213 ‒ Liquid biopsies and cancer detection | Max Diehn, M.D. Ph.D.

1. 0:00 – 0:08

   Intro

   1. PAPeter Attia0:00

      [upbeat music] Hey, everyone. Welcome to "The Drive" podcast. I'm your host, Peter Attia.

2. 0:08 – 8:45

   Max’s training that planted the seeds for development of liquid biopsies

   1. PAPeter Attia0:08

      Hey, Max. Thanks so much for making time today. So wonderful to see you again. It's been a lot of years, huh?

   2. MDMax Diehn0:16

      It's been a long time. Actually, I can't remember the last time we saw each other, but since we kinda-- We started together at m- in medical school at Stanford but then finished at a little bit different times. That, that kind of... I think it's been a while s-since we saw each other.

   3. PAPeter Attia0:28

      Yeah, well, uh, as I was sort of joking earlier, this is, this is-- Y- We're, we're gonna have, like, a whole subset of "The Drive" podcast which is based on the Stanford MSTP, uh, students between Carl, you, and Josh. So, um, you know, again, kinda speaks to the, to the quality of, of people that, you know, were a part of that program. Um, so, so let's tell folks a little bit about the kind of work you did. So as you mentioned, obviously, we started in medical school together. We have some really funny stories from the beginning of medical school, which I think we'll, uh, refrain from telling at this point in time. Um, but they, they-- I could have a whole podcast just on some of that stuff. Um, and then after the first two years of medical school, I went right off into the clinical stuff, and then you went off into the lab. Uh, tell, tell folks about whose lab you, you went to and what it is that you started pursuing and, and, and frankly, how you even made that decision.

   4. MDMax Diehn1:18

      As you pointed out, as part of the MD-PhD program here, I-- The, the way it is done in most programs, you split the curriculum with splitting medical school in half, doing the first two more classroom-based years first, then doing the PhD work in the laboratory and then going back for the clinical work. And so that is a-- There's an important transition point there when you're finishing, uh, the, the classroom part of medical school and deciding what lab to work in. And, um, I, uh, ultimately chose to do my, uh, dissertation with, uh, Pat Brown, who was a professor in, uh, of biochemistry here, uh, at, at Stanford. Uh, he is no longer. Uh, a lot of your listeners may know of him, though, because he's now a C-E-O of Impossible Foods, or was until recently. I think he just transitioned to a slightly different role, but, uh, founder and CEO of, initially of Impossible Foods. But he was here, uh, a faculty member, and, um, what really attracted me to his lab was that he had, um, you know, just around that time, invented a, uh, technology for measuring the expression of basically all the genes in the genome, uh, with a technique called DNA microarrays. And that at the time was revolutionary. Before that, we were always measuring everything in a one or a handful of genes at a time in experiments, and now we could measure tens of thousands in one experiment. And it just seemed to me that this was opening up whole new fields, that we would be able to learn so much, uh, and that's why I chose, uh, to pursue that lab.

   5. PAPeter Attia2:47

      So what did you do for your actual dissertation? What was the, what was the project that you worked on?

   6. MDMax Diehn2:52

      I, I had a little bit unusual dissertation in that I worked on many different project. Uh, it was, uh, a very unique time, uh, in, in, in the lab, in Pat's lab, as well as in labs in general, where... And I've seen this happen, uh, since, uh, several times, but when there's a new technology that's developed that's sort of transformative, it opens up so many doors simultaneously that, you know, you have this new tool that no one's ever had before, a new lens to view, through which to view biology. And so you can just immediately think of thousands of questions that would be interesting to ask. And so I, I worked on a, a variety of different things, the mainly two main areas. One is immunology, the other, uh, oncology or cancer biology, and those were my two interests coming into medical school. And so, um, you know, I, I felt fortunate that I was able to do projects in, in some of both. Um, as far as, you know, maybe I just hi-highlight one or two of them. Uh, one of the projects was, uh, focused on, um, T cells, which are a type, of course, of, uh, white blood cell or lymphocyte that, um, are a critical part of the adaptive immune system. And, um, so we had this new tool to measure all the genes in the human genome at one time to see how they go up and down after you, uh, you know, perturb cells. And so we were very interested to see what genes are turned on and off in T cells when you activate them, when you stimulate them through their receptor, the T cell receptor, either by itself or with a co-stimulatory signal.

   7. PAPeter Attia4:14

      You would look at the T cell. It's stimulated, and you're... Are you actually measuring protein, um, or you-

   8. MDMax Diehn4:21

      RNA. The, the-

   9. PAPeter Attia4:22

      You're measuring RNA?

   10. MDMax Diehn4:22

       This technology, RNA. This-

   11. PAPeter Attia4:23

       Okay.

   12. MDMax Diehn4:24

       Yeah. This technology is, was a way of, uh, measuring RNA, uh, so transcripts-

   13. PAPeter Attia4:28

       Yeah

   14. MDMax Diehn4:28

       ... so the intermediate between DNA and proteins, of course. And, um, so, uh, we could see, uh, you know, hundreds and thousands of genes changing as we, uh, manipulated, um, the cells, and so building a catalog of all the genes that, uh, that turn, uh, are turned on or down-regulated, turned off, uh, when you activate T cells using different signals. And that, um, catalog then of course has been very helpful for subsequent studies in, in better understanding and teasing apart, um, the, uh, mechanism of, uh, T cell activation, which has gotten increasingly more interesting, of course, now with the advent of immunotherapy.

   15. PAPeter Attia5:08

       Yeah. So there's a lot of things there that are interesting. One of them is the instability of RNA. This is gonna become relevant as we start to talk about the differences between cell-free DNA and RNA. But at the time, um, how difficult was it to basically keep all of this RNA intact as you cataloged the, uh, generation of mRNA from DNA as this, uh, signal of gene expression? I mean, was that one of the big technical challenges of this technique?

   16. MDMax Diehn5:38

       That for sure is, is a, a hurdle in all work that's focused on, uh, RNA, which as you mentioned, is chemically, uh, relatively unstable, particularly when you compare it to DNA, which is much more stable. Um, and so one has to be very careful, uh, in any experiment, whether you measure a single RNA or whether you measure twenty thousand RNAs, to really try to preserve the sample in a way such that you minimize the potential chemical degradation of RNA. So there are, m- you know, laboratory methods to do that, of course. So for example, if you're in that example of the T-cell stimulation experiment, um, you know, you-- the cells are alive, the RNA of course is maintained. It's really once the cells die that, that the, the degradation issue starts happening. So you, you design the experiment in such a way that you're very careful to immediately, you know, add, uh, solutions that protect the RNA after you kill the cells at the end of the experiment so that there's no time for chemical degradation processes. I, I had a, I, I... My, my other main project in PhD was actually-- we really stru-struggled with this because it was a separate project from my, my rotation project, which is the first project you do in a lab when you're kind of trying to figure out if you wanna go there, where we, um, developed a method to isolate self, uh, so RNA that's stuck to the, uh, endoplasmic reticulum inside cells, um, because that is where RNAs go for genes that are ex-excreted from the cells or that are surface proteins in the cells. And we were interested in cataloging those because those are interesting for diagnostic purposes and therapeutic purposes. So we had to purify the, the subset of the RNA that was, uh, stuck to the, these organelles called the, the endoplasmic reticulum. And to do that, you ha-- there's a very long procedure where you have to build special gradients and float, you know, gently slice the cells, and then th-float the membranes that they're stuck to to various levels in a test tube. And that required a lot of work in a cold room where you, you know, you're basically... You have test tubes in a four-degree room, but so is the experimenter. You, you know, you have to wear jackets and stuff 'cause you're working basically in a fridge. It's all to maintain the integrity of the RNA, as well as you can add chemicals to try to, to stabilize the RNA. So this is something that was critical at the time to really try to work out methods to stabilize the RNA.

   17. PAPeter Attia7:44

       And how much of an insight could you get into non-coding sequences of genes where they're not making proteins, but we now know, of course, that these non-coding segments can be very important as well? Could you, could you gain any insights there?

   18. MDMax Diehn7:59

       So with the technique we were using at that time, uh, we could not directly because we were focused on measuring the coding portions of the, the, of the genes that, uh, of the transcripts that, that, that code for proteins. Um, and we had to decide at the beginning of each experiment which genes we measured, because basically, we had to create a probe for each.

   19. PAPeter Attia8:19

       You had to have the primers and, yeah.

   20. MDMax Diehn8:20

       You had to have the primers for it, and so you had to make a decision. So at the time, we were not focused on that. Um, but the, um, the, uh, subsequently, there, um, this, this approach was used, uh, uh, early, ear- in the, some of the early work on long non-coding RNAs, and that was, um, largely led by a former postdoc from, uh, Pat Brown's lab who was there while I was a grad student named Howard Chang, who is, uh, a professor here at Stanford, as

3. 8:45 – 16:00

   Max’s decision to specialize in radiation oncology

   1. MDMax Diehn8:45

      you, as you may know.

