfired, I am not so sure— any other film maker may have even had less sympathy for her. The visual of him and her swimming together— a brilliant cut by the editor— leaves the viewer wondering long after the film.

If you enjoy this writing, subscribe

Subscribe now

The New York Times spoke to a half dozen lung cancer experts about the pope. I was surprised to find that despite reaching out to so many, the ultimate reporting was lopsided and inaccurate. It lacks humility.

Here's what we know.

The Vatican states that a lung nodule was discovered during routine imaging the pope was undergoing.

Since this is 2026, and since the pope is a VIP, it is possible they are referring to some type of whole body imaging or a coronary CT scan. If you're a 70-year-old who's otherwise perfectly healthy, and has no complaints, and is a never smoker, reasonable people can disagree about whether or not you need any of these scans. Whether or not any will benefit you. That calculus shifts if you've had symptoms. Then 99% of people think you need the scan. As do I. By all accounts the Pope did not have symptoms.

There are three possible scenarios for what the pope is going through. The first is: the pope was fortunately found to have a localized, early stage lung cancer, which has undergone surgical resection, and may or may not require some adjuvant therapy, but he is cured. Were it not for the providential diagnosis, this cancer would have spread and resulted in untimely death.

The second scenario, which we all hope is not the case, is that the pope indeed was diagnosed with an early stage lung cancer, adenocarcinoma, but despite these efforts made at this moment, the cancer has already spread microscopically and will reveal itself in a matter of time-- ranging from months to years. In the absence of the early diagnosis, the pope would have been diagnosed at this future date when the cancer had spread. Thus, the scan has merely added lead time to the Pope's diagnosis without possibility of benefit.

The third possibility is that, by virtue of the Pope being a non smoker, by virtue of the biases of VIP medicine, he has been diagnosed with a nodule that would not have progressed or caused harm in the rest of his natural life. Had the pope not undergone the scan, He would have lived happily for many more years to come. As a result of this imaging, he has undergone unnecessary surgery and is left with a scar, and psychological burden, but is no better off. To be clear, I'm confident a pope would be the least troubled by considerations of his own mortality, as I am sure he has transcended these mortal thoughts. But for many, this psychological burden can be fierce.

The truth is not even the Pope knows which one of the three scenarios happened. Only God knows. Certainly not the experts quoted in the times.

My objection to the New York Times coverage of this topic is that they are utterly unable to consider the third possibility. None of the experts raise it. In fact, they overwhelmingly think the only thing that could have happened to the Pope is option 1. And the article ends with a quote about how we all should be so lucky to get the scans the Pope gets.

Should help a 70-year-olds who have never smoked line up to get CT scans of the lung? Whole body. MRIs? My answer is no. They should not for the simple reason that these interventions have not demonstrated benefit.

Even in a population most likely to derive benefit- long time smokers-- ct screening for lung cancer is disputed. It has failed in a follow-up publication by Pinsky regarding the nlst to sustain a mortality benefit. The Nelson trial is a disaster. It's protocol amendment after protocol amendment with no all cancer mortality reduction. It's like duct taping a rocket and trying to fly to Mars.

Medicine is not an exact science. One of the hardest things to know in medicine is whether your patient did well because of what you did, or in spite of it. Did they do poorly despite all you did, or because of it. These are deep questions. They are nearly religious in their nature. They require open-mindedness and equanimity to consider. Those are the traits of all popes, including this one, but sadly not the news, reporters or experts quoted in the media.

I wish the very best to the Pope.

He is missing nothing. Two decades ago, leading science figures like Collins promised that genomics would affect all aspects of medicine— in sickness and in health.

Sequence your genome to learn what blood pressure pill or heart burn medicine to take— that was the promise.

Yet, Brian is correct that for healthy people (or those with common conditions, e.g. HTN) the answer is: genomic knowledge has little to no practical value.

It isn't that you are perfect and can’t improve. It's that we all need to do the same things. Diet, exercise, medication, lose weight, (for Brian, swapping blood from you teenage son, wearing a red light, swallowing a thermometer and going in the sauna)— these all can be recommended to everyone regardless of genome.

Of course with disease it's different.

In cancer medicine, the genome is use commonly. Mostly we use the Cancer's genome. Not yours. Only a tiny handful of HLA specific therapies, transplant, use a portion of your genome. Many drugs target the Cancer's genome— regardless of yours.

The biggest drugs of the last two decades have nothing to do with the genome at all. Their large market share is buoyed by the permissiveness of the label. Pembro makes hundreds of billions precisely because it works so indiscriminately.

Rare and childhood diseases have also benefitted from genomics in diagnosis— though treatments often lag.

For some diseases, we now have bespoke gene editors. For conditions that present with childhood symptoms, genomics is often vital.

Fixing small defects in enzymatic processes is one thing, fixing large structural defects, where the defect proteins pile up in cells over years, and then hoping to undo the damage is quite another.

