Making Sense - Sam Harris - #485 — The New Science of Cancer

00:00:00.000 I am here with Siddhartha Mukherjee.

00:00:23.460 Sid, it's great to see you again.

00:00:24.980 Pleasure, man.

00:00:25.760 So we have a lot to talk about.

00:00:27.140 You have an updated version of your Pulitzer Prize winning book, The Emperor of All Maladies,

00:00:32.760 A Biography of Cancer, which came out 15 years ago, but you've updated it.

00:00:37.200 And I think there are four new chapters in the new paperback.

00:00:41.060 So I want to focus on that.

00:00:42.760 I want to spend some time on how our thinking about cancer has changed in the interim.

00:00:48.140 And I think we'll break this into three chapters, prevention, detection, and treatment slash

00:00:54.040 cures.

00:00:54.500 But also you have an AI startup, which I want to talk about because the utility of AI here in any one of these stages is obviously something that people are hoping for.

00:01:07.220 And I'm glad to see you're trying to push that forward.

00:01:11.080 But let's start with just kind of the basic conceptual framework and maybe how that's changed in the intervening years.

00:01:18.120 How should we think about cancer as a disease?

00:01:21.520 I mean, this is a, are there, is it a hundred different forms of disease?

00:01:26.160 Is it one?

00:01:26.940 I mean, when we get a cure for this thing, is it going to be one cure or are there going

00:01:30.300 to be hundreds, do you think?

00:01:31.720 Well, I'm almost certain that there'll be hundreds, but there'll be common themes running

00:01:35.520 through it.

00:01:35.980 So one thing, one question that I, you know, try to answer very often is exactly the question

00:01:41.420 you asked, which is, you know, is it a hundred different things?

00:01:43.760 Is it one disease?

00:01:44.520 Is it many diseases?

00:01:45.760 If it's many diseases, why do we call them cancer in the first place?

00:01:49.120 Why shouldn't we just separate all of them out?

00:01:50.900 And the answer is somewhere in the middle. It's every form of cancer. In fact, every individual

00:01:56.620 form of cancer, every individual specimen of cancer is its own disease in the genetic sense.

00:02:02.880 So in the sense that, you know, a one woman who walks into your clinic with, let's say,

00:02:06.780 breast cancer has a particular spectrum of mutations. Mutations are changes in DNA

00:02:11.260 that drive the cancer cell's growth. The second woman might come into your clinic with breast 1.00

00:02:16.080 cancer looks the same under a microscope. It's called breast cancer, but her spectrum of 1.00

00:02:21.040 mutations, you know, maybe she has 100, maybe she has 20, her spectrum of mutations is slightly

00:02:25.840 different. So why do we call them all of them cancer? Well, first of all, there are some broad

00:02:30.620 physiological commonalities. So the broad physiological commonality is that in all cases,

00:02:36.880 the first woman, the second woman, the third woman, all with breast cancer, in all three cases,

00:02:41.240 the problem is that the cells don't know how to stop dividing. And in a few cases,

00:02:46.340 they don't know how to stop living or essentially they don't know how to die. But let's say that

00:02:51.620 most of the most part, they don't know how to stop dividing. And driven by that malignant growth,

00:02:59.240 these cells, these cancer cells have started co-opting, hijacking, you might call it,

00:03:04.340 normal pathways that normal cells use to survive. So just like normal cells use nutrients to

00:03:10.020 survive. Cancer cells also need nutrients to survive. You could say they need kind of a special

00:03:15.800 kind of nutrient to survive, special kinds of nutrients to survive, special pathways that

00:03:20.700 they've hijacked from normal cells. Just like normal cells in the body move around and go to

00:03:25.460 other places, cancer cells also acquire the property to move around. So there are deep

00:03:30.520 commonalities that run between all these diseases called cancer. And yet it's also true that each

00:03:36.580 individual specimen of cancer in its own cancer. Is there one conceptual bottleneck here that most

00:03:43.880 troubles you in our making progress? I mean, is there one question that if we had the answer to

00:03:49.300 it, you think it would unlock the greatest promise here for treatment or prevention or

00:03:53.980 detection or all of it? Well, I think we should really speak about prevention, detection, and

00:03:58.840 treatment differently. Let's start with treatment. I mean, the big unlock for treatment is always

00:04:04.480 going to be, can we find something in the cancer cell that's different from the normal cell? That's

00:04:09.360 always been the problem. Cancer cells are very close cousins, if you will, to normal cells. And

00:04:14.760 that's obvious because they're derived from normal cells. So the big conceptual unlock here

00:04:22.600 is, can we find one pathway, two pathways, five pathways, 10 pathways that are different enough

00:04:29.700 between a cancer cell and a normal cell.

00:04:31.840 And by pathway, I mean a series of, 0.97

00:04:34.440 you can think of it as a kind of baton race

00:04:37.960 between one signal and another signal.

00:04:40.860 Ultimately, all the signals are going to the same place.

00:04:43.000 They're telling the cell, grow, grow, grow.

00:04:45.560 But these pathways are unique to cancer cells.

00:04:49.240 And the job, one of the big jobs in treatment

00:04:51.940 is to find the difference, the unlock, as it was it were,

00:04:55.680 is to find the difference between

00:04:57.560 what the cancer cell is able to do

00:04:59.240 is doing and what the normal cell is able to do and is doing. If you can find that unlock

00:05:03.900 across not one, but multiple specimens of cancer, we'll have different treatments.

00:05:08.340 There may be some common ones, there may be some different ones, but that's the big unlock there.

00:05:12.840 Okay. Well, let's go back to prevention because it seems like the right thing to put first here.

00:05:18.880 So we know that lifestyle and other variables can affect one's cancer risk significantly. I mean,

00:05:26.840 there's the environment, there's lifestyle, there's vaccines, right? We have vaccines for

00:05:31.600 certain preventable cancers. Why is, in your view, is prevention kind of an afterthought? I mean,

00:05:37.600 is this a science problem or an incentives problem? And why do we think about prevention last?

00:05:44.100 Well, we shouldn't be thinking about prevention last. And to be totally honest,

00:05:47.300 this has been known for a while that it should not be an afterthought.

00:05:50.900 The problem is that prevention science is probably the most difficult science,

00:05:55.820 because you're trying to do something and not have it happen.

00:06:00.600 You know, scientists are used to, heuristically,

00:06:03.920 you can use a fancy word, epistemologically,

00:06:07.000 scientists are used to watching things happen

00:06:09.540 and then stopping them from happening or starting them from happening.

00:06:13.000 In prevention, what you're trying to do is trying to create something

00:06:16.200 that does not happen.

00:06:18.160 And so prevention trials, to give you one example,

00:06:21.240 tend to be very long because, you know,

00:06:24.100 you're essentially giving normal people something or exposing normal people to something or changing

00:06:29.460 normal people's behavior and making sure that they don't get cancer as a result of that change.

00:06:34.820 And you can imagine, given us, if the incidence of cancer is relatively small, let's say it's,

00:06:40.760 you know, a hundred in every hundred thousand people, you can imagine that that trial stretches

00:06:45.040 on for 10 years or five years until you really understand how to prevent cancer. Now, you can

00:06:49.700 take shortcuts. You can take people with high risk disease or high risk for cancer, high genetic risk

00:06:55.420 for cancer, and then you can have a shortcut to getting a better study. But that's always been

00:07:00.100 one of the big questions in science. The other problem is that there is really no surrogate,

00:07:06.640 and I'll tell you what a surrogate is, but there's really no surrogate for the development of future

00:07:11.300 cancer. I'll contrast it with heart disease. The huge difference in heart disease is that in

00:07:17.060 cardiovascular disease and when you have heart attacks, myocardial infarctions, we discovered

00:07:22.100 that there were biomarkers for myocardial infarctions. So in other words, if you had

00:07:26.380 high cholesterol of the wrong kind, you would have a higher chance of getting a heart attack

00:07:32.880 in the future. So now you have a biological marker called a biomarker or a surrogate in which you

00:07:38.640 say, well, okay, instead of waiting for the heart attack to happen, if I can lower that bad

00:07:43.040 cholesterol, that's a good trial. I can prevent a heart attack from happening, and the end point

00:07:47.680 of the trial is I'm going to lower the cholesterol. Another example, hypertension. We know that high

00:07:52.380 blood pressure is related to having heart attacks in the future. I can say, okay, well, lowering

00:07:57.860 blood pressure, which I can measure, is going to prevent heart attacks in the future. Unfortunately,

00:08:02.780 there isn't something like that. There isn't a hypertension or a high cholesterol for cancer.

00:08:08.200 You have to actually, unfortunately, for most cancers, wait for the cancer to happen.

00:08:13.040 And that has been a very difficult bar because, obviously, these clinical trials, any methods,

00:08:18.420 any discovery methods go on forever.

00:08:21.460 But there are basically two very broad ways that people try to figure out how to prevent

00:08:30.160 cancer or what causes cancer and how to take them away from our environment.

00:08:34.140 So one way is to look for, since cancer is a disease of mutations, one way is to look

00:08:40.200 for things that cause mutations.

00:08:42.400 So, and that's, there's a test,

00:08:43.920 a classical test is called the Ames test

00:08:45.980 after a fellow named Bruce Ames who invented it.

00:08:48.980 And that's a test that's essentially

00:08:50.400 a mutation trapping test.

00:08:52.420 So it says, you know, x-rays cause mutations.

00:08:55.580 X-rays, if you expose it, the Ames test,

00:08:58.600 it'll cause, it'll catch x-rays as a carcinogen,

00:09:02.140 a cancer causing agent.

00:09:03.500 The other way is to do animal studies.

00:09:06.080 So you expose animals to whatever agent

00:09:09.860 that you're concerned about.