   2. PAPeter Attia8:46

      So I, I think you finished your PhD in about three years, correct?

   3. MDMax Diehn8:50

      Well, yes, it took about three years, yeah.

   4. PAPeter Attia8:52

      Yeah, yeah.

   5. MDMax Diehn8:52

      Three and a half years, something like that.

   6. PAPeter Attia8:53

      And then, and then you went back now and you started your clinical rotations. Did you have a sense when you went back for the last two years of medical school that you definitely wanted to be a clinician? So which would mean not just finishing medical school, but then doing a residency. Um, or 'cause, you know, some people in the MT PhD program, you know, just say, "Look, I just wanna be a pure scientist, not a physician scientist. I'm gonna finish medical school and get my MD, but I'm not gonna do any clinical training." D-- Where were you on that spectrum at, at the beginning of that final two years?

   7. MDMax Diehn9:24

      I was set on doing a residency. I wasn't sure yet in what, but I was set on doing it. You know, actually, my, my sort of, um, decision point was for me was made prior to med school when I was at, uh, an undergrad. Initially, I thought I was gonna do graduate school and go for a PhD and focus on research for my career. But then my father was diagnosed with lymphoma while I was a junior in college. And the, you know, the, the journey that he went through, uh, in interacting with the oncologist and the medical team as part of that and really seeing how little we knew about, uh, many things and, uh, how suboptimal treatments were really convinced me that, um, that number one, I, I now, I wanna be on the doctor side, not just on the patient side with my father and, and be able to help, uh, people, help patients at the, you know, the time when, of their life what might be the worst time of their life, as well as to try to move the field forward to improve treatments that, you know, while they worked somewhat, obviously were not good enough.

   8. PAPeter Attia10:23

      And so does that mean that even coming into medical school, not only did you know you wanted to be a physician and a scientist, but y- did you already have kind of a, an inkling that oncology was where you wanted to be?

   9. MDMax Diehn10:33

      Yeah, that's exactly right, yes. I knew I wanted to work, um, uh, in one of the cancer, uh, specialties. Um, what I didn't realize when I was a first-year med student is how many options there are in that. Of course, uh, you know-

   10. PAPeter Attia10:44

       You were probably looking at it mostly through the lens of medical oncology, um-

   11. MDMax Diehn10:48

       That's right

   12. PAPeter Attia10:49

       ... whereas, of course, you have surgical oncology, radiation oncology, and other sorts of avenues. Um-

   13. MDMax Diehn10:54

       That's right.

   14. PAPeter Attia10:55

       And remind me, were you born in Germany or were you born in the US?

   15. MDMax Diehn10:59

       No, yeah, I was born in Germany. I was born in, in Munich, and I, I didn't come to the US till I was eleven years old.

   16. PAPeter Attia11:04

       Okay. And you grew up on the East Coast?

   17. MDMax Diehn11:07

       Yes. We initially moved to South Florida. Um, uh, and then when I was a sophomore in high school, we moved to, uh, Connecticut, uh, the Northeast.

   18. PAPeter Attia11:17

       Okay.

   19. MDMax Diehn11:17

       And then I went to Harvard as an undergrad, so I, I was in the, the, the Northeast until I came out here, but then I've never, I've never left anywhere, and now I've lived longer here than anywhere else in the world.

   20. PAPeter Attia11:25

       Yeah. Yeah, I know that feeling. Um, so you go into the clinic, and now it's basically a decision of, you know, medical oncology, radiation oncology, or God forbid, something like surgical oncology, though I'm guessing that was probably third on the list of three options.

   21. MDMax Diehn11:41

       You know, I did honestly strongly consider it. And, uh, I, I, I do like the, the procedural aspects of it, and that's the one reason why I chose ultimately what I chose. Um, so I did, and I, you know, every surgical, uh, on my surgical rotations when I had a choice, I always chose the, you know, oncologic subspecialties and, you know, did some surg onc, uh, did some gyn onc, uh, s-surgery ro- rotations. But, um, ultimately chose not to do that in large part 'cause I s- you know, I saw that, um, I thought it would be difficult to, um, balance with, with research, and of course, people do it very successfully, but, but I think it is harder in general in the surgical specialties.

   22. PAPeter Attia12:18

       So how did you ultimately decide on radiation oncology then?

   23. MDMax Diehn12:23

       Yeah. Uh, radiation oncology, I, as you mentioned, I, I was sort of leaning towards medical oncology 'cause that's what I'd seen through my, my dad's, um, uh, journey. Um, but I, uh, I did a rotation in, uh, radiation oncology actually in large part because my wife, who was also a medical student at Stanford and who you of course, uh, know, um, uh, did a, a research year in the Department of Radiation Oncology. Sh- uh, she at the time was also interested in oncology, and so that actually is how I got exposed to the field. Um, while we were dating, I would often visit her in the lab and so, you know, I kinda got, uh, exposed in that way to, to, to the department and the field, and then so it, it made me want to try it as a, a rotation, and then I really enjoyed it. I... N- a v-variety of things I enjoyed about it. From a patient care standpoint, um, the... It seemed to me that radiation oncologists, uh, had a little more time in clinic to spend with each patients. You know, w- uh, we generally see less patients in a day than, than might be seen in medical oncology, uh, and, uh, that seemed very attractive to me. The other thing I, um, really liked was the technology aspects. I was, I've always been interested in technology aspects, uh, you know, both, you know, new technologies in the research space, but then also, you know, technologies for treating patients, and radiation oncology is very procedurally heavy. It's of course a field where we do a lot of imaging to see where tumors are in a patient's body, and then we have fancy robots that deliver the radiation very precisely. Um, so it just seemed like, uh, a great field to combine my, my love of oncology and patient care with my interests in, uh, technology development. And then lastly, I kind of saw it as a field where there really, compared to medical oncology, there h- there wasn't as much work being done in the laboratory at the molecular level, um, and it seemed like an opportunity to, to, to make a difference in a field where there, there weren't so many people working.

   24. PAPeter Attia14:08

       You know, one of the things I talked about with Carl was the challenge of keeping his hand in the lab during those last two years of medical school and then during his psychiatry residency. Um, he mentioned that he was... He really found himself in a great situation when he did his residency where he was effectively allowed to do kind of a postdoc, and he was freed from, you know, certain things. He didn't go to, you know, lab meetings and journal clubs and things like that, but he still got to do a little bit of work. Were you able to do anything like that during your residency, or were you really strictly focused on just the clinical training for... It's, was it four years?

   25. MDMax Diehn14:49

       Uh, yeah. So radiation oncology is, uh, I did, um, a internship in medicine, uh, here at Stanford, and then it's four years of radiation oncology, so five total.

   26. PAPeter Attia14:56

       Five years, yeah.

   27. MDMax Diehn14:57

       Um, d-during internship, of course, uh, as, as you I'm sure remember, [chuckles] there is no time to do an- uh, anything else, although I do remember trying to write some papers in the call room, uh, you know, to finishing up work from Ph- from the PhD. Um, but, um, the, um, uh, the, for the r- uh, radiation oncology, uh, uh, residency programs have a research track for individuals who are interested in laboratory-based research called the Holman Pathway, which is, uh, in other fields kind of called short tracking or fast tracking.

   28. PAPeter Attia15:23

       Mm-hmm.

   29. MDMax Diehn15:24

       It's a similar idea where basically one can trade in some of the clinical training time for research time. Um, and so that's what I did, which gave me, uh, a, a postdoc time of about two years during my four years of radiation oncology residency. During that postdoc time, I d- I had clinic activities about half a day to a day a week, um, uh, and the rest of the ti- other days were in the lab. So it actually was a really good preview of what, you know, my life would be like once I finished all the training and, and was a faculty member myself. Uh, and so that, that allowed me to kind of keep a foot in the clinic while al- mainly focusing on, on the postdoctoral research.

4. 16:00 – 18:50

   A culture at Stanford that values research and physician scientists

   1. PAPeter Attia16:01

      So when you finished your training, obviously the first decision you had to make is do you want to stay at Stanford? Uh, I'm guessing that between, you know, the connections you already had there, Jenny was probably by this point already out of her training and in practice. That was probably a very, uh, high activation energy to leave. Um, tell me about the Department of Radiation Oncology at Stanford. Uh, was it a natural fit for the types of problems you wanted to solve, uh, on the research side?