Ultimately, the idea that the human genome is like the letters in a book that tells you how to build and run a person— that analogy has done a lot of damage. The human genome sequence is like taking a Lego sculpture in a lego exhibit- one with 10 million pieces— putting it under a hydraulic press, and then looking at the patterns of colors, trying to deduce the shape and function of the original object. DNA is a moving, living molecule with 3 dimensional interactions we can scarcely understand.

It is nothing like letters in a book.

If I had to steel man the case for why the trial is in fact phase 1, I would argue that the authors built in a mechanism where a third party could abort the patient's use of AI if they felt it was unsafe. This did not happen. The AI did not try to get any patient to abandon his wife and fall in love with the machine instead. Something which definitely happened to a New York Times reporter.

But if I put on my common sense hat, I would argue that it's not a phase 1 study. Phase one studies are nonsensical in this context. Patients are already using AI to navigate their questions and understanding on medicine and every other topic under the sun. Do we run phase 1 trials of breathing air, using the toilet, or eating oranges?

An uncontrolled study of a AI clinical application is better off called: uninformative, lacking appropriate controls, non-randomized, not too exciting— these words are more accurate descriptors than ‘phase 1’

I struggle to understand what we're doing in medicine regarding AI. Why are we asking questions to three different AI machines and then having random doctors grade the results? By the time that paper is published, the AIS are all different. What does that answer exactly?

If you're talking about integrating AI into the hospital workflow: does it need trials at all? Most of the changes in medical practice I've witnessed in the hospital have just been debuted.

One day someone decided to put the shittiest* desktop computer on a push cart and attached a car battery to it. They called it a cow. This was widely debuted and no one studied whether or not it was wise to ruin the walk and talk. Having a half dozen people exhibiting the shopping cart sign and staring glass-eyed at their computer became the norm. There was no phase one study.

A few years later we had to stop calling it a cow. This was around the time we had to stop saying pregnant woman and start saying pregnant person. Before people ran phase 1 trials of AI, they kept busy by thinking of words you couldn't say anymore. It's now a wow— workstation on Wheels.

Before the cow it was the pager, after the pager, it was epic messaging on your phone. Did anyone run a phase 1 trial of epic messaging on your phone? Definitely not. It would have hit MTD at the first message.

Grade 5: medical doctor threw his phone across the room. Iphone 11 dead.

Instead, many believe that AI improves patient outcomes. In that case, you should do… spoilers to follow… drum roll please.. a randomized study.

It turns out it's simple. You randomize people to doing things the way you're currently doing them, or to integrate the AI. Here's the tricky part: you measure something that actually matters to patients.

You measure their length of stay in the hospital, or their 30-day mortality. You could use the endpoint of a survey that patients complete about their experience, but such an endpoint would be pretty dull. It would likely heavily depend on the phrasing of the question and might not be a good measure of global quality of experience.

But we don't have those studies. I’ve yet to see a single one. In the meantime, AI is the cow. Someone's pushing it wherever we walk.

Should we start r chop on this patient?

Let me ask open evidence

Wait, did you do a phase one study?

For r chop? I can ask open evidence.

No! Are you crazy? A phase one trial of asking open evidence.

Do you want me to ask open evidence?

No, I want you to run a phase one trial of asking open evidence before you ask open evidence, which obviously is the most pressing research question of our times.

Should I do that before we round on the pregnant person?

* Sorry for profanity but it isn't a cow unless the computer that's on it is incredibly slow, unresponsive, with a poor internet connection. That's mandatory.

Instead, he envisions a scenario where AI cheats and achieves a phase three victory even though a drug doesn't work. Is this possible?

Yes.

One way would be to actively falsify data for the phase 3. This has happened and been detected a few times— one recent example got 0 news coverage oddly— and it may happen more often but we don't know because despite all the regulation on the world, science is based on trust. There is no system that verifies if 100% of primary study reports actually reflect what happened to a person. Such a system would require an insane amount of oversight.

Another way to fake a phase 3 result would be to use an endpoint that is vulnerable to informative censoring and then pair it with a toxic chemical that causes drop out not at random.

What if the endpoint is PFS and what if your drug made sicker people in the experimental arm skip future scans? This would create a false PFS of the experimental arm. While I cannot prove this has ever happened, there are several drugs that appear to have all these properties, and certainly smell like it.

Creating a fake OS read out in intention to treat fashion could be achieved with a subpar control arm or inappropriate post protocol care. Both of these have also already likely happened.

What if a (non cancer) drug has serious toxicity and drop out and the manufacturer tries to use a missing at random analysis or an imputation that doesn't assume worst case—- this happens.