00:09:11.520 and you ask if animals get cancer. Now, obviously you can realize that there are some things that

00:09:16.660 you can't make a mouse smoke, for instance. So you have to find a way to paint the mouse with

00:09:23.740 tar to get the mouse to see if that causes cancer. And the third way is a large epidemiological

00:09:31.900 study. So in other words, you follow a large population of people and you could ask the

00:09:36.800 question, is there a higher rate of cancer among those people? For instance, is a higher rate of

00:09:41.500 of lung cancer and mesothelioma in people who work in asbestos factories. So you say, okay,

00:09:47.380 asbestos is a carcinogen. How can I prevent those mesotheliomas? I'm going to take asbestos out of

00:09:53.040 the environment. If the AIMS test, the one, the first I referred to, suggests that x-rays cause

00:09:59.440 cancer, how can I prevent cancer? I'm going to try to reduce your exposure to mutation-causing

00:10:06.480 x-rays. You have a substance that causes cancer in animals. How do I reduce cancer? How do I

00:10:12.720 prevent cancer? I'm going to take that away from exposure to humans. So those are the three broad

00:10:17.840 ways by which we can, and I've left out a couple, but those are the three very broad ways that one

00:10:23.200 can understand how to prevent cancer. But whatever happened to the cell phones

00:10:27.000 cause cancer story that hit the news about 20 years ago? This actually predates the smartphone.

00:10:32.300 I remember we all had our flip phones, and we were all terrified about stories of lateralized

00:10:37.780 brain tumors that seemed to be skyrocketing.

00:10:41.220 And then I think it's been decades since I've heard a story along those lines.

00:10:45.040 Do we all just have more brain tumors and just we're so attached to our smartphones

00:10:48.600 that we can't talk about them?

00:10:50.420 Or what's happened?

00:10:52.500 Quite the opposite.

00:10:53.180 So if you look at the incidence or if you look at mortality from glioblastomas or brain

00:10:59.340 tumors in the United States, it has remained flat over multiple decades. One can have lots

00:11:07.960 of arguments about some people who have fancy mechanisms by which they claim that cell phone

00:11:14.780 and cell phone radiation causes cancer. Just to be very clear, the radiation that is coming out

00:11:20.740 of your cell phone is completely different. Physics-wise, it's completely different from

00:11:26.380 the radiation that you get from x-rays, for instance. They're both called radiation because

00:11:30.940 ultimately they're forms of energy transferred through radiance, but they are completely

00:11:38.380 different. They are completely different in energy. They're completely different in their

00:11:41.560 properties. And so mechanistically did not make sense. And the ultimate proof of the pudding is

00:11:48.420 that the use of cell phone has skyrocketed in the world and in the United States and the mortality

00:11:56.240 from brain cancers has remained largely flat.

00:12:00.160 Well, that's one piece of good news

00:12:01.260 we can dispense to our audience here.

00:12:03.880 Yes.

00:12:04.340 What about chemo prevention?

00:12:06.080 I mean, are you anticipating a time

00:12:07.620 where we're going to take a pill

00:12:09.160 that substantially reduces our cancer risk

00:12:12.360 or are we nowhere near even thinking about that?

00:12:16.100 Actually, we're closer and closer,

00:12:18.020 maybe closer than many people think.

00:12:20.280 So let's take a step back

00:12:21.900 and let's air out some, I would say, some laundry

00:12:25.780 whether dirty or not, some laundry from the prevention world. So this fact often surprises

00:12:31.620 people. The surprising thing is until recently, and I'll talk about what recently means,

00:12:37.100 we really have not found a chemical carcinogen with large human impact, a preventable chemical

00:12:46.480 carcinogen with large enough human impact to make a real difference in cancer prevention

00:12:52.800 since the 1960s. So just take a minute to swallow that fact. Billions of dollars have been poured

00:12:59.360 into prevention research. And certainly we found chemicals that cause cancer that can be removed

00:13:05.800 from certain environments. I'll give you a couple of examples. I'll give you one already, asbestos.

00:13:10.000 I'll give you another example, formaldehyde. But usually these are in niche populations,

00:13:16.500 and they're in populations where asbestos workers, woodworkers exposed to formaldehyde.

00:13:22.080 So there really hasn't been an absolute revolution in which I can say, here is a chemical widely present that you are exposed to and I'm exposed to, which increases the risk of cancer substantially.

00:13:36.820 That is changing. I'll tell you about the change in a second. But before I do that, you could ask the question, well, why not? Why haven't we found them?

00:13:43.200 Well, the answers could be many. Number one is that it could be that there aren't so many.

00:13:49.740 That would be a difficult answer for us to swallow because we all want to prevent cancer.

00:13:53.720 Number two, we don't have the right methods to look for them.

00:13:56.420 You know, the tests I told you about, the Ames test and the mouse animal tests and the

00:14:02.340 epidemiological studies just aren't strong enough to find these kinds of carcinogens,

00:14:07.540 or maybe we need a different kind of test to trap these kinds of carcinogens.

00:14:11.040 And then, you know, it's also possible that it's a death by a thousand cuts problem.

00:14:16.100 So they do exist, but they just sort of fly under the radar of all these tests.

00:14:22.400 And it's the combination of them, somehow or the other, that's causing cancer.

00:14:26.360 And finally, the one thing I said, just to remember, I made an important caveat.

00:14:30.540 I said, chemically preventable carcinogens, we have discovered since that time, since

00:14:35.480 the 1960s, viruses that cause cancer.

00:14:38.420 A great example would be human papillomavirus, and there's a great vaccine against it.

00:14:42.380 So that falls in the viral category.

00:14:46.100 But just to remind you of what I said, since the 1960s or 70s, we have not found a preventable

00:14:52.160 chemical carcinogen of significant magnitude to make a difference in human cancer mortality.

00:14:58.540 So that would be a sad statement if I were to continue that line of thought, but that's

00:15:03.260 changing.

00:15:04.140 And that's changing because we've discovered recently, we've begun to discover a new class

00:15:09.180 of chemical carcinogens.

00:15:11.220 And this class of chemical carcinogens will not be caught by the Ames test.

00:15:16.100 This class of chemical carcinogens is unlikely to be caught by animal tests because the reason

00:15:22.460 is that it doesn't cause mutations.

00:15:24.660 What this class of carcinogens is, is it changes, if you think of cancer as a seed and its environment

00:15:32.380 as a soil, it changes the soil.

00:15:35.300 It doesn't change the seed as much.

00:15:37.060 It changes the soil around the cancer and thereby enables the cancer cells that were

00:15:42.140 previously dormant or asleep, it encourages them to start growing. And there's been a recent spate

00:15:49.320 of studies, most importantly, a study around particulate air pollution. So very small particles

00:15:55.400 of air pollution, which are now coming out to be a preventable human carcinogen because you can

00:16:00.940 remove the air pollution. And the way that particulate air pollution seems to work is not

00:16:06.440 the way that we think standard carcinogens work. So it does not cause mutations in cancer cells

00:16:12.800 and thereby unleash cancer by causing mutations like, I told you before, like x-rays or potentially

00:16:20.440 formaldehyde, but it rather changes the soil around the cancer cell and thereby unleashes

00:16:26.780 the growth of a previously dormant cancer cell and makes the tumor grow. And particularly air

00:16:32.300 pollution is an example of this, there's a strong suspicion that asbestos is an example of this.

00:16:37.140 For a long time, we didn't know why asbestos, even though it was a very potent carcinogen,

00:16:42.860 we didn't know why asbestos caused cancer. We think that this is potentially how asbestos

00:16:47.740 causes cancer. So that leaves the question of what exactly is happening? What is it doing?

00:16:52.800 And the answer is that these new substances, in my book, I call them inflammogens,

00:17:00.000 These new substances cause a particular form of inflammation, not any kind of inflammation,

00:17:05.440 but a very particular form of chronic inflammation.

00:17:09.120 And cancer cells love to breed and grow in the soil of that chronic inflammation.

00:17:16.300 And that leads to two very important consequences.

00:17:19.320 Number one is that if we could find a measure of chronic inflammation, this particular kind

00:17:26.120 of chronic inflammation, we may be getting closer and closer to finding that magic thing that I

00:17:31.480 talked about before, which is a biomarker for future cancer. So that's one thing. And secondly,

00:17:37.620 of course, we can devise tests. Just like we devised a test for capturing x-rays and other

00:17:43.780 things that cause mutations, so-called mutagens, we could devise a test for inflammogens. And these

00:17:51.360 inflamogens are potentially things that we could remove from our environment, remove from our 1.00

00:17:56.140 bodies, remove from our body, physiological states in our bodies, and thereby reduce the risk of 0.98

00:18:01.400 cancer. So it'd be a new way of thinking about prevention. And this would actually be a very

00:18:06.360 revolutionary way of thinking about prevention. What about anti-inflammatory drugs that, I mean,

00:18:12.160 whether you can control the variables in the environment or not, what about just bringing down

00:18:16.740 inflammation generally in the body?

00:18:19.660 So it's, again, you said the word generally, and it's the generally part that doesn't work.

00:18:24.480 This is the very specific kind of inflammation. It's mediated by a particular kind of cell in

00:18:30.340 the body called a macrophage. A macrophage is named because it's a, a macrophage really means

00:18:36.440 big eater. It's a cell that goes around the body sort of scavenging all sorts of things like dust

00:18:42.640 particles, I should say, particles of pollution. Macrophages, you know, often have, in asbestos

00:18:48.880 workers, you can see them sort of trying to eat the asbestos, the tiny, tiny needles of asbestos.

00:18:54.620 So it's a very particular kind of inflammation. And yes, if we could find a way to find people

00:19:00.760 who had exposure to that kind of inflammation or had high levels of that kind of inflammation

00:19:05.840 and potentially prevent that inflammation in those people, yes, there would be a chemoprevention.

00:19:11.180 Now, one last important point here.

00:19:13.540 There is a very good chemoprevention for a very particular kind of cancer that works

00:19:17.860 very well, which is estrogen receptor-driven breast cancer.

00:19:22.200 So we know that breast cancer, we've known this for a long time, breast cancer is some

00:19:27.080 kinds of breast cancer, not all kinds of breast cancer, but so-called ER positive.