   2. MDMax Diehn16:29

      Uh, yes. I mean, the, the department at s- here at Stanford has a very long history. Um, it was actually one of the very first departments of radiation oncology, um, in the US. Radiation oncology grew out of radiology, which is, as, uh, your listeners probably know, is a diagnostic, uh, arm of, uh, radiology where you do imaging ma- mainly to, to diagnose, uh, disease. Um, and, um, initially back in the '50s, '40s, '50s, it was part of, uh, radiation oncology was part of those departments, but then here, our first chairman, Henry Kaplan, who is a very famous, uh, radiation oncologist and physician scientist, um, became the first chair of radiation oncology and sort of led the, the, the movement to develop it as its own specialty. Um, and then he quickly put the department on the map because, um, of his work in curing Hodgkin's disease with, uh, radiotherapy, which, um, you know, was, uh, this was, uh, in the, around the '50s where, um, you know, those were some of the first successes of taking patients, especially young patients with a Hodgkin's disease, which is a type of lymphoma that was incurable pr- previously, who now, uh, you know, the majority could be cured with radiation. Made the New York Times, you know, at the time, and all, you know, it was headlines, and at the time, you know, there were articles about how, oh, now radiation will cure all cancer and look, we're on the path. Uh, you know, that of course didn't turn out to be that way. We still have, uh, uh, many, many patients we can't cure. Um, but it sort of was the, one of the f-first examples of this, the hype that you get at each time a new therapy comes where, you know, then think, "Oh, okay, now we've really solved it," but ultimately, you know, it gets more complicated. But that, of course, established the department as one of the very leading departments in the world, and it has continued that the department here throughout its history has, um, can, uh, had a very strong, uh, interest in laboratory-based research. This is all thanks to Henry Kaplan, who was doing laboratory research at the time also while seeing patients. And so there were already faculty that were, uh, lab- laboratory-based, even just PhDs who were fully laboratory-based, which wasn't the case in many radiation oncology departments. And so I really liked that here, that within radiation oncology, this was a place where I could see that the kind of research I wanted to do was valued. Um, and, uh, there was already, you know, mentors that had done it and successfully, and th- those things are important at, when you're a young, uh, uh, physician scientist to have mentors that can show you, uh, if you run into trouble, how to, how to overcome obstacles.

5. 18:50 – 26:47

   The motivation to develop liquid biopsies

   1. PAPeter Attia18:50

      At what point, um, in your evolution did the idea of liquid biopsies start to become of interest? Uh, obviously, that's something I really wanna talk about today because it's, it's something that I think people are just starting to hear about. I mean, it's something I've been following for about six years. Uh, obviously you've been following it for a lot longer than that. But I think even in the, uh, in the, in the sort of broader public eye the past year, I think a lot has happened to bring this to the fore. But, but I still think for many people it's still a bit of a black box. Um, so, so at, at what point, how many years ago did this sort of become something that landed on your screen as an area where you wanted to put focus?

   2. MDMax Diehn19:35

      Yes. Um, so as these things often happen, um, I did not start my lab with the idea of focusing on liquid biopsies. Um, I really, uh, got there by following the results we, we got from some experiments and, you know, I think one of the, uh, keys to being a successful scientist, whether you're a physician scientist or a, a laboratory, purely laboratory-based scientist, is following the data, following the data that you generate, um, uh, to where it leads you as opposed to fo- you know, saying, "I have to work in one area." So I was actually gonna work on a dif- different area initially. I started this, the whole, this line of work in my laboratory initially be- out of a, uh, clinic, a clinical need I saw, and that's in general how I run my laboratory. All, all the research projects we do, we start with a clinical need or a clinical, um, an area where there, it's suboptimal we're doing in the clinic, whether it's done for diagnostic purposes or treatment purpose or whatever, with the idea being that that, you know, that's of course one of the things that having, being a physician and taking care of patients, uh, is one of the things that I spent a lot of years learning about and also what I have insights into that others might not. And those, that that's one way I can, you know, be a useful contributor to the, to the field in general. And so we always start with clinical, areas of clinical need. And here, the, this, in this case, it was one of the main things I struggled with as a, a junior faculty member, which is that I, I decided to specialize in treating lung cancer clinically. So one day a week, once I was on faculty, I was doing clinic one day a week, and I saw lung cancer patients exclusively. And, um, I was very frustrated by the fact that, um, after I treated patients, so I treat a lot of early-stage lung cancer patients. These are patients that are stage one or two lung cancer, which means as far as we can tell, the lung cancer hasn't spread yet. It's still, it's just in, in the lung where it started and hasn't spread anywhere. And, uh, with h- targeted high-dose radiation, I can cure the majority of those patients. Um, but there's, let's say about 20, 25% of those patients who ultimately develop recurrence, uh, where the cancer comes back. Um, but at, after the, after I finish my treatment in that first, you know, follow-up visit, let's say three months after I do the radiation, I cannot, I could not tell who was gonna ultimately have the cancer come back and who had been cured. And so state-of-the-art at the time, and actually, you know, is still state-of-the-art today in many ways, is just to see the patient every three, six, 12 months as you get further out and to do i- scans and see if the cancer has come back without any way of predicting in whom it will, right? It's really just, it's a very reactive approach and, and very unsatisfying, and this is something we went through with my dad also. So this w- was a major frustration for me.

   3. PAPeter Attia22:11

      A- a- and also, and also very difficult because of the resolution, right? I mean, like, ma- give people a sense of the resolution we have with traditional imaging. So if, if, I assume, or s- where, where are we at on CT scanners? Are we at 256-bit? Are we-- What, what's the, what's the resolution and speed of a CT scanner today?

   4. MDMax Diehn22:32

      There is a wide variety of that and that, that, that you may be better, uh, ask somebody who, uh, works in that specific field of building those scanners. But the, I think, I, maybe, uh, to rephrase your question a little bit is, you know, what's, how small of a tumor can we see?

   5. PAPeter Attia22:46

      Yeah, th- that's where I'm really going, which is, you know, it, it, it's, if you can see a one by one by one centimeter or a one centimeter diameter tumor, that's, you're doing pretty well, right? You can probably see a little smaller than that.

   6. MDMax Diehn22:59

      That's exactly right. So it is hard to see things much smaller than, uh, you know, one, uh, centimeter in diameter, maybe eight millimeters. It depends a little on the location in the body.

   7. PAPeter Attia23:08

      Right.

   8. MDMax Diehn23:08

      Some areas, of course, it's easier to see smaller things than others, but that's in generally true.

   9. PAPeter Attia23:12

      And so how many cells is that?

   10. MDMax Diehn23:15

       Yes, so that's great question, and, uh, you know, the general sort of rule of thumb is that's about a billion cells already.

   11. PAPeter Attia23:21

       Hmm. That's great.

   12. MDMax Diehn23:21

       Um, which is, of course, a vast number of cancer cells, you know, again, at the time when we can just barely detect it in a patient.

   13. PAPeter Attia23:29

       A- a- and, and I think it's important for patients to understand the non-linearity of this. So you have everything shy of a billion cells is by conventional detection methods undetectable, right? And then, but a bill- but, you know, outside of the most rare tumors, anything at a centimeter is not posing a direct threat to the organism. You know, maybe a really badly placed brain tumor at one centimeter is problematic. But for the most part, if you talk about a lung cancer, liver cancer, pancreatic cancer, colon cancer, et cetera, breast cancer, one centimeter is irrelevant. At what point does a tumor become in and of itself threatening to the host? I mean, you could argue by the time it's 10 centimeters by 10 centimeters by 10 centimeters, the tum- the burden of tumor itself is fatal And that's how many cells there roughly when you're at the point where the burden of the tumor is actually fatal?

   14. MDMax Diehn24:28

       Yeah. So the, of course, whether a tumor is fatal or not, as you point out, greatly depends on the location, right?

   15. PAPeter Attia24:34

       Sure.

   16. MDMax Diehn24:34

       Um, and, um, [lips smack] in certain areas, one can have a ton of tumor without it being fatal. In other areas, you have very little, uh, room for that. But, um, you know, uh, obviously it's, uh, it's a vol- it's a volumetric, right? So, uh, it grows by the cube, um, uh, by the, uh, radius multiplied by itself three times. And that, uh, is, uh, of course, nonlinear. It's g- It grows much faster. The, the number of cells, uh, in a mass are, are much more, you know, than just the linear increase in the diameter of the lesion.

   17. PAPeter Attia25:05

       Yeah, and I think that's the part that's just hard to understand for-- 'cause it, we don't th- we think linearly pretty well. Uh, we don't think exponentially very well. Um, but again, I think I want people to just anchor to this idea that from zero to a billion cells, we're pretty much flying blind.