The problem with Tao’s analogy if interpreted this way is that you don't need to espouse AI. We are plenty good at making drugs and trials that project a false benefit for marginal and ineffective products. We do that with bad control arms, differential censoring, poor post protocol care, changing the SAP after we looked at the data, and, in rare cases, cooking the books. We put very little effort to catching this. You don't need AI at all— in fact what a waste of tokens. It's cheap and easy to do.

And all that said, most cancer patients would still be willing to try even accepting the possibility of gaming. Tao’s analogy still falls short.

But don't try to use a cancer analogy if you don't understand the field.

This week Terrence Tao argued that patients would want the reassurance of knowing a human cancer expert understands an AI derived cancer cure before getting it. This betrays a deep ignorance about cancer medicine.

1. There are many commonly used cancer drugs where we do not fully understand the mechanism of action. If you think you know how Adriamycin works, I think you haven't read enough about it. We still debate how docetaxel works in prostate cancer. Is it microtubules that are trafficking AR receptor, or is it the cellular division?

There is no underlying theory that explains all of the places ibrutinib works it and where it doesn't. The same goes for venetoclax.

2. A drowning man will reach for the blade of a sword. If a phase 3 clinical trial is positive, as Tao postulates, but humans do not understand the mechanism of action that would be level one evidence to administer a therapy. Every cancer patient on Earth would want it. Every doctor would give it. Tao is not even close to the bleeding edge of what patients will want or not want. Patients will want a drug in a tumor that lacks phase 3 data if it has phase 3 data in a different tumor, and a compelling mechanism of action. Patients will want a drug that has phase 2 data that's very promising. Patients will want a drug that has been given to five people and has a promising response rate. These are all things doctors will do in the right situation. Patients will even want drugs that have never been administered to a person but have helped a mouse or fly. This is where some doctors would give it and others wouldn't. Here I would warrant caution, but I just want to point out the depth of desperation.

   A drowning man will reach for the blade of a sword, and we hope cancer patients do not go as far as to seek unproven remedies that can only harm them. But they certainly do. If AI generates a cancer combination cure that no one understands, but passes a phase 3 randomized control trial, there's not a single cancer patient on Earth who would say no thank you. We currently accept far less.

3. What Tao is missing about cancer medicine is something fundamental and deep that many are missing it. People know how computers work. People know the edges of mathematics and can check the theorem produced by open evidence. There is no one who knows how the human body works. We have a much deeper understanding than we did a hundred years ago, but we don't know if we have understood 5% or 50% or .05%. . We don't even know what we don't know.

4. The human body is a collection of trillions of cells, and is the constant interaction of the three-dimensional shapes of countless intracellular and extracellular products that interact in emergent ways that we have not even close to grasped. We are the prisoner in Plato's cave looking only at the mere shadows thinking we understand the world.

5. Despite this blindness, through serendipity and trial and error, we have deduced interventions that do improve outcomes. In many cases we have some plausible story as to how they work. But even these stories are partial and incomplete and not fully vetted. In other cases we have no idea how they work. No one knows how inhaled sevofluorine knocks you out. At the same time I know with confidence I don't want you operating on me unless I'm breathing it in.

6. If mathematicians want to complain that computers are replacing them, they should seek the comfort of chess Masters complaining that they have been bested. But they should not look for comfort from doctors. We would do anything to help our patients. We would do anything to extend survival. We already do things on a wish and a prayer and a mechanism of action and a hint of bioplausibility.

If AI conjures up a cocktail to cure cancer and we don't know how it works, but it passes a phase 3 trial, you can bet we will be on our knees thanking it, or more likely giving a standing ovation at our plenary session.

What is the correct control arm?

An ethical trial randomizes participants to the best available standard of care versus an intervention you think might possible improve things. That intervention could be on top of the best standard of care, or in leu of it.

Many trials meant to inform the US practice are unethical because the control arm is something we have not done in the US for quite some time— a standard that was already beaten. This is the norm in oncology, where it is much easier to be a dilapidated and antiquated treatment then the best standard of care. We documented 17% of FDA pivotal studies used such an unethical control arm (Hilal, Prasad, JAMA Onc).

Other trials are conducted in low and middle income countries, and meant to understand the impact of therapies in those countries. For instance, you can randomize participants in sub-Saharan Africa to a mosquito net or no mosquito net and then measure malaria. That's because (prior to your study) the current standard of care was not to have a mosquito net, and you are asking whether or not it provides a benefit. In fact, at least 20 of these studies have been conducted and they work. But it was also possible they didn't work. Maybe people wouldn't go under them enough. The trials settled the question.

Randomized trials are particularly important for bitter and divided topics. If half of doctors think stenting stable coronary lesions saves lives, and the other half vehemently disagree, running a randomized controlled trial is of tremendous value.

On matters of public health, we unfortunately have to deal with a group of ignorant zealots. These are people who believe that they know what works based on mechanistic science and low quality observational studies. They know, for instance, that HEPA filters would save lives if we just installed them broadly. They know that cloth masking a 2-year-old saves lives. They often “know” these things with religious fervor.