00:19:32.260 So estrogen receptor-positive breast cancer thrives on estrogen.

00:19:36.640 Estrogen is a natural hormone made by the body,

00:19:38.940 but you can give drugs that modulate or modify the cancer cell's response to estrogen,

00:19:46.420 and those are very good chemoprevention for patients who are at high risk for breast cancer.

00:19:52.080 We don't give them to everyone because they have significant side effects.

00:19:55.860 A good example of such a drug is tamoxifen.

00:19:58.360 We don't give it to everyone, but for patients who are very high risk for breast cancer,

00:20:01.960 There have been several studies now that show that if you give them these anti-estrogenic

00:20:06.420 pills, they have to deal with many side effects.

00:20:08.840 But if you give them these anti-estrogenic pills, you can actually have chemo prevention

00:20:12.540 for breast cancer.

00:20:13.860 So here's a final question on the prevention topic.

00:20:17.600 How should people think about their risk when they have some information about it?

00:20:23.680 It might be family history or a polygenic score or some other information that they

00:20:29.260 have, which makes them feel like, you know, I think in this case, accurately, that they have

00:20:35.100 more than a normal risk for a certain kind of cancer. How do you recommend people process that

00:20:41.540 information personally without, you know, falling into fatalism or despair or some form of panic?

00:20:47.600 I mean, you as an oncologist, how do you walk people through that?

00:20:50.960 So first of all, I think, you know, I've written a lot about this. I think, you know,

00:20:54.240 there's a whole phenomenon. It's a chilling Kafkaesque word called previvor, which has

00:21:00.300 entered the vocabulary of cancer. And a previvor is sort of derived from the word survivor,

00:21:06.800 but a previvor is a person who thinks they're going to get cancer, but they don't have it yet.

00:21:11.580 They're driven by the anxiety and the fear that they're going to get cancer, but they've not had

00:21:17.240 it yet. They're not a survivor of a cancer. They're a previvor of a cancer. And the number

00:21:22.180 of previvors is increasing dramatically in the world because all sorts of tests surround all

00:21:27.240 sorts of genetic tests and other tests surround and sowing a lot of fear in people's eyes and

00:21:33.600 brains. The way I advise people to think about this is to really have a, even if it's a grayscale

00:21:39.800 quantification, a grayscale understanding of their risk. And by grayscale, I mean there are some

00:21:46.380 people where their genetics and their family history is very strong. So a great example would

00:21:54.000 be patients with the BRCA1 gene or the BRCA2 gene. Those patients have, or those people,

00:22:00.080 before they become patients, they have a very high risk of getting, for instance, breast cancer.

00:22:04.480 It's especially higher if in the context of when they have a very strong positive family history

00:22:09.720 of breast cancer. So those patients, I advise going to a genetic counselor and seeing the

00:22:14.840 genetic counselor and seeing if they should enroll in one of many trials that are now available

00:22:19.400 to screen them more effectively, potentially to put them on, you talk a little bit about

00:22:25.260 chemoprevention, or potentially to put them on a trial for a novel chemoprevention for those

00:22:31.560 cancers. So that's the advice I give those people. Then there's a second, again, moving along the

00:22:36.360 grayscale, there's a second category of person who has what you call a polygenic risk score.

00:22:41.280 Now, polygenic risk score, let's unpack that word.

00:22:44.460 So polygenic risk score is a person who doesn't have one of these genes like BRCA1 or P53

00:22:51.180 mutations or one of these inherited mutations in genes where their risk of getting cancer

00:22:57.060 is ginormous because the genes are mutated, but they've inherited a mutated gene from

00:23:02.680 their parents.

00:23:05.060 Polygenic risk scores, you can think of them as, if you think of the BRCA1 gene as a

00:23:10.900 something that shoves you towards getting cancer. I apologize using that analogy or metaphor.

00:23:16.540 These are genes that nudge you little bit by little bit towards higher and higher cancer risk.

00:23:22.120 They're quantifiable. So you can quantify them. If you sequence a genome, you can quantify them.

00:23:27.720 And patients with very high polygenic risk score, for instance, for breast or ovarian cancer,

00:23:32.900 I usually assuage them. I generally tell them that these polygenic risk scores

00:23:36.800 are still early in their study. And I tell them to see a genetic counselor, but not to be as

00:23:43.060 worried as patients with, for instance, the BRCA1 or the P53 mutation. And then there are

00:23:47.860 patients who have no family history, no polygenic risk score, no real risk upfront of getting

00:23:53.700 cancer, but they are still worried. And I say to them, well, it's a risk that you have to take

00:23:58.160 with aging. Cancer is a disease of aging. We all are at risk. And if you feel that you have

00:24:05.780 a particular exposure in your childhood. For instance, your father was an asbestos worker.

00:24:12.800 Those are the patients that I send for deeper testing, genetic counseling, etc. But in those

00:24:18.080 patients, we're really a little bit stuck in some ways. The last one, the last category I've left

00:24:25.100 always separate. I always leave it separate because it's a very unique category. And that is

00:24:29.240 if you're infected with a virus that causes cancer. And a great example of that is human

00:24:35.580 papillomavirus. So if you have a human papillomavirus infection, you are indeed at a higher

00:24:41.500 risk to get cervical cancer, depending on the strain of human papillomavirus. Not all strains

00:24:46.040 cause this, but some strains, we call them the teen strains, increase the risk of human papillomavirus.

00:24:53.880 If you have that strain, if you're infected with human papillomavirus, then you should certainly

00:24:59.260 be seeing a gynecologist who should follow you to make sure that your risk of cervical cancer

00:25:04.420 is decreased and they may have to do a biopsy or even potentially invasive surgery to decrease that

00:25:09.880 risk. Which brings me to a side point, which is that there are incredibly effective vaccines

00:25:16.560 against human papillomavirus. And please don't believe the nonsense that's been perpetrated

00:25:24.100 about vaccines against cancer. These are extremely effective. I believe that both men and women,

00:25:30.460 young men and young women, young boys and young girls should get these vaccines.

00:25:34.420 And a massive study in Sweden showed that if, and this was a randomized controlled study,

00:25:40.360 the hardest, most rigorous kind of study that exists, a massive study in Sweden showed that

00:25:46.520 if you gave the appropriate age, the appropriate number of vaccines for human papillomavirus,

00:25:53.280 the dangerous strains, the risk of getting cervical cancer in adulthood goes to zero,

00:25:59.560 zero so so we are committing a terrible i would say you know it's a terrible tragedy that across

00:26:08.500 the the global world there are still women dying of cervical cancer caused by papillomavirus this

00:26:14.780 is a completely preventable cancer well when you qualify it saying young men and young women is

00:26:21.420 that just a matter of uh kind of population level um triaging of resources or you actually think

00:26:29.060 the utility of being vaccinated goes way down as people age. Well, this is a very particular

00:26:36.620 situation. So human papillomavirus is a sexually transmitted disease. And obviously, young men and

00:26:43.000 young women are the most at risk because they are the most likely to have an infected partner

00:26:50.480 or have multiple partners, one of whom carries an infection. So it's just a consequence of human

00:26:55.520 behavior. It's a behavioral risk. But if someone is 40 years old and they don't show any titers for

00:27:01.260 HPV and they're, you know, single and sexually active, is there any reason why you wouldn't

00:27:07.360 recommend that they get vaccinated for it? So there's no reason that they wouldn't get

00:27:11.860 vaccinated for it, except that those populations have not been studied because the studies have

00:27:17.220 been done in young people. But biologically or physiologically, there's no fundamental reason

00:27:23.320 that if they were negative for human papillomavirus to start with, that they would not

00:27:28.100 respond to the vaccine. There's no biological reason to think that all of a sudden their

00:27:31.640 immune system will flicker off and not be able to drive a response against human papillomavirus.

00:27:36.000 Right. Well, we won't ask RFK Jr. his advice on this topic. We might get to our political

00:27:43.200 moment eventually. So let's talk about detection. So there's been a lot of excitement around

00:27:49.340 so-called liquid biopsies, you know, blood tests that detect so-called cell-free DNA that I think

00:27:55.540 they're up to, you know, 50-some-odd cancers, specific cancers they claim to detect. I have a

00:28:01.000 little bit of experience with this. I have taken a couple of these tests, one of which was positive.

00:28:07.380 It turned out to be a false positive, but I, you know, then did subsequent scanning and, you know,

00:28:12.820 spent a week imagining that I had a greater than 50% chance of having, you know, one or two or

00:28:18.340 three different cancers. So I've experienced some of the downside of this. Give me your thoughts

00:28:24.080 on around the risks of moving too fast on this and where you think it is headed and what will be

00:28:30.060 the stable point if it's achieved where we're not continually running the risk of over-treatment

00:28:38.120 and, you know, kind of painful encounters with misinformation. Well, so people think that your

00:28:44.800 case is, your anecdote is atypical, but in fact, it is the most typical. So the most typical

00:28:51.260 anecdote, which is not being publicized, there are a thousand companies that do work on cell-free

00:28:57.000 DNA. And I'll try to distinguish the ones that are doing actually good work, but your case is

00:29:01.760 actually typical. But to understand why it's typical, you need to understand something about

00:29:07.020 nothing. It has nothing to do with cancer, but it has to do with mathematics. And this is pure

00:29:11.140 mathematics. I wrote a piece in the New Yorker. I got all sorts of hate mail for it. But the

00:29:15.540 problem is that you can't argue against pure math. Math is math. And the math is very simple

00:29:21.280 in this case. And the math, I won't give you the formula, but it's based on the observation

00:29:26.460 of a very important man whose work has inspired computer science and pure mathematics and

00:29:34.760 statistics, and that's Thomas Bayes. So Thomas Bayes lived in Greater England, and he made a

00:29:42.800 very simple statement, which he then made into a mathematical formula. And the simple statement is

00:29:48.240 that what he called the posterior, what was later called the posterior probability that you have

00:29:55.980 cancer. In other words, whether you do have cancer or not, whether in a population, if I'm measuring,

00:30:01.580 if I have a test that's measuring whether someone has cancer or not, depends on the prior probability

00:30:08.680 that there's cancer or not. So how do we explain this? Here's a simple analogy.