   18. MDMax Diehn25:22

       That's exactly right. And it's actually, um, and relevant to, to the work, some of the work that we're now doing. Um, if you th- you realize that many cancers, once they have metastasized or spread, which is really ultimately what ends up, uh, killing patients is once the, usually when the cancer spreads. If the cancer is only localized, usually a surgeon or a radiation oncologist can cure that patient. But once the cancer's spread and is in many places in the body, you can't remove it all or radiate it all, right? So but if you think about, uh, a patient who has what's called micrometastatic disease or microscopic disease that has spread elsewhere in the body, you could have m- dozens or even hundreds of these micrometastatic deposits-

   19. PAPeter Attia26:01

       Right, that are under-

   20. MDMax Diehn26:02

       ... in different organs

   21. PAPeter Attia26:02

       ... under half a centimeter.

   22. MDMax Diehn26:04

       Un-- Yes.

   23. PAPeter Attia26:05

       Yeah.

   24. MDMax Diehn26:05

       Exactly.

   25. PAPeter Attia26:06

       A- a- and so you technically-

   26. MDMax Diehn26:07

       Still millions and millions

   27. PAPeter Attia26:07

       ... have tens of billions of cells, and it's still-

   28. MDMax Diehn26:09

       Yes

   29. PAPeter Attia26:09

       ... undetectable. Yep.

   30. MDMax Diehn26:11

       And it's still, exactly right. And th- that is a critical issue where that, that's a l- a limitation, a blind spot of our imaging methods because again, we, we can't, the instance they're all spread out, we can't see them all. Um, but that is one thing that, um, that, um, one could potentially get a handle on if one had a, a test that could, uh, be measured, let's say, in the blood, where one could measure contributions from all these, you know, dozens of micrometastatic deposits, um, and that that one might get m- uh, higher sensitivity. And that exactly kind of was the motivation for the liquid biopsy work in, in my lab. I had no idea how I was gonna do that when I started it.

6. 26:47 – 34:49

   History of blood-based cancer screening and understanding the predictive value of tests

   1. PAPeter Attia26:47

      N- now let's hit pause for one second, Max, 'cause I think it-- Let's also give people a little bit of the historical context of where there are at least a couple of areas where this was sort of being done, although it wasn't great. So maybe explain for people how PSA, CEA, and CA 19-9 have been used historically. Because even when we were in medical school, and, and bef- long before you and I got to medical school, those three biomarkers were used. Now, you were always cautioned that at least the CEA and the CA 19-9 could never be used for screening, but they could be appropriate for following. PSA has a very complicated history, which we could go down the rabbit hole of about how it could be used for screening, but, you know, you can be misled. But tell folks a little bit about what those things are and how effective you think they were for what they were trying to do and w- which elements of those you wanted to replicate for lung cancer and where you wanted to improve upon it.

   2. MDMax Diehn27:47

      Yeah. Great question, and for, uh, ab- you're absolutely right. Uh, when we start thinking about, you know, a bl- developing a blood-based test for initially lung cancer, um, we, of course, thought about the exactly examples you pointed out, which are the, the markers you mentioned like PSA, CA 19-9, are protein biomarkers. So these are proteins made by the cancer cells that can be shed into the blood and that where one can then, uh, measure them, uh, in, in a non-invasive fashion. Now, the main, uh, or a large issue with, with these protein biomarkers is that they are actually not that specific for cancer, meaning that these are also proteins that normal cells can make. N- now, cancers often tend to make more of them, which is why they are potentially of interest, but the problem becomes that having low levels of any of those markers does-- You c- uh, when you, when a patient has that, you can't know whether that means they have a little bit of cancer in the body or whether it's just that's their baseline, what the normal cells are making in their body. And that lack of specificity is what it's called, meaning you're not so sure when you see it that it's cancer or not cancer, was the major Achilles' heel of the protein-based biomarkers. And a lot of work, of course, had been done in lung cancer previously looking at similar biomarkers, um, including CA actually, which sometimes w- in some subset of patients can be a good biomarker if the tumor just pumps out a, a lot of it and the patient has a decent amount of disease, so the total level is significantly higher than what a normal patient would have. But in most patients, that was not the case. And so that was, you know, the protein-based approaches were really state-of-the-art, um, but had this major weakness of a lack of specificity.

   3. PAPeter Attia29:24

      Let's maybe use this as a moment to explain to people sensitivity and specificity, which are going to become very important as we then factor in prevalence when it comes to screening to understand positive and negative predictive value. But I think for now, if we can just start with helping people understand what specificity and sensitivity are in terms of probabilities of false positive and false negatives. Again, I th-

   4. MDMax Diehn29:50

      Yeah

   5. PAPeter Attia29:50

      ... I find this stuff to be so important, and yet if you don't, I think, explain it over and over again, it's easy for people to kinda get lost in the details. And yet it's so important for what we're about to talk about today that we'll probably end up doing this twice, three times.

   6. MDMax Diehn30:06

      [chuckles] Yes. No, and I agree with you. That, these are im- very important, uh, topics, words that are thrown around a lot and, and can be, uh, can be confusing. S- um, so, uh, sensitivity, um, is, uh, something that's also known as or a synonym for it is true positive rate, which is basically the, uh, likelihood of, uh, having a positive test when the patient has the actual condition. So let's say in the example of cancer, sensitivity is out of you have 100 cancer patients, you have a new blood test, out of those 100 cancer patients, you know, how many of them is the blood test positive? So that's sensitivity.

   7. PAPeter Attia30:45

      Right. So a perfect test would obviously have 100% sensitivity, which means 100 out of 100 would test positive. In reality, to give people a sense of this and, you know, a mammography might have an 85% sensitivity, which means if you do a mammogram on 100 women who are known to have breast cancer, it might only tell you that 85 of those 100 do. So it's correctly identifying 85. It's erroneously-- it's giving you a false negative in 15 of the 100.

   8. MDMax Diehn31:17

      That's exactly right. Yep. Um, and in general, it's important to realize that there are no perfect tests really, um, particularly when you start, uh, pushing the envelope to seeing-- trying to see smaller and smaller tumors. So while it would of course be great to have a test that's 100% sensitive, meaning it catches every cancer patient, that is really re-unrealistic in the vast majority, if not all cases. And I think this is something that's sometimes not understood by patients where, you know, you, you have a-- you go get your mammogram or your, your CAT scan for lung cancer screening and, you know, you still-- patients sometimes still develop a cancer even though they did all those things. And, you know, why is that? And that can lead to a lot of frustration, but that's because these tests, they aren't perfect and they can't-- they're not always gonna pick up a cancer even when it's present.

   9. PAPeter Attia32:02

      My f- one of my former analysts, Bob Kaplan, came up with a great thought experiment to explain this to people, which was the reason you can't have perfection is if you drive something to a very, very, very high sensitivity, the specificity must go down. And here's the silly example, and this is his example, so I don't wanna take any credit for it, but it's brilliant. If you wrote a letter, Max, to 1,000 women randomly and told every one of them they had breast cancer with no additional knowledge, you technically have a test with 100% sensitivity because let's assume, I don't know, 10-- let's assume 50 of those women have breast cancer. You have correctly identified them all. There are no false negatives in that group, so you have a 100% sensitive test. The problem is your specificity is in the toilet. You have so many false positives that the test is utterly useless. So you can have 100% sensitivity if you really wanna push that envelope, but it, it has no clinical utility. So you have to m-- and of course, I can play the experiment the other way. You could send 1,000 women a letter that says, "You absolutely do not have breast cancer," and technically you have no false positives, but you clearly are going to have false negatives. So I love that experiment 'cause it explains that if you, if you pull full throttle on sensitivity or specificity, that's fine, but it doesn't make for a good test because you've undoubtedly compromised the other parameter.

   10. MDMax Diehn33:38

       Yep, that's absolutely right, and th-that's exactly-- and I think I agree with you. That's a great example, um, to, uh, to explain h-uh, how that works. And, uh, actually that explains a lot of what happens in the, in the, in the research, the diagnostic research field in general, um, that, you know, you try to push sensitivity, but that never is meaningful to report if you don't also report the specificity to look at specificity because it's kind of like a yin and yang. If you, you push on one, the other, you know, usually one goes up, most of the time the other goes down. Um, that, that is, uh, in general how it is unless you really have a dramatic new insight or a new advance where you can just increase sensitivity without also hurting specificity. And then I guess just to finish off-

   11. PAPeter Attia34:20

       Yeah, just to finish up specificity, yeah

   12. MDMax Diehn34:21

       ... specificity, we haven't defined it. Yeah. So that is sort of the inverse of sensitivity, which is that for a patient who does not have cancer, so a neg-- uh, really is a true negative, what we, they call it, and doesn't have cancer, the test reports that, that the patient doesn't have cancer. So it's sort of the exact inverse of sensitivity, but for the patients who don't have cancer. And as, as you mentioned, um, they are connected and, um, you can, uh, game it. You can always push the sensitivity higher, uh, if you kind of ignore the specificity.