When it comes to hep B vaccines, there are many important questions.

For the US: there has never been a randomized controlled trial that gives neonates, whose mother tests negative for hepatitis B while pregnant, and who are at low risk (i.e. not an IV drug user), a. Hepatitis B. vaccine on day zero versus delaying that vaccine by 2 months, or 10 years.

Such a randomized study would be imminently important in the United States, where this issue is heavily debated, and neither side has gold standard evidence, but the zealotry side uses bizarre arguments. Before we did universal day 0 shots, hep b killed x number.

Ok, sure, before firefighters poured water on a burning building, it was on fire, but now that it has stopped, we might ask if we can install a sprinkler system and turn off the hoses.

Researchers from Denmark however, are interested in a different question. In the GB, hep B vaccination at birth is not universally done. Instead they do it at 6 wks, 10 wks, etc. They want to randomize infants to having this added at birth or usual care. They will measure all cause mortality. In the absence of the study GB won’t launch universal hep B vaccination till 2028.

It appears the trial is ethical, feasible, practical and would be quite useful. GB has more hepatitis than the US, if the trial will show net benefit anywhere it will be there. If it fails to show net benefit there, or if there is net harm, then the US’s pre 2025 policy would be certainly net harmful.

If you ask me, I think running the trial in the US would actually be the better trial. It would more accurately impact US thinking and practice. I would be the US study is negative — i.e. totally fine, and probably better to omit Day 0 hep B if the mother is low risk and screens negative.

The GB trial is more uncertain. I suspect it may even give a favorable outcome, but I don’t know for sure. Yet, that’s why we run randomized studies.

Yet no one gets more publicity for being snubbed than Rosalind Franklin. Science is back again with a new piece saying she knew she had found a helix. Do you hear that? She knew it!

I can imagine a pompous James Watson saying, well “If you guys were the inventors of Facebook, you would have invented Facebook”

“If you had elucidated the structure of DNA, you would have elucidated the structure of DNA.”

I've been hearing about how Rosalind has been wronged for my entire career in medicine 05- present. Now that she's dead, Crick is dead, Watson is dead, Pauling is dead, Wilkins is dead, the Nature editor is dead, the 1962 Nobel committee is dead, I honestly don't see how it matters… but sure, I'm happy to say she deserved more credit.

But what I find odd is there is no love for the living Rosalind Franklin… Dr Pan. Who independently discovered optogenetics slightly earlier or contemporaneously with this week's Nobel prize winner.

Because Pan was at a (a) low ranked university, because he is (a) not a politician (b) self promoter, (c ) networker, he walks away with no recognition. Just a paywalled stat article.

Just like James Watson trash talked Rosalyn, Stanford trash talks Pan

“A far cry”. That seems technically incorrect and particularly bizarre that it's being made by a university spokesman. It seems oddly similar to the way James Watson talked about Rosalyn Franklin.

As much as I appreciate Science trying to correct a historical Injustice, I wonder why Science is not interested in current injustice.

Yes I too would like to solve the unsolved murder from 1953, but a dude was killed on Monday maybe we should look for his killer too.

You can read the full Dr. Pan article here.

https://archive.ph/FAqsp

Supposedly in Russia, a lab worker dropped a tube of the plague— yersinia pestis— and died of pneumonia.

Here is what you need to know…

Some online speculate that because Russia has gone hard on containment there is more here than meets the eye

That’s entirely possible. If indeed we were talking about garden variety Yersenia, then it seems a much lighter touch could contain spread. But on the other hand, Russia might just be erring on the side of taking this seriously… unlike our friends in Wuhan.

Even if the Russians have juiced Yersenia, making it MDR, I have my doubts that it could turn into a global pandemic. The organism itself has a fundamental biological R0 limit. In contrast, coronavirus is massively contagious at baseline.

It is hard to get worked up these days given how much of public health is endless fearmongering. Monkeypox, hantavirus, ebola, measles in Pennsylvania… the public health apparatus seems keen on rekinding the panic of 2020. Some legitimately seem sad to learn that hanta-virus was a nothing burger.

I see some already predict that mRNA will be the solution to this problem.

Let’s try to learn from the last rodeo, and next time, let’s be skeptical of products without robust randomized evidence. Thankfully Yersinia is unlikely to be that next time.

Bottom line: the real lesson is to give up studying these deadly pathogens in labs where error is inevitable.

Before I work up to to my main thesis let's agree on a few things/ subtheses.

Everyone agrees some people are snubbed

First, consider yesterday's Nobel. Everyone agrees that some people were left out who should have been included. Ed Boyden. Feng Zhang (snubbed also for crispr). The Wayne State University professor who actually did everything first who wasn't well published because he was at a second tier school. I forget his name—that’s how bad he was screwed over. Etc etc. The list goes on and on. I am even snubbing many of those who were snubbed yesterday. Sorry about that. But I made my point. Have to move on.