00:30:13.420 Let's say you make a genius detector, a needle detector, and you're looking for a needle in a

00:30:18.340 haystack. So you have a massive haystack and there's one needle buried in it. And you have

00:30:23.320 a good detector. It's 90% sensitive and 90% specific. In other words, it means that 90%

00:30:30.800 of the time when it says it's got something, it actually turns out to be a needle. You go into

00:30:37.880 the haystack and you start, the detector beeps, and you find that it's actually not a needle,

00:30:43.400 but actually a piece of hay. You go into the haystack again, it beeps, and it's another

00:30:47.780 piece of hay. And the third time, and the fourth time, and the fifth time, and that's because

00:30:51.500 there's only one needle in a massive haystack. The prior probability, this haystack was stacked,

00:30:57.320 as it were, with stacked poorly against you. And no matter how good your detector is,

00:31:03.740 no matter how smart your instrument is, it's always going to detect more hay than needles.

00:31:09.480 Does that make sense? It should be very obvious. Yeah, yeah. I recommend that people take,

00:31:14.060 we can't do it here, but take a little time to understand Bayes' theorem and Bayesian reasoning.

00:31:19.340 But I mean, the background frequency of the thing you're trying to detect obviously changes the

00:31:26.600 likelihood that you produced a, however valid the test, you've produced a real positive as opposed

00:31:34.180 to a false positive. So just to finish up, yeah, go ahead. Yeah, but in this case, like we have a

00:31:39.320 company that is advertising its false positive rate, you know, its type one error rate of one

00:31:44.920 in 200, right? So they put the, you know, their Bayesian reasoning, you know, given the incidence

00:31:50.120 of cancer and the cancers they're trying to detect, they're advertising their false positive

00:31:56.220 rate as, you know, half of a percent, and therefore you as a consumer get a positive

00:32:01.980 finding and you think, well, okay, there's a 1 in 200 chance this is wrong, but I don't

00:32:09.620 find that very consoling because the report is telling me I now have a 57% chance of having,

00:32:14.800 you know, either, I think in my case it was, you know, kidney and bladder cancer or prostate

00:32:18.860 cancer or both. And so presumably we're going to get to a place where we're going to find that

00:32:27.140 the error rate is low enough so that, I mean, obviously we have to live with some false positive

00:32:34.000 rate and also there's also the false negative rate, which is real cancers that are undetected.

00:32:38.760 The fundamental mistake that we're making here is the one you just actually,

00:32:42.280 just, you exactly enunciated what the mistake was. The fundamental mistake is that most of

00:32:49.100 these companies are advertising their sensitivity and specificity. So in other words, they're saying

00:32:54.500 our test is really sensitive and it's really specific. What they're not telling you is what

00:32:58.880 Bayes would call the prior probability. The prior probability, the base rate of cancer

00:33:04.680 is low. And until no test, I suppose you can make a test that's a hundred percent

00:33:12.280 specific and 100% sensitive, but that's sort of an impossibility at this point in time.

00:33:17.440 But no test will ever change the prior probability because the prior probability of something is a

00:33:24.000 given. It's how much cancer is there in a population. That's a fixed number. So the

00:33:30.720 answer, I'm going to twist this around and give you a positive answer to the question.

00:33:36.140 If these companies, many of these companies were less greedy and if they were less driven by

00:33:41.340 trying to screen everyone and make money out of everyone and put anxiety into everyone,

00:33:46.700 this is actually the basis of my long piece in The New Yorker, and it's actually in the book as

00:33:50.520 well. If they were less consumed by consumerism, which is to say, I'm going to use this for

00:33:58.400 everybody, then they would identify patients who are truly at a higher risk for cancer. So in other

00:34:04.580 words, they would find populations where the base rate was higher. And sure enough, in those

00:34:10.300 populations, I'm absolutely confident that tests like cell-free DNA will be helpful.

00:34:15.720 So who are these people? Who are these people who have higher base rates of cancer? Well,

00:34:20.640 we talked about some of them already. If you have a mutation that is likely to cause a higher risk

00:34:26.460 of cancer, possibly, I'm not sure about it, but possibly if you have a polygenic risk score,

00:34:31.260 so the NUD genes that I talked about, which increase the risk of cancer, fine, you take

00:34:35.880 those people. You could take people who have had prior cancer before and ask the question,

00:34:40.940 is that cancer relapsing? That's another population where the base rate is higher.

00:34:46.040 And in all those cases, if the trials had been done with patients with all those cases,

00:34:51.060 then the chances of detecting a stage one or stage two cancer, something that actually you

00:34:55.940 can do something about, would have been much higher and is much higher in the small numbers

00:35:00.280 of trials that have done this. So that's the answer. It's a very simple answer. Thomas Bays

00:35:04.720 knew the answer 200 odd years ago. And it's amazing to me that in 2026, we're having a

00:35:11.940 conversation, not you and me, but the global public is having this anxiety-ridden conversation

00:35:16.820 about, oh my God, should I not test or should I test? The answer is, well, what is your prior

00:35:22.040 probability? What do you think? What has moved the needle? Where are you on the grayscale?

00:35:26.660 If you think that you're farther on the grayscale, your father had prostate cancer,

00:35:30.860 your grandfather had prostate cancer, you're worried. Yes, a self-free DNA test might be

00:35:35.780 useful. If you're just someone... But presumably some reduction in the rate of type 1 errors,

00:35:43.380 false positive errors, you could bring it so low that you wouldn't feel that you had to assess

00:35:49.700 your prior probability by being part of some special population of heightened risk. You'd say,

00:35:54.520 I'm a homo sapiens. There's some rate of cancer out there. And if they're giving me a 1 in 500,000

00:36:03.040 false positive rate, that's very different than 1 in 200. And I could do the calculation.

00:36:10.420 Fair enough. But actually, if you do the calculations, again, I would encourage people

00:36:15.220 to just... You don't even need to do the calculation yourself. You can go into Google

00:36:19.500 Gemini and ask Google Gemini to do the calculation for you, but you don't need to do the calculation.

00:36:25.020 But yes, absolutely right. At some point of time when the rate of type one error would be reduced,

00:36:31.660 yes, that test becomes relevant. The problem here is that the ultimate positive predictive value,

00:36:40.300 the number you're really looking for is if the test is positive, what are the chances that I do

00:36:44.700 you have stage one or stage two cancer, right? That is the ultimate, that's the answer you're

00:36:50.000 looking for. So again, to repeat the answer, if the test is positive, what are the chances that

00:36:55.000 you have a stage one or stage two cancer, not stage three, not stage four, but stage one or

00:36:59.560 stage two cancer for which I can actually do something? That number is highly, highly dominated

00:37:06.880 by the prior probability. So even if you increase or decrease the type one error,

00:37:12.740 that number will continue to be dominated by prior probability. And yes, of course,

00:37:16.960 in the envelope of time, in the envelope of things, if you decrease the type 1 error, yes,

00:37:22.400 you're going to start getting a situation where the test is worthwhile doing. For stage 1 and

00:37:27.160 stage 2 cancer, we're far from that yet. Along these lines, obviously different

00:37:31.260 technology, different, maybe every relevant way, but how do you feel about whole body MRI scans as

00:37:38.180 a prevention technology or detection technology. Basically, almost same story, except unfortunately,

00:37:44.740 I feel that the rate of what you call the type one error, or in other words, something is found,

00:37:51.500 but it's not really cancer. That kind of error is even higher. So I don't see that moving in the

00:37:58.440 right direction. I do... But Sid, I think that I completely understand the liability there,

00:38:04.780 but that seems to apply to a first scan. Yes, I was just going to come to that.

00:38:09.960 But if you've had a first scan as your baseline scan, then every subsequent scan is a measure of

00:38:15.660 change against that first scan. So I was just going to come to that. So there's a temporal

00:38:19.100 quality to this, which is what you're talking about. And that we have, to be totally fair,

00:38:25.980 we have not fully tested yet. So right now where we are is we're testing one scan at a time and

00:38:33.960 you know, whether you get a stage one cancer detected or not. It is probably fair that the

00:38:40.260 type one error reduces over the temporal axis over time and potentially reduces to a point

00:38:48.660 of time or to a point or a number where it actually is worthwhile potentially doing more

00:38:54.160 invasive tests like a biopsy or another kind of test. Let me just clarify. I want to make sure

00:39:01.020 everyone understands the distinction we're making here. So with a first scan, the problem with

00:39:04.760 getting your first full-body MRI, let's say, you know, you're a 50-year-old man and you're worried

00:39:10.280 about cancer and someone, you know, your doctor has advertised to you the possibility of getting

00:39:14.420 a full-body MRI. This is, you know, there's no ionizing radiation. It's totally safe. Why not

00:39:20.220 do it? It's $2,000, but it, you know, you get every voxel of your body scanned in an hour

00:39:26.440 looking for tumors. The problem with the first scan is that if you see something, you don't

00:39:32.400 know whether it's been there for 30 years and it's nothing or whether it's a quickly growing

00:39:37.500 cancer. And the prospect of being led on a wild goose chase that entails, you know, biopsies of

00:39:44.840 organs or more invasive scanning, all of that is a clear liability here. And the question I've

00:39:52.500 just asked you, Sid, is yes, but you price all that in, you get your first scan, it's clear.

00:39:57.440 Now your second scan seems to promise some much more valid information wherein anything that has

00:40:03.920 suddenly emerged in your liver or lung or anywhere else suddenly seems like this is new and worth

00:40:10.260 checking out. Well, let me challenge you back with two scenarios which may complicate that

00:40:16.660 answer a little bit more. First of all, let's say your first scan actually does find something.