7. 34:49 – 46:46

   Current state of lung cancer and the need for better screening

   1. PAPeter Attia34:50

      So now let's go back to your clinical problem. Uh, by the way, Max, w-- uh, you know, we've talked a little bit about lung cancer on this podcast, but I, I can never resist the opportunity given the prevalence of lung cancer and the, you know, what I refer to as the loss given default, meaning the severity of the disease. Do you wanna give people just a sense of where we are today in 2022 with lung cancer? Is it the first or second leading cause of cancer deaths for men and women?

   2. MDMax Diehn35:19

      It's the number one cause of cancer death, which, uh, I guess I think a lot of people don't, uh, realize because it doesn't get nearly as much attention, um, at least historically as some of the other cancers that are, uh, maybe more common, especially if you just look just in men or just in women like breast or prostate. But where actually the outcomes are much better. You know, we can cure a, uh, a very large fraction of patients with breast and prostate cancer. Of course, not everyone, and th- and that is a critical area where we still need to make more advances. But the difference with lung cancer is that the vast majority of patients, you know, historically can't be cured, and so that's why it is the number one, uh, cause of, uh, of death, of cancer-related death. So it is a major problem.

   3. PAPeter Attia36:04

      Is that going down or, or going up or flatline? I mean, 'cause smoking rates are coming down and about 85% of lung cancer by incidence, I believe, is, um, small cell or, or sorry, is smoking related, so some combination of smoking related, uh, 15% is in non-smokers. So wh-where do we-- what's the trend line on, on, um, lung cancer mortality?

   4. MDMax Diehn36:29

      Yeah. So that's one of the good news, um, uh, in, in oncology in general is both the incidence of lung cancer has, has been going down in both men and women. Uh, it, um, initially, uh, lagged in women, but it's also coming down in women now, and this is all related to, as you mentioned, the, um, the uptake and then now decrease of smoking, uh, in the population. So that means because there's less smoking and because smoking is such a strong risk factor for developing lung cancer, um, it, uh... there's less lung cancer overall. Um, on top of that, our treatments have gotten much better, uh, and, uh, as has now slowly the advent of screening. We have a way of screening for lung cancer, which is relatively recent development. So those things combined have also decreased the number of patients who die from lung cancer once they get it. Even with all that said, it still is the number one cause-

   5. PAPeter Attia37:22

      Yeah

   6. MDMax Diehn37:22

      ... of, uh, cancer deaths. But the trajectory is, is good. It's starting to head this way, and of course, we're very eager to accelerate that slope to make it, uh, make it even less of a, of a, of a problem, a health problem, uh, ca- or a killer of patients. Um, it-- I think it is important, I want to point out one other point, which is often comes up in these kind of discussions, which is that, um, while smoking is the, you know, by, by and far, uh, far away the largest risk factor, it is not the only risk factor, and really anybody who has lungs can get lung cancer. Um, and the, uh, you know, that can happen for reasons that related to, um, pollution, you know, and that has been well described that, um, uh, pollution or other things in the environment like radon gas that, you know, people are exposed to, um, can cause lung cancer, and there's genetic associations also. Um, for example, uh, in the Asian populations, there's a much larger, um, incidence of, uh, non-smoking individuals who get lung cancer, particularly females. And so, um, we are never gonna get rid of lung cancer, uh, just by-- even if we could magically make smoking go away today, you know, across the globe and, you know, make all cigarettes disappear and never make a new one, um, there will still be lung cancer.

   7. PAPeter Attia38:36

      So, uh, let's go over kind of the distribution of how these things work out. Um, you have small cell and non-small cell is the big division, correct? And then within non-small cell, you have adenocarcinoma, you have squamous cell carcinoma, large cell, and carcinoid. Is-- Are those the big ones on the large cell or in the non-small cell? I'm sorry.

   8. MDMax Diehn38:58

      That's right. Yeah, I mean, carcinoid is not considered a non-small cell, uh, lung cancer. It's, it's own category.

   9. PAPeter Attia39:03

      It's its own thing. Okay

   10. MDMax Diehn39:04

       But, um, but, but you're right. It is one of the, um... It's, it's not a small cell.

   11. PAPeter Attia39:07

       Yeah.

   12. MDMax Diehn39:07

       So that, that, that's true still. Um, but that's right. So, and the dominant, the most common lung cancer is the non-small cell subtype, of which the, uh, the majority have adenocarcinoma, but also a significant fraction have squamous cell carcinoma.

   13. PAPeter Attia39:22

       And when a non-smoker gets lung cancer, it's usually adenocarcinoma. Is that correct?

   14. MDMax Diehn39:30

       That's correct. Mo- the vast majority of the time. It can sometimes be, you know, uh, squamous cell carcinoma or even occasionally small cell, but that is extremely rare.

   15. PAPeter Attia39:38

       And as you mentioned, being a woman, being Asian seem to elevate risk. Uh, on the, on the, on the, um, in the case of women, it seems to be potentially related to differences in estrogen, testosterone. That's been proposed as one plausible explanation. Uh, do we know what genes might account for this in Asians versus non?

   16. MDMax Diehn40:00

       Uh, we, we do not, and that's sort of one of the big mysteries, um, that we have not identified the main genetic drivers. It-- There've been-- N-And that's not for lack of trying, of course.

   17. PAPeter Attia40:11

       Hmm.

   18. MDMax Diehn40:12

       Um, so, um, it su-strongly suggests that it's multifactorial, right? That it's one of these, um, as, as a, as are many, of course, of our chronic, uh, illnesses that have genetic components that have, uh, like our-- the cardiovascular diseases, for example, right? Where it's not about a single gene the vast majority of the time. It is contributions probably from, uh, multiple, uh, uh, genetic variants that in aggregate elevate the risk. And then there may also be, of course, on top of that then some further, um, environmental factors that aren't smoking that, that may then interact with that genetic background.

   19. PAPeter Attia40:47

       Am I correct in saying, I think you mentioned radon, we talked about that is definitely a big, uh, environmental exposure. What do we know about the sort of PM2.5 exposure and, you know, uh, I've certainly seen data that's, that show that people who are in a... And just for folks listening, so, so, you know, particulate matter less than two point five micron is small enough to make its way into the most distal part of the lung. Um, there's clearly an increase in all-cause mortality associated with higher, uh, PM2.5 exposure. Do we know if that's also true specifically for lung cancer? Is that something that we can peg to an increase in lung cancer?

   20. MDMax Diehn41:25

       Yeah, that's a great question. I mean, there, there are, yeah, associations with that. So, you know, um, people have looked at individuals living in cities versus rural areas and, you know, and correlating with the particulate, um, contamination. Uh, and that is also associated with more lung cancer in, in, in, in the cities that have more of the, the particulate, um, uh, in the ambient environment. So there's definitely an association, um, in that regard. You know, as with epidemiologic studies, uh, to really prove it in a human is difficult, of course.

   21. PAPeter Attia41:54

       Yeah.

   22. MDMax Diehn41:54

       Um, but, um, but it does seem quite strong, and there's some biological rationale for that, where that if that is causing irritation and chronic inflammation, those kind of things, of course, we know that those things are as-- can be associated with developing cancer in, uh, many organs. So it all makes sense, and I, I think it's likely true. But, um, there also are, of course, chemicals in, you know, in smog that, you know, sort of along the lines of what's in to-tobacco smoke, uh, similar class of, uh, of chemicals that, that can be direct carcinogens.

   23. PAPeter Attia42:23

       Do we have a sense of what the dose-response curve is in s- in pack years of tobacco to, uh, relative risk increase in lung cancer? So You know, does a person who smoked a pack a day for 10 years, so they have a 10-pack year history, um, relative to a non-smoker, is that 2X the risk, 50% more risk? I mean, do we have a sense of how nonlinear that curve is as well versus, you know, the 40, 40-pack year smoker, the five-pack year smoker, et cetera?