Did Optogenetics save lives? (Aka who decides which discoveries are great?)

Optogenetics is cool— very interesting biology— and might someday save lives, but so far hasn't lead to any commercially approved products. (Am I wrong?). GLP-1s saved a lot of lives and might win in the future. TKIs saved many lives but probably Brian Druker will never win, having missed his window (aka annoyed someone in Stockholm). Jim Alison won for CTLA-4 but that is a way worse target than PD-1. All recent winners did cool science (not true historically—the lobotomy dude was crazy) but not all of it has equally improved the human condition. Don't @ me. It's true. My point is just which discoveries matter is ultimately subjective.

If you deleted all yesterday's winners and all those who were snubbed, would Optogenetics have been discovered? If so when?

If you deleted the top 30 people who worked on optog—- Optogenetics would still have been discovered (+ or -) 5 years. If you deleted Brian Druker, Lydon, Sawyers, CML would still be drugged (+/-) 3 years. If you deleted Doudna/ Charpentier, Crispr would still be a thing +/- 0 years.

I think the lay person doesn't realize that science is often a forced move. Because of advances in computer chips, and a theory of neutral networks, it is imminent that someone makes an LLM, Dario and Sam did, and so did Mark and Sundar, and China. Etc etc.

Similarly a lot of science is just around the corner whether by the winners or someone else. This threatens the hero narrative of science but is totally logical that technology builds on each other.

Nobel prizes aren't like being the first to climb Everest, they are like being the first to sleep under this tree.

Edmund H and tenzig norgay (snubbed) were the first to summit Everest. They had a team helping them, but I can't remember their names… sorry Wayne State professor and Ed Boyden. Walter Bonati never got credit for K2.

But everyone knows what a mountain is. And it was known that Everest was the tallest. Science is often more unpredictable. The Nobel isn't like being the first to top Everest, it is like being the first to sleep under this tree. Not that tree.

Imagine I find an oak tree in California and said: who was the first to sleep under this tree?

Well in 1883, 17 people slept under the tree but 2 didn't get their legs fully under. So Ed Boyden and the Wayne State professor don't win the prize.

The reason the tree analogy is better is often in science we don't know what is going to be the useful discovery when it is first made. Cripr immediately had implications but this wasn't true of all prizes. For others the importance took decades to realize. Ironically crispr work in the 80s fits this bill. Some discovered were like Everest (watson and crick) but others were like this Oak tree.

Nobel winners aren't the smartest or best they were smart and also lucky

Not only were they lucky not to be snubbed, they were also lucky to sleep under the right tree. Many great scientists never slept under that Oak. But lots of scientists sleep under trees.

Nobel prizes reflect poor scientific understanding

Sometimes I see a smart person in medicine saying something embarrassing. Our observational study shows that flu shots sare associated with less Alzheimers — ergo flu shots slow Alzheimers! (Not healthy user bias). I ask myself, how does a scientist not know this? How can they make this mistake?. Then I realize you can be a scientist with a poor grasp of the history and implications of science. Dare I say belief in Nobel prizes is the same thing. It shows someone who doesn't understand the retrospective nature of how importance is assigned (it's a tree not Everest), the luck to sleep under it, the luck not to be snubbed (your head wasn't under), and that discovery is not the measure of being a good scientist— which is about a process and not an outcome, which is about a lifetime of work, and not 2-3 years of effort (in your early career)

Should we ban Nobel prizes?

No of course not, but we should treat them the same way we treat that observational study linking the flu shot to Alzheimers. With a grain of salt.

I look forward to your angry comments.

PS. Read my nytimes essay. I said this a decade ago. Sadly Brian probably will never win. He probably angered the wrong voting person or they think his work is too clinical (another irony). Silly Nobel committee.

Academic politics are the lowest kind bc the stakes are so low.

Mr Sasse has been courageously upfront about his therapy for pancreas cancer. Which included some chemo, Daraxonrasib and now is gem abraxane.

After Rasoncue and before Von Hoff, Ben said this

He underwent a major liver resection to “clear out” cancer and send tissue for molecular testing.

I often hear cancer doctors complain about the charlatans who prey on our patients. They give febendazole and high dose vitamin C and anti neoplaston therapy. Interventions that lack careful randomized trials showing benefit. I agree with this criticism.

But strangely my colleagues don't extend the— you don't have good evidence argument —to metastasectomy for metastatic pancreas cancer, oligo met radiation for many tumors and dozens of off label tkis— all of which happen daily at top medical centers.

I don't see a bright line between vitamin C and unproven surgery. In fact, arguably vitamin C is more benign, and less invasive.

Why is their snake oil bad but yours good?

Everyday the line between huckster and cutting edge medical center blurs.