00:40:23.320 It's a spot. Actually, Dhruv Kullar wrote a nice piece on this in the New Yorker,

00:40:27.040 if anyone's interested in reading further about this. But anyway, I think the company in that

00:40:32.380 case was called Prenovo. There are many out there. Anyway, your first scan, let's say it

00:40:37.420 actually does show something. The question you want to ask yourself is how many people are

00:40:41.060 totally comfortable sitting and waiting for their next scan at, let's say, six months from that time

00:40:47.540 and not doing a biopsy. And if you ask people, I see patients in real time, I see real people in

00:40:53.600 real time, the number, you'll be surprised. No one wants to sit and wait. Wait and see what

00:40:59.380 happens and wait and see it grows. That number is very small. So already you're committing a kind of,

00:41:04.840 you know, you're pushing people down the pathway of invasive tests, biopsies, and so forth. But

00:41:09.600 fair enough some people might say okay you know the first scan has is has shown a spot i want to

00:41:14.680 see if that spot is really growing or not if it's going at what speed and you know if it's cancerous

00:41:19.240 or not fine the second thing i would say about this is is is a point that is often missed but

00:41:26.620 is is is very important so i'm going to try to say it a little slowly but try to make people make

00:41:32.280 sure people understand when you have so let's say we decide that this these tests are a full body

00:41:39.100 scan is a useful test, or even cell-free DNA is a useful test. Let's say we decide that.

00:41:44.000 So then the question becomes, well, how do you judge whether it's really useful or not?

00:41:48.580 And someone's answer, not your answer, but someone's answer is going to be, well,

00:41:52.480 we should just measure survival. How long has someone survived once their scan has detected

00:41:57.680 something positive? But that's the wrong answer because it's a classic pitfall or a bias in

00:42:03.540 statistics called lead time bias. And in other words, what you've done is the person who didn't

00:42:08.160 get scanned, may also have had a cancer, but because they didn't get scanned, we don't know

00:42:12.640 when they get the cancer. Whereas your clock starts ticking the moment you get the scan.

00:42:17.480 So if your clock says that you lived three years after the scan and someone else who didn't get

00:42:22.920 scanned dies at the same moment, you'll think that, oh, you lived three years, that person

00:42:27.520 lived shorter times because their cancer was detected much later, so-called lead time bias.

00:42:33.880 you'll say, oh God, this test is wonderful. But in fact, that's not true. It's just a biased test.

00:42:38.680 So what you need to measure, if you really want to measure, is mortality. And measuring mortality,

00:42:44.900 just again, numbers, pure math, measuring mortality is hard because people die at a

00:42:50.300 certain base rate. And so you have to have a massive number of people in your trial to measure

00:42:54.500 mortality. So those are the two caveats. So if you were to tell me that people are comfortable

00:43:00.260 with having a spot in their bodies, wherever it might be, a lung, a prostate, liver, et cetera,

00:43:06.100 if they're comfortable getting repeat scans without biopsies, and if you tell me that there's a trial

00:43:12.740 that shows that invading on those growing things, whatever they were, actually decreased mortality,

00:43:19.940 I would say yes. But those are very high bars. And you say that research hasn't been done,

00:43:24.780 right? And also there's lots of confounds here. Anyone who's getting a full body MRI at this

00:43:30.160 point is obviously in a very specific population and it'll be hard longitudinally, it'll be hard

00:43:35.320 to separate all of that. I mean, they're doing all kinds of other things. Those studies haven't

00:43:38.740 been done. You know, those studies will probably never be done. So again, what do I, what is it,

00:43:43.780 how does it translate into actual advice along very much the lines of what you're saying? I,

00:43:49.560 you know, obviously if you're at a higher risk, I told, I talked about the grayscale of risk,

00:43:53.960 you know, I say, fine, go ahead and get your scan. And those would be things like family history,

00:43:58.000 exposure history, some particular reason that you suspect that you have a higher risk of getting

00:44:05.040 cancer. Secondly, I almost certainly advise people to do, if they're going to do a scan,

00:44:10.800 I advise them, even if they have a positive somewhere or the other, I advise them to get

00:44:14.860 an orthogonal test. By an orthogonal test, I say to them, well, okay, you've gotten the scan,

00:44:20.180 you've gotten this, let's try to see if you're also positive, if you also pick it up, for instance,

00:44:25.980 with the cell-free DNA, because two completely different tests are unlikely to have the same

00:44:31.340 type 1 error, obviously. Then, if that's still not satisfactory, I say, well, let's get at least

00:44:37.620 another scan six months later to see. And the number of people who balk at that is enormous.

00:44:43.020 People will say, no, no, I just want to get tested. And I just remind them that study after study

00:44:48.180 after study has shown that invasive tests, now if it was a superficial thing, like someone found a

00:44:54.380 spot in their skin and it's a simple skin biopsy, fine, I'll say, yes, fair enough,

00:44:59.300 go and do a simple skin biopsy. But if it's deep in the liver or it's somewhere in the lung and

00:45:03.340 there's a chance of puncturing the lung or bleeding out from the liver, I'll say, well,

00:45:08.200 there are real risks here. Do you want to really take the risks? I quantify those risks and then

00:45:13.380 give them all the information and ultimately, of course, let them make the decision themselves.

00:45:17.200 Right. Well, what do we actually know about dormancy or kind of the minimal residual disease

00:45:24.920 of somebody who's had cancer and is in something like remission? How close are we detecting

00:45:30.460 those states reliably? And how do you think about that in this picture of having or not having

00:45:36.560 cancer? So that's a very good question. So you've pinpointed the right population now. So this is

00:45:43.080 the population that I'm most interested in, I think most serious cancer biologists are

00:45:48.680 most interested in, which is you've had cancer, you've got into remission with first-line

00:45:53.620 therapy, and now we know that your cancer had some suggestion or there is a general

00:45:59.180 suggestion from the population that your cancer or your type of cancer is likely to relapse.

00:46:04.200 Can we monitor you for potentially what you're calling minimal residual disease?

00:46:08.720 By minimal residual disease, it means by all visible tests, you don't seem to have cancer,

00:46:14.680 you know, MRIs, x-rays, and whatever tests.

00:46:18.440 But in fact, there is some cancer lurking in your body.

00:46:20.880 We may not know exactly where it is.

00:46:22.640 We may not know if it's growing out in the same site where it was originally found.

00:46:26.860 We don't know if it's going out elsewhere.

00:46:28.940 The most important thing about minimal residual disease is to think about it as a tool, not

00:46:33.860 an alarm.

00:46:34.520 By tool, I mean, we now are using minimal residual disease to see if you can use early

00:46:42.480 treatment, once minimal residual disease has been detected, to use early treatment in a

00:46:47.740 population that deserves early treatment.

00:46:49.580 In other words, let's say, you know, Jim and Tim both unfortunately develop myeloma, both

00:46:55.500 go into remission after their first therapy.

00:46:58.880 These therapies obviously all have liabilities.

00:47:00.940 These are chemotherapies.

00:47:02.160 They may have side effects.

00:47:03.440 So we stop the chemotherapy. We say, you've finished with that. And we watch. And Jim does not develop minimal residual disease. In other words, let's say his cell-free DNA comes back over and over again, and there's no sign of recurrent myeloma in Jim. Tim, on the other hand, six months later, starts to have a little blip of cell-free DNA that shows the recurrence or the presence of myeloma.

00:47:28.460 So again, remember Bayes, our old friend, what we've just done is we've shifted the

00:47:34.780 Bayesian probability, prior probability that that blip that was found in this unfortunate

00:47:41.080 fellow Jim is actually recurrent myeloma.

00:47:44.440 And what we use that for is, we can use that for is, is we can use that as a biomarker

00:47:50.240 for the recurrence of myeloma.

00:47:52.760 And we can use that for testing new therapies or potentially tried and tested therapies

00:47:58.280 now in an early setting. And that has proved to be a very good strategy. In fact, myeloma is a

00:48:04.540 disease where this has actually proved to be a particularly good strategy. And the reason behind

00:48:10.420 all of this is that minimal residual disease picks up very few cells. The chances that those cells

00:48:15.760 will acquire or have acquired resistance to second-line therapies is therefore fewer, and

00:48:21.400 therefore the chances of curing the cancer or beating the cancer completely are higher. So that

00:48:26.800 is the setting. That is exactly the setting where I do use prevention, and that's a very good

00:48:32.380 setting to use preventative therapies in. All right. Well, let's talk about treatment and

00:48:36.900 cure. Is there anything in recent years, let's say, since you wrote the first edition of your

00:48:43.640 book, where a cancer has moved from being very high mortality to effectively being cured? Has

00:48:53.440 Has there been a radical breakthrough in the last 15 years for any specific cancers?

00:48:57.420 There's several radical breakthroughs for several cancers.

00:49:00.220 So people often say, oh, you know, let's take a great global view.

00:49:03.680 The very global view is people have a very dismal view of many cancer, cancers in general.

00:49:09.360 And of course, it's a scary disease.

00:49:10.780 It's a second largest killer about to become the largest killer of people in the United States.

00:49:17.600 So it's absolutely a scary disease.

00:49:19.720 That said, overall, mortality from cancer has been decreasing over the last 20-odd years.

00:49:26.980 So 20-odd years ago, it was 200 deaths per 100,000.

00:49:31.440 That's gone down to, what, 140 deaths per 100,000.

00:49:34.200 So there's absolute progress being made.

00:49:36.440 It's a mixture of prevention, some early detection, and some treatment.

00:49:39.720 Largely driven by prevention, some early detection, some treatment.

00:49:43.640 But let's talk about treatment.

00:49:44.960 So big radical changes in some cancers.

00:49:48.520 Immunotherapy, everyone's heard about immunotherapy, using your own immune system to direct it

00:49:53.840 against cancer, using ways to, you know, cancer cells have mechanisms to conceal themselves

00:49:59.260 from the immune system.

00:50:00.900 These medicines take those cloaks away or they make the immune system point towards

00:50:06.600 the cancer.

00:50:07.680 There are several of them now.

00:50:09.320 These have been radically effective for some cancers.