   24. MDMax Diehn42:56

       Right. Um, there are associations between the number of pack years and, uh, the risk of developing lung cancer. Um, now a pack year is, is, uh, it's not an ideal clinical variable because, uh, uh, of course you can't measure it in a, uh, in a completely unbiased way, the way it just so your patients understand. But when we... How we figure out, you know, how a doctor tries to figure out how many pack years a patient has smoked, meaning we basically multiply the number of years of smoking, how many years did they smoke for, by the number of packs per day that they smoked. Well, how do we get that information? Well, we have to ask the patient, and the reality is, uh, number one, patients smoke different amounts of ti- uh, right over their, over their time, and number two, you know, they don't always remember perfectly. So it's, it's not a great, um, uh, metric in that regard. So the, the... It's difficult to really pin it down, uh, perfectly. I have seen studies that have suggest- for sure suggest that there is an, you know, increase when you go-- as you go from zero to, to higher levels. But then around 20 or 30 pack years, some studies argue that it then kind of plateaus, that you kind of, you may be sort of saturating. But, um, it, it, they, um... It's, that's not exactly my expertise, so I don't have the exact, you know, the slopes of those lines I don't know off the top of my head. But, um, but it is not a perfect variable, of course, as, as it-

   25. PAPeter Attia44:11

       What do the data say about secondhand smoke, which I, I would imagine is even harder to quantify than pack years? But when, you know, for example, I have some patients who grew up with parents who smoked. They themselves have never smoked. Um, and let's say those parents continued to smoke all the way until those kids, now adults, went to college. Do we have a sense of how to quantify what their, their smoke exposure is, and should they be treated as former smokers in terms of cancer screening, for example, with low-dose CT and things like that?

   26. MDMax Diehn44:39

       Yeah, very interesting question. So yeah, I agree. Uh, it's very difficult to measure. It's even harder, harder to measure than the pack years we'd, uh-

   27. PAPeter Attia44:47

       [chuckles]

   28. MDMax Diehn44:47

       ... we just talked about. Um, and you know, we know that cer- you know, certain professions that, that were exposed to a lot of smoking like, you know, waitresses and waiters or stewards and stewardesses, that, that there was increases, you know, epidemiologically. Of course, there's always other things that go along with that, with that is-issue also. So, um, I think the link is quite clear how, you know, how to quantitate it and how to know, you know, uh, has one had enough secondhand smoke exposure that one's risk is significantly elevated. I think that's a major problem. Actually, in a way, it would be very useful to have a, a biomarker that could be quantitatively measured to actually, like, measure, like a, a clock of, of amount of, uh, exposure you had. We don't have such a thing. Um, and currently, that is not one of the, um, uh, criteria f- that would get a pa- a patient eligible for, uh, uh, lung cancer screening by low-dose CT, so secondhand smoke is not considered. Um, the way that we decide on, you know, those criteria for who, who should get this test and covered by insurance and who shouldn't is really about trying to enrich for a s- you know, the highest risk population. You know, you want at least, you know, a certain percent, half a percent, a percent of risk of developing, uh, lung, uh, the cancer that you're gonna screen for 'cause otherwise you will screen lots and lots of patients who will never get it, and then the specificity becomes a problem we talked about earlier. Uh, and so that's... Even though there are things like, you know, uh, exposure to pollutants, secondhand smoke that we know increase the risk, we don't-- then it can be frustrating enough for patients who are, who are in those categories to, um, to know whether, you know, why, why can't they get access to screening? But it, there are, you know, sort of public health reasons for that.

   29. PAPeter Attia46:22

       Yeah, which we'll then get to something we'll talk about in a little bit, which is you can know the sensitivity and specificity of a test, but you then need to know the prevalence or the pretest probability to know how to interpret the result. You, without the prevalence, you can't impute the positive and negative predictive value, which is what tells you when you have a positive, how confident are you it's positive, and similarly, when you have a negative, how confident

8. 46:46 – 53:21

   Low-dose CT scans: an important tool for managing lung cancer but with limitations

   1. PAPeter Attia46:46

      are you there. Let's talk a little bit about low-dose CT because, um, I would say this has been certainly in the last 10 years, one of the major changes in how we manage lung cancer. It's been about 10 years, correct, since, uh, since the data have, have, have made the case for, for the use of that in, as a screening technique?

   2. MDMax Diehn47:03

      That's right. Yeah. So that was, um, you know, there'd been a, a long effort at trying to develop a screening test for lung cancer given that it is the number one cause of cancer deaths. Um, and there were a lot of failures along that road. A lot of in- initial studies focused on doing chest X-rays when the, when the, when that was the main thing that was available, so a much lower resolution way of imaging the lungs, um, where those studies were unable to show a benefit. But then there was a landmark study called the National Lung Screening Trial, um, that, um, proved that, uh, if you take, uh, the... In that trial, they took, uh, patients and randomized them either to get a low-dose CT scan, which is now a, a more high-resolution way of looking at the lungs that can see smaller nodules, uh, versus getting a chest X-ray, which was this technique that previously already we had kind of known that wasn't, didn't work. And if you do that, uh, the patients who get the low-dose CT scan, the higher resolution imaging, ha- uh, in that, in that group, there was significantly lower, uh, rates of, um, lung cancer deaths, you know, a relative risk reduction of about 20%, um, that, um, sugge- you know, argued that that's a major win for a screening test to, to actually be able to decrease the number of deaths from the disease you're screening for.

   3. PAPeter Attia48:10

      Do you remember what the absolute risk reduction was?

   4. MDMax Diehn48:13

      That was small. It was in the single-digit percents. I don't remember it off the top of my head.

   5. PAPeter Attia48:18

      But it was single-digit percents, so you, you would still have an... Let's say it was 5%. Your NNT would be 20. So you would, you could save a life for every 20 people. That still seems pretty good.

   6. MDMax Diehn48:30

      It's, uh, it's actually lower than that. Yeah, no, um-

   7. PAPeter Attia48:33

      So maybe it was like 1% to 2%

   8. MDMax Diehn48:34

      ... it is, uh, the number needed to treat is... Yeah, it's in that. Yeah, it's in the hundred. Uh, the hundred. I think I'd say one or 200, so maybe it's .5% to 1% based on that-

   9. PAPeter Attia48:40

      Okay. Okay. Yeah

   10. MDMax Diehn48:40

       ... uh, ar-argument, yeah.

   11. PAPeter Attia48:41

       Yeah.

   12. MDMax Diehn48:41

       Now the, um, that gets complicated also because, um- Uh, you'd think doing a CT scan, uh, it... And then you either see a nodule or you don't, so it should be pretty easy. But of course, it's not that easy, as CT scan's a complicated thing to read. And so how you interpret the scan and what you consider a positive will, of course, a- affect your, um, sensitivity, uh, and specificity and will affect the number needed to treat down- downstream. So, um, the, you know, the way that we now read scans, the low-dose CT scans, is different than what the, the initial study did, um, because of a high risk of false positive in if-- the way that they was doing the original study. So, um, the, it's not, you know, the, the low-dose CT is not my area of expertise, um, so that I don't have all the numbers at my fingertips, but those are the sort of salient issues that, that are in that field. So even though we have this test, uh, and it is great, um, and is this major home run for screening 'cause we can save a lot, you know, cancer deaths, um, it's not a perfect test.

   13. PAPeter Attia49:41

       And do, do you know by any chance how many millisieverts of radiation the low-dose CT provides? I'm, I'm guessing it's like s- sha- you know, one millisievert or less maybe.

   14. MDMax Diehn49:51

       I think that's right. Yeah, I don't know that off the top of my head, again, 'cause I, I don't do those procedures myself. But it's dr- significantly less than the, the scans that we routinely do for patients who already have cancer.

   15. PAPeter Attia50:00

       Yeah.

   16. MDMax Diehn50:01

       And this is an important concern that patients have, is that, well, you know, radiation can cause cancer, so why, um, should I, uh, you know... Is, is the risk worth it, right? Um, the risk of, of, of the, the lower risk of cancer. But it-- the, it's called low-dose CT because the amount of radiation used is much, much, much less than was historically done for a CT scan, and so therefore the risk of causing a cancer is, is much, much lower.

   17. PAPeter Attia50:25

       You know, the NRC says that we shouldn't be exposed to more than 50 millisieverts in a year. W- w- w- you know, is, is that sort of like saying, you know, you, you, you shouldn't drink more than, you know, 10 drinks in a day? Like, do you have a sense of, you know, what would be your personal threshold for how much radiation you would wanna receive a year from imaging? 'Cause you're gonna get radiation from, you know, being on airplanes, although that's relatively small. You know, just being at sea level is probably one to two millisieverts a year. If you live in Colorado, you probably double that. Um, but do you ever-- whe- when do you start to think about how much radiation a patient is getting from a, from a practical standpoint? You know, do you start to worry at 20 millisieverts in a year? Uh, which is obviously pretty easy to do with a PET CT scan, for example.