The line between a fringe layman and medical expert is increasingly hard to discern

The future of medicine is bleak as we either have to believe nonsense from a random person on the internet or nonsense from a provost at Penn. The uneducated uneducated and the highly educated uneducated competing in a race of poor thinking— each confident the other is the fraud.

The truth is: both often are.

** If someone wants to say that emerging evidence supports multiple tumor resection/ debulking for second line pancreas cancer. Please read about the limitations of selection bias and retrospective observational studies.

The blood pressure effects are massive

Literally, the only concerning signal is S22; Its amazing that despite the boatload of weight shed, and the plummeting BP, the HR is still elevated years later. (I am assuming the HR also includes the fact that ~half stop taking the drug due to too much weight loss & intolerance) So the elevated HR among those taking the product is probably 6 bpm higher.

If I were a betting man, I would bet that these folks will live longer and better on Reta than placebo. Just because the benefit on BP and LDL and weight will far outweight the potential harm of faster HR. Consider:

The mean entry BMI is 40, and entry weight was 250 lbs.

Now lets compare against tirepatide. Cross trial— i know— but slightly less weight loss, slightly less HR elevation

Now semaglutide, a little less HR elevation still, with less weight loss/ BP reduction

If you take an overweight 45 year old, should they start on high dose reta, low dose reta, tira, or sema, and what about an oral?

Which is the best in the long haul?

That’s the big clinical question for which we need a large, pragmatic, multi-arm, un-conflicted study.

We also need to know what happens when someone is 36 and their BMI is 26 and they start taking the product? This is already happening.

Finally, there are still the unknown unknowns— what happens after taking the medication for 15 years?

Only time will tell.

The former senator has been open about his treatment, which included some chemotherapy prior to the Ras inhibitor and is now on Gem-Abraxane, which is sadly not curative in this setting (2+ line, metastatic pdac). Here is one study

There is deservedly a lot of enthusiasm in pancreas cancer, after decades of no new options Ras appears to be one, and Tango is reporting impressive results. Yet, the headline obscures some facts.

92% response in pancreas cancer!

92 % is 11 of 12, but only 9 of 12 are confirmed objective responses — so the response rate is more accurately stated as 75%.

Oh, but the trial is only for MTAP deleted and Ras mutant which is about ~10-20% of pancreas cancer , and 8 people were followed for less than 14 weeks and excluded

See the fine print

“59 patients with previously treated MTAP-deleted and RAS-mutant pancreatic ductal adenocarcinoma (PDAC) or non-small cell lung cancer (NSCLC) were treated with a vopimetostat-based combination with either daraxonrasib (n=20 PDAC; n=5 NSCLC) or zoldonrasib (n=34 PDAC)”

So the response rate is 45% RR 9/20 in MTAP-D and Ras-m PDAC, compared to 30% for daraxonrasib alone in the relapsed setting.

This is not meant to be discouraging, but to point out that there is still an unmet need here. The median DOR is not reported. It is 8.2 months for daraxonrasib alone.

And sadly, what is not stated, is that 100% of these patients will eventually have disease progression and die.

If you're 53 years old and you get pancreas cancer, it's certainly better to live 2 years than one, but both are a tragedy. And we should be honest that a person may lose 29 or 30 years of life by having this disease, irrespective of the new advances.

Three of the biggest public health issues right now that are not discussed:

Who will be screened imminently with multi-cancer early detection tests— will they experience net benefit or net harm— when Grail is approved and later covered by Medicare (Congress already authorized the payment)— what will happen?. What about when imaging is added— like midjourney or prenuvo? What about all the downstream investigations?

Which GLP-1 will save the most lives? Are some dangerous? Will the increased HR of retratrutide mean that it is worse than tirezepatide? Running a head to head trial of these agents powered for hard endpoints would be a boon— and will soon be a vital question. Raising HR is generally not a good idea with weight loss products.

What if: a widely used prescription medicine that is supported by disease specific regulatory endpoints increases mortality. This could be catastrophic. How could public health detect and stop this?

These three questions will soon account for a trillion dollars in health care spending, and many millions of disability adjusted life years. The importance of them is staggering.

But public health doesn’t talk about these things.

Instead, I hear endlessly about a measles outbreak in Pennsylvania, and some contaminated lettuce from Mexico. I hear panic then silence about Hantavirus on a cruise ship, and that I should take an annual flu shot. I hear about the importance of pregnant women taking SSRIs and tylenol.

In some cases, I don’t disagree. MMR is important, contaminated lettuce is bad, and pregnant women should minimize medications (like SSRIs), but the evidence linking Tylenol to autism is weak. At the same time, taking Tylenol for no good reason is silly.

But none of these are the most important public health questions of our times. Public health is so nakedly liberal and reactionary that it only talks about issues in response to what the Rs are doing, and can’t seem to formulate ideas on its own.