00:50:13.280 You know, I used to have a bet when I was a fellow that we will never have cures of advanced stage lung cancer in my lifetime.

00:50:22.340 And I lost that bet.

00:50:23.780 So now there are the word cure is a complicated word.

00:50:27.580 I rarely use it because, you know, sometimes, you know, 10 years later, something might relapse and come back.

00:50:33.600 But in lung cancer, for instance, non-small cell lung cancer, we're seeing a situation where some patients and we don't know which patients and why.

00:50:42.680 but some patients, about 20% of the patients are living out five years when they're given these

00:50:48.020 immunotherapy drugs. Bladder cancer is another example where there's been a lot of progress on

00:50:53.180 immunotherapy. We talked a little bit about breast cancer, you know, breast cancer with

00:50:58.240 advanced therapies, some immunological therapies, some antibody therapies. Again, we're seeing

00:51:04.320 cases in which people are living 5, 10, 15 years after their initial diagnosis of breast cancer.

00:51:10.660 And by 5-2-3, I don't mean sort of, these are real dignified years.

00:51:14.800 These are people who are working.

00:51:16.760 They are, you know, functional.

00:51:18.360 A couple of more examples, myeloma is a multiple myeloma.

00:51:21.920 You know, if you plot the survival rate of multiple myeloma based on what year you were

00:51:26.380 diagnosed, I know I'm using the word survival rate, but I'm using it in a very specific

00:51:30.360 context.

00:51:31.420 But if you plot that in 1990, 95, 2000, 2005, every five years, people diagnosed with multiple

00:51:38.640 bieloma, same stage, live longer and longer. And the last one I'll mention is the one that

00:51:44.880 sort of the story of where the story of chemotherapy begins, and that's acute lymphoid

00:51:50.080 leukemia in children, ALL in children. So by the 1980s, 80 to 90% of children with ALL were being

00:51:59.800 cured by very toxic but conventional chemotherapy. But that still left about 10 to 15% of children

00:52:06.640 who were called relapsed, refractory. They had relapsed and they were refractory to chemotherapy.

00:52:13.420 There are now new treatments. They're called CAR T-cells or T-cell treatments. This is a T-cell

00:52:18.320 that's been weaponized to kill that cancer cell. And we're seeing cure rates in these patients.

00:52:24.000 So again, five-year survival after therapy of around 50% to 60%, maybe a little bit larger,

00:52:31.240 a little bit more than that. For some cancers, quite a few cancers, I would say, we've seen

00:52:36.540 radical changes in treatment and potential cures. I've heard that CAR-T therapy has been very good

00:52:44.680 with blood cancers, but it's been challenged against solid tumors. Is that true? And if so,

00:52:50.920 what is it about tumors that poses a special obstacle? So first of all, it's true. CAR-T

00:52:57.980 therapies have been very successful in liquid tumors. In fact, I'm very involved in the field.

00:53:05.000 I made one of the first CAR-Ts in India for against lymphoblastic leukemia, ALL, that same

00:53:11.780 disease. I've made CAR-Ts against other forms of leukemia as well. So the sad answer is we don't

00:53:17.540 know. There's something different about liquid tumors and solid tumors, something in the so-called

00:53:22.740 microenvironment. Remember I said tumors don't grow in a vacuum. There are seeds that are

00:53:28.920 surrounded by soil. And in the case of solid tumors, there's a lot of soil. They're surrounded

00:53:35.160 by themselves or each other. They're surrounded by blood vessels. They're surrounded by immune

00:53:39.400 cells. They're surrounded by supportive cells that support their growth. So this thing is

00:53:46.080 called the microenvironment of a tumor. And for some reason, CAR T cells don't seem to be able

00:53:51.940 to penetrate the microenvironment of a solid tumor and deliver their kill. So that's changing

00:53:59.760 over time. We're actually combining CAR T cells with therapies that can make the microenvironment

00:54:06.460 less resistant. But for some reason, CAR T cells have never really fully grown to show their

00:54:13.040 promise in solid tumors. Now, cancer drugs, there might be some exceptions here, but my understanding

00:54:19.160 is that just as a class of drugs, they're notoriously expensive. As treatment becomes

00:54:25.100 more personalized and sophisticated, is this synonymous with them growing more expensive

00:54:31.740 still? And what are the social or scientific implications of this?

00:54:37.880 Well, the social and scientific implications are well known. I mean, we are spending billions of

00:54:42.680 dollars of money on cancer drugs, and they're expensive. Mostly, and we'll talk about why

00:54:48.940 they're expensive in a second. But the good news in some ways is that some of these drugs,

00:54:54.560 some of the most very promising drugs, like the immunotherapies that I talked about,

00:54:59.140 are going to come off patent soon. And generic versions are going to be available. There's

00:55:04.380 always a fight between legacy companies that have made the drug that will keep saying that

00:55:11.260 the original drug is actually still the better drug. That's mostly not true. The FDA ensures

00:55:16.000 is that the generic drug that emerges, which is usually one-tenth the cost or should be one-tenth

00:55:21.700 the cost, is actually just as effective as the pioneer drug. So that's one piece of good news.

00:55:27.120 There are many cancer drugs that are coming off patent, and that should decrease the price

00:55:32.140 dramatically, which is a reminder to us that we should be respectful of the patent cycle.

00:55:38.700 So I think it cuts both ways. We should be respectful of patents, but we should also be

00:55:43.020 respect for the patent cycle. So which means that when someone makes an invention, pours sweat,

00:55:47.680 blood, and tears into this invention, does a clinical trial, they get protected from

00:55:53.100 infringement for, depending on the particular class, for let's say 20 years. After those 20

00:55:59.460 years, these efforts to continue to extend the patent life cycle of a drug, we should be resistant

00:56:05.760 to that because they've gotten there 20 years. They've made their ample amount of money. They

00:56:10.280 should have spent that money on innovation and making new drugs. And if they haven't,

00:56:14.720 that's their problem. They should basically give in to the generics, as it were.

00:56:18.520 Sid, you must have seen this article by Catherine Ebon, who I think was in Vanity Fair maybe eight

00:56:24.960 years ago, that suggested, with a fair amount of research, that the pipeline for generic drugs

00:56:31.060 in particular, but really even the precursors of brand label drugs, was far less reliable than

00:56:38.500 anyone would hope. And I mean, if memory serves, something like 30% of generic drugs didn't even

00:56:47.460 contain the advertised compound. And there's just all kinds of, I mean, it detailed this

00:56:54.220 kind of a litany of corruption where, you know, labs in India, generic labs in India were tipped

00:56:59.200 off once a year when the FDA is going to come inspect their lab, et cetera. So there's kind

00:57:04.120 of a Potemkin village of laboratories. I mean, how aware of that problem are you? Has it been 1.00

00:57:11.440 exaggerated? And more importantly, if it was real, is it less real today?

00:57:16.660 Well, it certainly was real for a while. It's become less real. So the solution to this is

00:57:23.720 not to have spot audits, but to have continuous audits and to have continuous checks. This is not

00:57:30.760 a difficult thing to do. For instance, there are multiple mechanisms by which you can keep checking

00:57:36.060 whether a generic drug coming usually from India, from China, from South Korea, less from China

00:57:41.760 because of geopolitical reasons, but from South Korea. Sometimes they come from very diverse

00:57:47.920 sources that they actually contain the active ingredient. It's actually not hard to check this.

00:57:53.620 This is a relatively simple check. You can put it through a machine like an NMR or other kinds

00:57:59.240 machine, which will ensure that the parent drug and the generic drug actually are actually the

00:58:04.900 same. And in fact, since the so-called multiple scandals that have erupted because of this,

00:58:12.560 the typical scandal was a factory, let's say in India, would get tipped off that there's an FDA

00:58:17.880 inspection coming and all of them would just sort of clean up the factory, put on their coats and 0.99

00:58:22.040 start making the real drug as it were, and then go back to their old ways as soon as the FDA

00:58:27.180 inspector had left. So that's why continuous audits are helpful and also continuous checks,

00:58:33.500 quality QC checks made independently by an independent organization, whatever you want

00:58:37.560 to call it. And I'm very much aware of the original article in Vanity Fair that really

00:58:42.340 pointed out this as a major problem. So that's one solution, which is the genericization of

00:58:49.000 high value, high impact patented drugs should bring the cost down. The other solution, and we'll

00:58:54.840 Now switch a little bit, talking more about new technologies and potentially AI.

00:58:59.360 The other solution is, you know, part of the reason that the cost of drugs is so high is

00:59:03.200 that most pharmaceutical drugs fail.

00:59:05.540 And most pharmaceutical drugs fail because they don't have the right research apparatus.

00:59:10.740 You know, they're basically two or three reasons.

00:59:12.680 But let's say the two big reasons is that they've got the wrong target.

00:59:16.000 In other words, they're targeting the wrong protein.

00:59:17.940 Protein is the machinery that drives the cancer cell, or they've got the wrong chemical.

00:59:22.840 The chemical is not good enough to target that protein.

00:59:24.840 or the wrong biologic or protein to target the original protein.

00:59:28.960 So either they're missing the target or they're missing the protein.

00:59:31.820 And occasionally, it's because they run the wrong kind of study.

00:59:34.760 Now, what's interesting is that in the new world, we have, and I'm involved in this very

00:59:40.000 personally, so I should give that as an important caveat.

00:59:43.540 In the new world, we are making more and more drugs through a combination of virtual means

00:59:50.940 and real, you know, we don't take a virtual drug and put it into human patients. It has to be then

00:59:56.220 tested on animals and potentially then go through a human clinical trial and ultimately becomes a

01:00:00.960 real drug. But in all that, in that life cycle of the birth of a new medicine, we have new

01:00:07.460 technologies, including most importantly, perhaps AI as a new tool to make drugs, to test drugs,

01:00:15.060 to test drugs efficiently, and hopefully bring the cost of a trial down or the life cycle of a

01:00:20.220 drug down so that you can actually make cheaper, better, faster drugs.