   18. MDMax Diehn51:15

       Right. Um, so y- you know, we don't routinely track that at the individual patient level, and, um, I think the, you know, the way in more it's tracked in a less, you know, I guess, systematic way in that, you know, we don't do imaging tests unless we feel that the, the benefit of doing that test outweighs the known but really low and actually very difficult to quantitate harm, right? Uh, 'cause, uh, you know, it's actually very difficult to quantitate exactly how much risk of, let's say, a second malignancy there i-- a future malignancy there is from a given, uh, uh, protocol of a CT scan or whatnot, just 'cause the, to get that data and, uh, the number of patients you'd need-

   19. PAPeter Attia51:57

       Yeah

   20. MDMax Diehn51:57

       ... you could never do it randomized, all those things, right? Um, and so, you know, the, the way that medical practitioners who order imaging studies, which of course most or many fields, in many fields they do, not just radiology, um, uh, is the, the, the calculus is always, do I need this test, right? Will this, uh, help me ma- uh, uh, help me manage the patient in a way that's beneficial to the patient? We absolutely should not be doing imaging if there's no point. Like, if, if the, no matter what the scan shows, I'm not gonna change what I'm doing-

   21. PAPeter Attia52:26

       Right

   22. MDMax Diehn52:26

       ... then I should not be ordering that scan. And, uh, and there are, that does come up quite frequently that there are, you know, you're sort of, uh, the question comes up, should I image, should I not? And, you know, the way I and, you know, and all of us, you know, in this, in these areas think about it is, is it worth, uh, doing? And so, you know, some patients who have lung cancer, particularly when they get radiation therapy, which is very high doses of radiation, they have astronomical amounts of radiation that if you were not a cancer patient, you would never want those amounts of radiation.

   23. PAPeter Attia52:52

       Yeah.

   24. MDMax Diehn52:52

       But the vast majority of those patients will not get a cancer from their imaging or radiation treatment, and their cancer, if we leave it untreated, will kill them, right? And so, uh, or if we don't catch, if it's gonna come back, if we don't catch it early where we still might have a chance to, a second chance at cure. Um, so that's how we, you know, we do the calculus. It's not that there's, you know, badges, ba- badges that, that patients wear that, um, that me- that track this so that, um, that, uh, that and if you get to a certain point, you couldn't do no more imaging.

   25. PAPeter Attia53:20

       Yeah.

   26. MDMax Diehn53:20

       It's always a sort of a case-by-case

9. 53:21 – 1:12:25

   Using liquid biopsies to identify circulating tumor cells

   1. MDMax Diehn53:21

      basis.

   2. PAPeter Attia53:22

      So let's go back now to the, the clinical problem that is at the, at the root of this whole situation, which is you have a patient who has either had a resection and/or radiation to follow. They're clinically, as we would say, NED, no evidence of disease, but you know that the recurrence rate is there. It's, let's say it's twenty-five percent. Actuarially, twenty-five percent of these pat-- of these hundred people who are currently NED are going to have a recurrence. And we know that the sooner we catch the recurrence, the better the odds are for treatment. The lower the tumor burden, the lower the burden of mutation in the patient, the greater the odds of therapy. So if this was two hundred years ago, we'd be hosed, right? Because we'd have to wait until, you know, they had foliating, fungating masses. They were s- coughing up blood. I mean, we'd be, obviously that would be crazy. And we've already now just put in, you know, great detail that look, even giving high-resolution CT scans, at best we have to wait until a billion cells are cancerous. That's the best case scenario. So now we wanna talk about something else. So, so walk us down the path of how you would or how you did think about going after another type of biopsy, a liquid biopsy.

   3. MDMax Diehn54:45

      As I mentioned, you know, I was frustrated by not being able to diagnose a recurrence earlier, and, uh, that was an unmet need, so I wanted to start, you know, a part of my laboratory working on that problem. But to be, to do, be able to do that You know, there's two ways you could think about doing that. You could start with mouse models to build mouse models, let's say, or, or you know, what they're called preclinical models, in which you try to, you know, grow tumors in the animals and then see if you could, you know, have an i- you know, hypothesis for what, what might be a good biomarker, try to test that and, and, uh, see if it works in a mouse, and then ultimately go back to the human. And as I mentioned earlier, sort of the approach that we take in, in my group is more, um, very tied to the clinic, very what's called translational research. And so for us, really the, the human being is the model organism. It's the final model, of course, it's the model we, we wanna get the answer for. Um, and so I thought, let's-- we should do this work directly on blood samples because if we find something, then it likely will be directly applicable. Otherwise, you'd, you'd always still have that second step where it works in the mouse, let's say, but you don't know. So, so I started just by collecting blood samples. I actually didn't know what I was gonna do with them.

   4. PAPeter Attia55:48

      And, and by the way, who was funding this work at the time, Max? Was this something you went to the NIH and said, "You know, here, I want an RO1 and then this is why I wanna study it"? Or, or at the outset, who, who was the, uh, who was, who was bearing the risk financially?

   5. MDMax Diehn56:01

      At the, at the outset, the, the-- it was basically me. It's, uh, when you, um, when you're a new faculty member, when you start a new lab, when you get that, that first job, you get what's called a, a startup-- you get startup funds, which is basically funds your department or university gives you, provides you sort of as a to kickstart you because you don't have grants at that moment. You haven't, you're just starting. You haven't had a chance to write them. Um, and so I was using that, that money basically. So I was footing the bill on, on sort of that, that, that pot of money I got to start my lab. Um, because as you know, and I'm sure your listeners have heard it many times that, um, one of the things with academic research and, and funding is that, uh, oftentimes you kind of have to show it already works before you can get funding for it, and that's a very common story because, you know, grant funding is limited and so, uh, the, the organizations that fund research are often quite conservative that if the ideas sounds great, but you have no proof that it works at all, you know, uh, they are not gonna give you money. So I, I used money that I had from my startup funds. Um, so, um, so yeah. So I just started collecting blood and then, um, started thinking about what biomarkers might one-- would be good. So I read the literature on the protein biomarkers as we talked earlier.

   6. PAPeter Attia57:10

      And, and sorry, did you, um, collect blood in patients prior to resection and post-resection or just post-resection?

   7. MDMax Diehn57:17

      Both. Always at, you know, prior, prior to resection, uh, uh, or radiation, whatever the treatment was, and then, uh, in, in subsequent follow-up s- visits, usually, you know, at the time of their follow-up scans when they're coming back to see us. Um-

   8. PAPeter Attia57:31

      Yeah

   9. MDMax Diehn57:32

      ... and, um, so the, uh, the protein biomarkers didn't seem very useful. I then spent maybe about a year, uh, looking at, uh, uh, something called circulating tumor cells, which is one subtype of what this field called liquid biopsy, um, which is actually looking for intact cancer cells that can circulate in the blood of patients. Um, [lips smack] the-- I tried that with a couple different, uh, techniques, um, and, uh, s- quickly realized, you know, something that many others in the field have also observed, which is that while if one could really do that, it would be super powerful, m- it is very complicated to actually measure those cells, uh, for a variety of reasons. Uh, the two, two biggest ones probably are, uh, or maybe three biggest ones are there, uh, the cells are not very abundant when they are there. There's very few of them. And they're-- so they're very difficult to purify, like, you know, to actually, um, puri- you know, a good purification, they might still only be one percent or less of the cells in your purified sample because it's just so hard to purify. It's finding a, you know, needle in a haystack. And, uh, because it's, uh, intact cells, you have to process the samples really quickly. You have to process them basically the same day or within an hour or two of, of the patients having the blood sample drawn. That makes it very difficult to build up biobanks of frozen cells, um, of frozen samples that you could, you know, get a large cohort you could actually, uh, study because you have to process them immediately. So, uh, and then the last thing was that in-- we did some control experiments. This is critical obviously. You always wanna, you know, ask, is the thing I'm working on really working? And where we drew blood from healthy individuals who don't have cancer, you know, uh, and some of these, the tumor, certain tumor cell, uh, methods we're using found circulating cells, um, but those patients don't have cancer. So they were picking up something else, so there was this lack of specificity. So they were technically complicated. Seemed like it's gonna be hard to get this into the clinic anytime soon, and there was some weaknesses. So that's then when I went back and tried to really see what else could we do.

   10. PAPeter Attia59:30

       So let me go back to the first one. Let's just talk about the protein 'cause that's makes sense as the obvious place to start because we already kind of do that with PSA, notwithstanding the limitation that normal cells, normal prostate cells, non-cancerous cells make prostate-specific antigen. Same with CEA, CA19-9. Did you find any proteins that were made, that were expressed by lung cancer cells that were unique to them?