Measles: Sorry to break the obvious news, but the die was cast for vaccine hesitancy. Forcing people (many who already had the disease ) and those at low risk to take a extremely reactogenic, novel product that didn’t help third parties (whoops!) was a public health error. It undermined trust. You can’t walk that back. A CNN news article or Josh Shapiro presser— won’t change the arc of hesitancy. In the next decade, many will get measles and other preventable illnesses, a fraction will have bad outcomes, and then the arc will bend towards vaccines. Nasty lettuce is a perenial problem, and hanta virus on the cruise ship was a hysteria.

Public health is obsessed with these things but the delta public health (ability for anyone to change the future) is limited. Instead, multi-cancer screening, optimal therapies for weight loss, drug safety— these are massive under-explored topics.

People forget that public health being synonymous with extreme liberal politics— doesn’t just mean that it makes frequent cognitive errors to sate short term political preferences— but also that it is incapable of scanning the horizon and deciding which issues are worth exploring in the first place. Instead, it craves the daily political news cycle.

Meanwhile universities continue to avoid fixing their own DEI policies. Cambridge has changed it’s plans. It is no longer investigating how a serial plagiarist/ fabulist become the youngest ever tenured Black professor— how many errors occurred to let that happen? No, instead, it is focusing solely on what Cambridge could have done to shield the professor from media attention and assist him as he contemplated suicide. For the record, I think an investigation of the latter is warranted. But surely, an honest institution must also perform the former. It appears Cambridge lacks the backbone to do this.

The issue is not confined to one side of the pond. Here in the US, the justice department is slowly working it’s way through medical schools to find title VI violations— where race is used as a sole or key factor for admission (which inevitably means discrimination to other races). These letters are now hitting close to home.

There is a major unspoken debate at the heart of this. Should we hire the people based solely on who you think can best accomplish the task— the MEI idea… Or should you take into account past injustice, and work to create a workforce that looks like the communities they serve (DEI— and remind people of that in mandatory annual seminars)?

I just want to point out some interesting things about the contours of the debate.

Not long ago being a professor at Cambridge or Harvard was one of the most desirable jobs— job protection, a guaranteed salary floor, working with the brightest minds, tremendous freedom to speak your mind, etc. Reasonable people can disagree— these jobs are still desirable— but it has become tarnished in a world of dwindling federal funds, cancel culture/ less academic freedom, and weak and numerous administrators.

Now, arguably, being a founder or first 20 in a successful company has dwarfed that. Twenty years ago it was unheard of for 10 young people to work on something for 2 years and be acquired for 60 billion in stock— and yet, now that occurs. In San Francisco, it doesn’t seem that odd. I know several people personally who went from net worth <1 mil to net worth ~1 bil in less than a decade. Many are linked to companies that are household names. This is truly astonishing.

Note: If you enjoy this substack— if you sometimes but not always nod— please become a subscriber— it’s strength comes from subscribers and nothing else.

Subscribe now

A core premise of DEI is to empower those who had not previously been empowered to obtain the positions in society of great influence, security and prestige. Yet, an irony is that MEI has come to dominate the burgeoning institutions that now house those positions. A pediatrician has always paid less than an orthopod, but was still far more than many other jobs, now even most entry level tech positions offer comp packaged 2 -3x as much— jobs in fintech or other more lucrative sections rival the salaries of the orthopod.

A decade ago, a colleague made the observation: too many past department chairs had been men. In an effort to combat this, the university worked to hire a woman— explicitly stating this was the goal (Never a good thing to say/ so undermining/ “my vice president will be a woman”). Then, a number of senior men in the department— arguably the runner(s) up to be chair— were given “new titles”— Endowed prof/ head of research/ etc. The Dept chair’s former power was dispersed. Thus DEI was just for show. The department chair retained only the worst parts of the job— scolding errant faculty. That was DEI 1.0.

2.0 took things more seriously, and was taken to extremes— the Jason Arday’s of the world were hired. Admissions policies openly flaunted the Supreme Court. But then, start ups started to boom. Biotech, and AI, and more, and we found that DEI carried out more faithfully dragged down institutions, with MEI used at those that rose.

My prediction is that no one has learned their lesson, and gravitating to a compromise is not in the nature of either political party. DEI will hide from the Trump DOJ, and come out like spring flowers when the next Democrat is hired. It will blossom and grow like never before, but then inevitably the retribution will be swift, and I suspect a future republican DOJ will remind universities why they should have made efforts to move to the center of the body politic. Just a prediction.

Ucsf Title Vi Violations

4.5MB ∙ PDF file

Download

Download

Obviously we don't facilitate it in medicine — by installing cameras and making it one click. God forbid, if there was a video recording of the doctor's visit, someone might be able to fact check the doctor's physical exam.

I've had patients ask me to audio record visits, and I've always said yes. But I've never known a hospital system to facilitate it/ make it easy. And certainly not video.