01:00:23.720 Okay. So let's talk about AI because I know you have your own effort here, which I want to hear

01:00:27.840 about. But the context that many people will have noticed is that there was a big piece of press

01:00:34.100 some years ago when AlphaFold solved the protein folding problem. And I forget what the color on

01:00:40.920 this was. It was something like, you know, had done the equivalent of, you know, 200,000 PhD

01:00:45.240 dissertations. I mean, it's like the equivalent man hours was just ridiculous. So that obviously 0.73

01:00:52.040 suggests that AI can be helpful in finding plausible targets for medications and crafting

01:01:00.440 molecules for those targets. Obviously, there's a prospect that AI will transform radiology and

01:01:09.800 and data analysis. What is still just promise or hype at the moment? And where is AI really

01:01:19.320 making a change to outcomes for people now? So if you look across the spectrum,

01:01:26.360 I think AI has already delivered promises in some parts and in other parts is about to or

01:01:33.160 has started delivering promises. So you should really think about not one AI, there are multiple

01:01:39.160 AIs for this. We're not talking about acquired general intelligence. We're talking about what's

01:01:45.700 called neurosymbolic AI or AI that's been taught on rules and then are taught on patterns in some

01:01:52.220 cases. And we're talking about AIs that are different in each and every case. So let's again

01:01:58.240 start with prevention. So in prevention research, there's not been a lot of use of AI yet,

01:02:04.400 but it's very ripe for AI research. The reason it's very ripe for AI research is that prevention

01:02:09.480 research, again, to remind people, the kind of study that would be very helpful in prevention

01:02:14.520 would be to figure out, you know, what is your background genetics? What are you exposed to? So

01:02:19.560 what's your exposome, as people call it? What is your, you know, you can add in other things like

01:02:24.880 what is your microbiome? What is your, you know, what are other large multidimensional features

01:02:30.580 that comprise you, genetics, exposures, behaviors, diets, and so forth, and then construct, as it

01:02:38.800 were, a multidimensional version of you and ask the question, if you construct that multivectorial,

01:02:44.540 multidimensional version of you, who is likely to get a higher risk of any one cancer? So that is

01:02:51.280 a kind of problem that humans are not very good at, but AI is quite good at because it's a highly

01:02:56.080 complex multidimensional problem, and ultimately produces a correlation. It's not going to tell

01:03:01.360 you why something is causing cancer, but it's going to tell you a correlation, and then you

01:03:06.260 can do subsequent experiments to figure out why. So that's one area. The second area you identified

01:03:11.620 was in detection and in diagnosis. So again, an area that AI has played a very strong role in.

01:03:20.560 So, as you know, mammography is routinely used to detect early breast cancer, and there's a miss rate. And the miss rate is because the radiographer hasn't seen or finds a funny pattern that they miss. AI is a very good, you know, I think of it as a person whispering across your shoulder and saying, well, are you sure about that little white spot?

01:03:42.100 So it's almost like having a companion with a human being.

01:03:46.280 And that's been more and more used.

01:03:49.260 That's true now for screening for lung cancer in high-risk patients.

01:03:53.340 It's true for their AI modules that look at a skin lesion and make a decision whether

01:03:59.580 it's a melanoma or not melanoma.

01:04:01.400 Is this something that people can take for granted now?

01:04:04.320 I mean, if you're going in to get any kind of medical imaging done, more or less anywhere,

01:04:10.380 let's just call it the United States. Can you safely assume that in part of the workflow of

01:04:16.380 data analysis, there is an AI component now, or is this only happening in bespoke places in the

01:04:22.300 biggest cities or in research hospitals? It's largely still in bespoke places.

01:04:27.320 Some of them are still actually in test mode, in beta mode. But the chances that this will succeed

01:04:33.360 as a companion. I often say the word diagnosis comes from the root of the word is learning

01:04:41.100 together. And this is going to be a companion mode. There are various ways you can think about

01:04:47.480 it. You can think about a triage as a triage mechanism. You can think about it as a second

01:04:51.740 opinion mechanism. But nonetheless, it's coming. I would say this is a likely given for radiology

01:04:58.920 and potentially for pathology as well.

01:05:01.620 So, you know, when you have a pathological lesion,

01:05:03.500 you put it under a microscope, you take a picture,

01:05:05.460 the pathologist says, I'm not sure if it's cancer or not.

01:05:08.440 The AI, the AI in that case,

01:05:10.660 has been trained on typically 500,000 images of a melanoma

01:05:16.000 or, you know, 500 million images of a melanoma.

01:05:19.000 So the chances that, you know,

01:05:21.140 whereas a pathologist may have seen 500.

01:05:23.460 So this is a great arena

01:05:25.160 where a companion diagnostic is very useful.

01:05:27.320 So let's now move on to drug discovery and clinical trials. So those are two other areas which are very interesting and important. So in drug discovery, this is what we do. This is what Manus AI does. That's my company. I co-founded it with Ujwal Singh and Reid Hoffman.

01:05:44.240 And what we do, the crucial insights that we discovered was that if you want to do drug

01:05:49.380 discovery with AI, you have to teach AI the rules of medicinal chemistry.

01:05:55.340 And that's not an easy task.

01:05:57.200 A medicinal chemist has a massive brain.

01:06:00.120 They've been trained for 20 years.

01:06:01.940 And when you find a pocket, they'll find a way to insert or create a drug for that pocket.

01:06:07.360 And AI doesn't know any of these rules.

01:06:10.380 It starts from scratch.

01:06:12.100 And the other problem is that there are not enough exemplars.

01:06:15.820 So just like I said, there are 500,000 specimens of myeloma sitting in some bank somewhere.

01:06:21.520 And AI can look at those and learn the images, look at those images and learn the pattern

01:06:25.400 and look at a new one and say, that's a myeloma or not a myeloma.

01:06:29.920 There's not enough teaching data on generative chemistry, on true drug generation.

01:06:36.760 So you have to teach it the rules.

01:06:38.800 And that's something very important.

01:06:40.200 it's difficult to do, but it's a very important thing. The second arena is target discovery. So I

01:06:45.780 just said, you know, every drug, every medicine works by binding to a target, usually a protein.

01:06:51.360 So on one hand, you know, AI can help with target discovery. Manus doesn't do that. We have

01:06:56.660 collaborators who do that. There are many academic labs who do that. So finding out, you know,

01:07:00.740 what's a good protein to inhibit, to activate, to, you know, what's the, the analogy is lock and key.

01:07:06.620 In one case, how do we find the locks, and then how do we find the keys?

01:07:11.300 So the way you find it, the AI is very helpful in finding the locks because the lock involves

01:07:15.880 taking, again, very multidimensional cellular data and finding out where the lock is, turning

01:07:22.280 the lock, turning of which will stop the cancer from growing.

01:07:25.860 So there's a big role for AI in target discovery.

01:07:29.620 Second role for AI, as I said, in molecular discovery, still to be fully proven out, but

01:07:34.800 as you may know, for non-cancerous diseases, a recent spate of papers have shown that for

01:07:40.240 non-cancerous diseases, and in fact, some for some cancerous diseases as well, you can use AI to

01:07:45.200 build a molecule or to find a molecule. One is a search algorithm and another is a build algorithm,

01:07:50.180 but you can use AI to find a molecule that actually would turn the lock, the key in the

01:07:54.800 right way. Finally, final note is about clinical trials. Clinical trials can be extraordinarily

01:07:59.360 powered by AI. AI can, for instance, to give you one example, go into hospital records under safety,

01:08:06.020 under HIPAA rules, et cetera, et cetera, go into hospital records and identify patients who are

01:08:10.520 likely to benefit from a particular trial or a particular drug. So that's a data search problem.

01:08:16.340 In fact, we already have language models that are able to scrape the web or scrape electronic

01:08:22.280 medical records and find the right kinds of patients. And secondly, we have things called

01:08:27.580 adaptive trials. Adaptive trials are trials in which basically over time, the trial itself evolves.

01:08:34.000 More people are moved to one arm or to another arm to ensure that there's a balance as we move

01:08:38.700 along. And the trial learns as it moves along. And that's another, as soon as you use the word

01:08:44.000 learning, it means that if human beings can learn that, then certainly AI can learn that.

01:08:48.660 So when you think about AI in medicine, in particular, if you think about different AIs

01:08:54.080 doing different things for different aspects of medicine, all of which are very empowering and

01:09:00.100 powerful. When you think about the future, do you think about it more or less being a foregone

01:09:06.900 conclusion that at some point cancer will be fully behind us and we'll look back on all of

01:09:14.680 those generations of people who lived in a world where cancer was more or less untreatable and

01:09:20.460 and just, I mean, we'll just feel the poignancy appropriate to that. It's just a contingent fact

01:09:27.140 of history that at one point we had no idea how to stop this thing, and now it's not even a thing.

01:09:34.120 I mean, are you anticipating that kind of future in some, you know, how surprised would you be not

01:09:40.620 to achieve a future like that, you know, if we don't destroy ourselves some other way in the next

01:09:45.980 50 to 100 years?

01:09:48.640 Well, hopefully we won't destroy ourselves.

01:09:50.100 But look, as far as cancer is concerned, I'm an optimist.

01:09:53.380 I've seen in my own lifetime, many cancers slowly transform into from incurable acute

01:10:01.900 diseases to chronic diseases and some to curable diseases.

01:10:06.520 So I'll give you a couple of examples.

01:10:08.860 Breast cancer, I talked about there, you know, somewhere between you and me and the people

01:10:15.100 in the studio. There's a woman who has breast cancer who's now lived her life with dignity and

01:10:20.840 with a good quality of life 15 years since her original diagnosis, 20 years since her original

01:10:26.960 diagnosis. A century ago, she would have been miserable with undergoing surgeries for advanced

01:10:33.260 breast cancer and having all the consequences of those surgeries. But some cancers we've had a

01:10:38.300 very hard time with. Acute leukemia, myeloid leukemia, not the kind that most of the children

01:10:44.040 get. But acute myeloid leukemia, we've had a very hard time with. So a lot of my own research has

01:10:48.940 been how to find new ways of treating acute myeloid leukemia that's different from the

01:10:54.020 current paradigm. In fact, we use CRISPR technology to try to beat acute myeloid leukemia.

01:11:00.000 Recent data, for instance, there's a big stir in the world because a very old target of cancer

01:11:08.220 called RAS. The gene is called RAS, a new medicine from a company called Revolution Medicine,

01:11:13.880 and in fact, several other companies are making them. A company called Revolution Medicine made

01:11:17.420 a RAS inhibitor. RAS is one of these so-called four horsemen of death of cancer, one of those

01:11:23.080 genes that keeps coming up across multiple cancers. And you can imagine RAS as driving

01:11:30.560 the cancer with its whip. This medicine essentially holds the whip and stops it from moving. And

01:11:37.300 therefore the cancer is no longer able to respond to that malignant signal from RAS.

01:11:43.000 In the clinical trial for pancreatic cancer, as you know, pancreatic cancer is a terrible disease.

01:11:47.560 We haven't been able to budge mortality from pancreatic cancer for decades. In a clinical

01:11:53.180 trial for pancreatic cancer, randomized patients who were given this drug lived 13 months versus

01:12:02.140 patients who were treated with standard therapy who lived six months. And you could say to yourself,

01:12:06.180 well, who cares, 13 months. But that's how cancer therapies evolve. The way cancer therapies

01:12:13.080 evolve these days is that it's a little bit like driving the first crampon into a mountain.

01:12:21.880 The first crampon is not the way you climb. You're not going to climb the mountain with

01:12:24.960 the first crampon. But the first crampon gives you a foothold on what the problem is. And it's

01:12:30.500 the first crampon that allows you to then put the second crampon on. So the second crampon in this

01:12:35.260 cases to say, well, okay, the RAS gene became blocked or stopped. What happened? Why did these

01:12:42.420 patients relapse after 13 months? So you put the next cramp on and then you put the next,

01:12:47.540 and this is how basically bit by bit by bit, many, many other terrible diseases, myeloma was a good

01:12:53.860 example. This is how these cancers became more and more chronic diseases and in some cases became

01:13:00.320 curable. So the big story is not that we increased the survival of patients by six months

01:13:09.260 with this new RAS inhibitor. The big story is that we planted the first crampon in 20 years

01:13:15.080 against pancreatic cancer, which was really not planted before. So do I think that cancer is going

01:13:20.940 to go away from human biology, from human history forever? No, that's impossible. As cancer is a

01:13:27.420 disease of aging. We can change our lifestyle. We can decrease the incidence. There are medicines

01:13:33.040 that will, for instance, obesity is related to cancer. So as the population hopefully becomes

01:13:38.860 less obese because of other medicines, lifestyle changes, healthy changes, yes, we will decrease

01:13:44.080 the risk of getting cancer. Obviously, you know that the decrease in smoking has been the largest

01:13:50.160 driver of the decrease in cancer mortality ever in human history. So do I think it'll go away

01:13:55.380 completely? No. Some people get cancer because of just plain old bad luck, and that will still

01:14:01.980 remain. Do I think that many of those cancers will become treatable in the future? Yes. Will

01:14:07.140 there be some that will remain sort of frustratingly out of our reach? Yes, but that number will be

01:14:12.780 fewer and fewer. All right. Finally, a political question that I gestured at in some disparaging

01:14:19.480 remark about RFK Jr. earlier. What is the Trump administration doing to medical science at the

01:14:25.680 moment? I mean, it was much, the vandalism was much discussed initially. Has much of it been

01:14:34.700 significantly rolled back quietly, or are we still in a state of just lighting everything on fire for

01:14:41.840 no good reason and defunding essential medical science and putting ideologues and conspiracy

01:14:49.700 nuts in charge of everything. I mean, just how much damage has been done and how much damage

01:14:54.600 has been quietly repaired, if any? Well, a lot of damage was done initially,

01:15:00.760 and that's evident by the fact that there's been, you know, across the entire academic

01:15:07.660 establishment and certainly across the drug development establishment, there's been

01:15:11.880 seeds of chaos were and have been planted. And that's been a huge problem. I can say

01:15:18.820 very globally speaking, and by globally, I mean in the United States in terms of institutions,

01:15:26.320 severe funding cuts to the CDC, threatened funding cuts to many other organizations,

01:15:31.320 and certainly no increases in budgetary increases, increased scrutiny of research where scrutiny was

01:15:39.740 not required, lack of scrutiny of research where scrutiny is required. So it's really been a kind

01:15:46.500 of, I would say, I would describe it as a minefield. That said, I think cancer is something

01:15:54.760 that affects all populations, Republicans, Democrats, regardless of your political leanings

01:16:01.480 and political spectrum. And the public has spoken. I think increasing that the public is speaking,

01:16:06.440 the public has spoken. It's speaking about other diseases. It's speaking about the fact that

01:16:11.020 in a highly civilized country, we have a thousand odd cases of measles. And there are deaths from

01:16:21.780 measles because people have become reluctant to give the measles vaccine. So without pointing

01:16:29.340 individual fingers, I would say the administration has been relatively anti-science, but also that

01:16:36.540 the voice of the people, and by that I mean the larger American people, has always been, in some

01:16:42.540 sense, a voice of sanity and a voice that says science has to be restored in order for us to

01:16:48.600 make progress. We just have to restore scientists. Trust in science has to be built. Some of the

01:16:55.100 fault is the fault of scientists themselves. They've locked ourselves up in ivory towers

01:17:01.600 that didn't fully communicate with the public about what's going on. Drug prices are a big issue

01:17:10.220 and people feel the pinch of drug prices and they feel very annoyed that pharmaceutical companies

01:17:15.020 are racketeering their way through all of this.

01:17:17.780 But slowly over time, I think some of this will be repaired,

01:17:23.240 is being repaired.

01:17:23.880 The problem, as you know, Sam,

01:17:25.780 is that when you destroy an institution,

01:17:27.880 it's very easy to do.

01:17:29.200 But rebuilding that institution takes years and years of work.

01:17:32.320 It's one fell swoop of a pen, and the USAID is gone.

01:17:36.700 One swoop of a pen, and half the CDC has been dispatched.

01:17:40.920 Restoring these people, because people lose their training,

01:17:43.500 they lose their jobs, they lose interest in coming back to the job and so forth. So restoring the

01:17:48.600 ecosystem will take years and years and years. I'm very concerned about the fate of US science.

01:17:54.860 And I'm also concerned about the fate of US innovation. Of course, we're doing a lot of

01:17:59.400 innovation in AI, but just to give you a number that will maybe stick in your head, in 2020,

01:18:05.760 the United States in license, in other words, brought in from China about $5 billion of drugs.

01:18:12.900 in 2025, that number will be $60 to $70 billion, expected to be $60 billion. So in other words,

01:18:20.460 most of the medicines that we're getting in the United States are really medicines that are

01:18:25.540 emerging from Chinese biotech companies and are being imported by the United States.

01:18:31.960 And we've just taken the most valuable thing that the United States produces, which is innovation.

01:18:37.480 I'm not even talking about the pharmaceutical industry in particular. We've taken the most

01:18:41.200 valuable thing the United States produces and made it and hobbled it. And that is going to be

01:18:47.780 very, very difficult to repair. I thought one of the indelible lessons from the COVID pandemic was

01:18:54.380 that we needed to onshore many of these supply chain essentials. Well, in fact, we've offshored

01:19:00.460 them. We didn't learn that lesson. You remember we discussed this exactly in a previous podcast

01:19:07.800 about onshoring of medical resources, onshoring of medical technologies, onshoring of manufacture,

01:19:15.200 really. I'll just give you another surprising fact. I have a great fear that some supply chain

01:19:21.960 disruption of some kind, and I can name many different kinds, will suddenly cause hospitals

01:19:26.880 not to have intravenous saline. Saline is sterile salt and water. And you cannot go into a hospital,

01:19:35.060 you cannot perform surgery. You cannot do a simple procedure without sterile saline.

01:19:41.760 And if that runs out, you can imagine the whole hospital with all this very fancy medicine,

01:19:47.560 $10 million MRI machines, et cetera, none of it will work. So absolutely, we need to make sure

01:19:55.140 the supply chains are resilient, they're robust. And the best way to make them resilient and robust,

01:19:59.900 of course, is to onshore them. It keeps manufacturing. The administration keeps saying,

01:20:04.200 we want more manufacturing jobs in the United States. Well, here's an area. Make more manufacturing

01:20:09.920 jobs for life-saving either medicines or life-saving pharmaceutical products and ensure that the

01:20:17.180 supply chain is not disrupted. Well, Sid, it's always great to get you on the podcast. Thank

01:20:21.860 you for the work you're doing and just the clarity of your communication around this issue.

01:20:26.400 My pleasure. Thank you. And thank you for your podcast, which always does a great service to

01:20:30.540 science. Oh, nice. Nice. Well, again, that reminded people, the new edition of The Emperor

01:20:36.260 of All Maladies is out there, four new chapters. It's a great read, and it won the Pulitzer for a

01:20:42.400 reason. And I hope to get you back here. The door's always open. When you feel like the story

01:20:49.260 has changed in any important way, whether it's with respect to AI or anything else, please give

01:20:54.060 it a knock and come back on. Yeah, I think the next time we'll come back probably is when AI

01:20:57.640 has produced a whole bunch of new medicines.

01:20:59.840 Nice.

01:21:00.080 And we can talk about how they were made.

01:21:01.500 Nice, nice.

01:21:02.260 All right.

01:21:02.620 I await that time with pleasure.

01:21:04.080 Great.

01:21:04.420 Thank you so much, Sam.

01:21:05.160 Take care, Sid.

Making Sense - Sam Harris - July 14, 2026

#485 — The New Science of Cancer

Episode Stats

Summary

Transcript