(What about concerns of privacy around the exam? For the sake of argument, I'm stipulating the patient wants it recorded nonetheless. Also it is rare these days for doctors to actually have the patient put the gown on, etc)

Enter the new battle

Do patients have a right to the audiotape recording if a doctor is using AI to generate a note (and if the recording is saved?)

Before I give you my answer. Let's think about body camera's worn by police officers. There are a number of studies, including some randomized trials. Researchers find that the use of body cameras reduces citizen complaints. It's unclear if it reduces police aggression. The footage has been useful in anecdotal cases. It is costly to collect. There are additional questions that are not fully answered because the sample size of the studies and scope is limited. Probably the vast majority of footage is not looked at.

With this in mind, medicine should embrace randomization when it comes to more accurate recordings and encounters. It is plausible that allowing patients to opt into a video taped encounter with a link provided to family to review later, will improve medication compliance, satisfaction, family member involvement, etc. It may even result in a reduction in level 5 billing, as doctors more accurately document their physical.

Just as citizen complaints go down, patient complaints may go down. Doctors may be less likely to sound impatient or be hurried or dismissive— the universal characteristics that annoy patients.

Malpractice is a disaster. Many doctors are sued who shouldn't be, and many cases of malpractice aren't brought. The patient may not even know something was done incorrectly.

It's not clear to me why investigating malpractice with the doctor's note is superior to having the raw audio tape of the encounter. Just like in cases of police shootings, everyone is interested in the body cam footage even if grainy and limited.

But just like body cams, we need to consider the broader effects of the policy. These effects may dwarf malpractice concerns.

The AMA filed an amicus brief to keep the AI recordings private and some companies auto delete them. If you are a patient tell me… why are you consenting to AI recording your encounter? What is in it for you?

I suspect there will be more court rulings and eventually patients will have a right to the raw audio just like they are free to review the raw radiology film and not just the note written by the radiologist. To combat this, more companies will discard the audio.

Eventually, someone may study the impact of audio and video recordings on patient satisfaction. Some have even been done.

Just remember. Survey data will be near useless (sorry Dartmouth). Single center studies will be useless. You really need multi-center, cluster randomization with adequate power to draw any real conclusions or to land up like a lot of the body cam studies.

In the meantime: audio recording is to AI note what radiology imaging is to radiologist note. I suspect legally it will be owned by the patient, and eventually hospitals may not be able to discard it.

Yes.

Can looking at secondary endpoints help you figure out if that's the case?

Again, the answer is yes.

But you have to be smart about it. You have to apply critical thinking skills to the question.

Consider two cases. Case 1:

You make a new cancer drug. You test against the best available care and the primary endpoint is PFS. Your response rate is 20 points higher than the control arm (stat sig). Your PFS is numerically 3 months longer but the p-value is 0.06. Your overall survival is trending favorably. Your drug has very little side effects, and is oral versus the IV control arm. Time to next treatment is prolonged, and quality of life is favorable.

Does your drug actually work? Yes, probably. Look at the story that's being told. A narrow miss on the primary. Multiple consistent secondaries. Why did you miss? Was the event rate in the control arm lower than anticipated? Is it possible there's a effect modifier, a biomarker? Totally worth exploring these ideas further.

Should you be required to replicate your finding? Use real world evidence to bolster your finding? This is a debate for another day. But reasonable people will agree you have a flash in the pan.

Case 2:

Now, what if your primary endpoint is Stone Cold negative? Not just a little negative but the hazard ratio is essentially 1.

Then you look at a secondary endpoint that is actually a sub-component of the primary endpoint. That's going the right direction— borderline stat sig.

But what is the shape of the curve? Is it consistent or does it wobble?

If the primary is Stone Cold negative and the secondary is erratic, could your treatment still work? Maybe, but the far more likely possibility is you have fooled yourself. You may even fool others too. But, as a wise man once said, you'll be unable to fool everyone.

Take a look at MCED.

The primary endpoint is 1.03 - stone cold negative.

It makes sense that a screening cancer blood test will initially have an excess of advanced disease, as you find stage 4 disease that is not otherwise clinically apparent.

It makes sense that there would be convergence in subsequent rounds.

But what doesn’t make sense is that the curves would get closer at 35 months, furthest at 39, closer at 44. And missing stage would be imbalanced between arms?

Also there is a difference between what readers think the figure means and what the figure means.

There are 2 ways to have metastatic cancer

You can be diagnosed initially with Stage IV disease. (A)

You could be diagnosed initially with I-III, and later get mets. (B)

Readers think the graph is showing the sum of these two things (A+B), but it is showing A only.

What K-M can help us?

A time to bone mets or mets graph. (B)

(This should show a separation like Panel D)

A death from cancer graph.

An all cause mortality graph.

Secondary endpoints can help tell a story, just like circumstantial evidence can convict a criminal— but only if the detective is impartial, unbiased, and skeptical— otherwise you might convict the wrong person.