   11. MDMax Diehn59:56

       We at the time did not. Um, the, um... We did not do a holistic screen ourselves, um, which, uh, but other people had, um, which, you know, one can do, but there's techniques now, uh, called, uh, mass spectrometry where one can take a sample of proteins, let's say purified from the blood, run them through this machine, and it basically tells you what proteins were there. Um, and those, uh, studies had not found convincing, um, markers that were unique to lung cancer cells. Now, they found, um, some markers that in some patients were elevated. Actually, CEA is one of them that can be, uh, elevated in lung cancer patients, and in ones where it is, it can even be a marker of recurrence like PSA can be in prostate cancer. But we, we know- we already knew that CEA has problems with specificity, right? So-

   12. PAPeter Attia1:00:42

       Yep

   13. MDMax Diehn1:00:42

       ... um, so that, that, that was sort of the, uh, the main concern. Um, so we didn't actually do, uh... But the main protein work we did was to actually I drew some CEA and a few of these other markers like CA19-9, some of the other ones that had been previously reported to be markers in lung cancers, uh, in some of my patients because there were some studies that claimed really great performance, uh, in the literature. But quickly realized they, those studies were, you know, maybe it was an artifact of that cohort or something else was going on where, 'cause we could not reproduce those results.

   14. PAPeter Attia1:01:12

       Yeah. Uh, l- l- let's talk about that a little bit, Max, 'cause that's, I think what people don't understand is how difficult it can be to reproduce results. And so, so like help people understand what that process is like. You or one of your graduate students or one of your postdocs reads a paper where somebody says, "Hey, we've identified this protein. It seems really specific. It looks fantastic," et cetera, et cetera. Presumably these people are collaborative. You can read the paper and get the methods. You can call the lab. They can tell you how they did it. Now you do the experiment. What do you find?

   15. MDMax Diehn1:01:48

       In many cases, uh, unfortunately one finds that one can't reproduce the result. Um, that often, m- uh, that can be not that it doesn't work at all, but that just it doesn't work as well as it did in the initial, uh, study. Um, and, um, that, but it works, you know, so much less well that it's not gonna be clinically useful. That's a very common, uh, event.

   16. PAPeter Attia1:02:13

       Now, do you still publish that result, Max?

   17. MDMax Diehn1:02:17

       Uh, the va- the lack of validation-

   18. PAPeter Attia1:02:19

       Yes

   19. MDMax Diehn1:02:19

       ... you mean? Yeah. Um, sometimes. It depends. Yeah, sometimes we do. Um, often people don't. Um, the reason, uh, is complex, of course, but one, one, uh, issue is that, um, you know, to definitively prove something doesn't work quite as well as, as it w- was initially reported, one actually has to do very large studies so that the, what are called error bars or our estimate of the, uh-

   20. PAPeter Attia1:02:46

       Yeah

   21. MDMax Diehn1:02:47

       ... the, let's say, sensitivity, that's what we care about, is very accurate, right? Um, but even though by doing a smaller cohort, you can already tell it's not gonna work out, but to really ultimately prove it to the level of rigor that one, you know, would need for a publication would require another year of, you know, and thousands of dollars, right? And so there is these, you know, decisions one has to make is, is this really worth it and what's the opportunity cost, right? Um, now-

   22. PAPeter Attia1:03:12

       But it, but it's interesting, right, when you think about it because it means that we have a bias system in the publication, uh, realm that it, we are biased towards positive findings that are often untrue because let's say one person publishes that they can make this thing work, okay. And again, we're gonna f- throw out fraud or anything like that, but we're just gonna assume-

   23. MDMax Diehn1:03:36

       Correct. No fraud. Yep

   24. PAPeter Attia1:03:36

       ... that in good faith, these are people who did something, but they had an artifact. They didn't catch it. The experiment looked like it was positive. They publish it. Let's assume that you and four other labs independently try to reproduce it, and none of you can reproduce it, but you all come to the same conclusion, which is, "Well, we only did it this one way and it didn't work. Technically, I would need to do it three or four other ways to be really sure. I'm gonna put those resources of time and money elsewhere and away I go." And everybody comes to the same logical conclusion. But of course, that means that that one finding gets disproportionately propagated as a positive finding, when in reality four labs have now found it to be highly unlikely, and that doesn't get published. Of course, I'm not being critical of you or anybody, Max, other than just the system, which is I wish there was a way that the incentive structure could be such that we prioritize negative findings as much as positive findings.

   25. MDMax Diehn1:04:34

       Right. I, and I agree with you 100%, um, and, um, y- as you laid it out, uh, is true, and that does happen. And in general, you know, the, we don't have a system where one lab could know that three other labs had done the same thing, you know? Maybe eventually you find out over dinner at a conference or something, right?

   26. PAPeter Attia1:04:53

       Yeah.

   27. MDMax Diehn1:04:53

       But, um, but how would... Because there is no, um, place to go to look up to see what other people have done, um, the, uh, there is no way to know, right? Uh, so, um, it is, it is a problem. I, I, I, it would be great if we could fix it. It's a very difficult problem, of course, because of the, the, the many factors inc- including, including the economic parts of limited research w- uh, resources and time, and, like, even, you know, to publish such a study would take, let's say, a grad student's, could still take six months of their time to g- get that out. But now, you know, they're not spending those six months on, on, on their main project, which is, you know, to develop the new test, right? So it, there are so many, uh, reasons why, why this happens.

   28. PAPeter Attia1:05:35

       Right. And, and, and I've spoken with graduate students who will say, "Look, it's really gonna be harder for me to get my PhD publishing negative studies. I mean, I'm gonna quickly turn my attention to where I think my committee is going to be looking at me more favorably." So yeah, I think the, I think the problem is pretty significant, but-

   29. MDMax Diehn1:05:52

       There's a selection bias for it. It, it, the s- selection bias of the positive result is a major w- issue, and so this, this is why re- reproducing positive results is so critical, uh, and something that also not everyone likes to do it because reproducing something someone else already did is not making a brand-new discovery, right? But that is critical. That is, you know, and that's one thing I would say particularly in the, in more in the medical fields we do do quite a bit of, right? There is quite a bit of studies that sort of try to reproduce, um, results, positive results, and then publish it when the results are also positive, and that's really to really prove something works, w- you know, to convince convincingly, uh, convince yourself that something is really real. It's really helpful to see multiple studies from multiple groups-

   30. PAPeter Attia1:06:33

       Yep
10. 1:12:25 – 1:21:41

    Liquid biopsy research moves from circulating tumor cells to cell-free DNA

    1. MDMax Diehn1:12:25

       clinically.

    2. PAPeter Attia1:12:26

       All right. So then what, where, where would you go? Where'd you go next? Did you look at cell-free DNA?

    3. MDMax Diehn1:12:31

       Yes. So the next, after that, we went to what's called cell-free DNA, which refers to DNA molecules that are found in the circulation, but outside of cells. So they're circulating in what's called the blood plasma, which is the, the non-cellular liquid portion of the blood. Um, and, uh, it's called cell-free 'cause it's outside of cells, free of cells. Um, of course, it comes from cells originally, but it, it's now circulating by itself without membranes around it. Um, and the reason I got interested in that was actually largely from reading about a different field, which is, um, prenatal diagnostics, where around this time there were, uh, there was a lot of work being done led by a number of individuals like Dennis Lo, uh, in Hong Kong, and, uh, Steve Quake, who was actually here at Stanford in the engineering department, um, basically showing that in pregnant mothers, you, uh, c- can detect DNA from the fetus in the woman's blood, cell-free DNA, again, not, not inside cells, but outside of cells. So It's pretty obvious that, I mean, but I wasn't the only person who had this thought that if, if we can do that in pregnancy, then, you know, couldn't we try to do the same thing in cancer patients?

    4. PAPeter Attia1:13:41

       And, and just to be clear, Max, when you spin this down, when you, when you take a, a tube of blood and you spin it, you get all of the, uh, cellular matter below the buffy coat, and then the plasma is above. Is the cell-free DNA in the clear plasma and not stuck-

    5. MDMax Diehn1:13:56

       That's right

    6. PAPeter Attia1:13:57

       ... within the cellular material? It manages to-

    7. MDMax Diehn1:13:59

       That-

    8. PAPeter Attia1:14:00

       Yeah.

    9. MDMax Diehn1:14:01

       Correct. So the, the, the cell-free DNA is, uh, in the plasma, and that's exactly where we isolate it. Uh, and we-- that's actually a critical technical aspect-

    10. PAPeter Attia1:14:08

        Yeah, because if it were in the cellular matter, it would be a disaster to find it

    11. MDMax Diehn1:14:13

        You can't get it, right?

    12. PAPeter Attia1:14:14

        Yeah.

    13. MDMax Diehn1:14:14

        And, and it actually, I-- we've thought about this idea. The, I think it's actually quite insightful idea you have there. Could there be more cell-free DNA mixed in with the cellular compartment? And I think it's quite likely that there could be-

    14. PAPeter Attia1:14:23

        Right, if, if you could put a gradient on that and try to somehow-

    15. MDMax Diehn1:14:26

        Yes

    16. PAPeter Attia1:14:26

        ... extract it, yeah, yeah.

    17. MDMax Diehn1:14:27

        Get it out, that you might be able to get it out, but-

    18. PAPeter Attia1:14:29

        And you could enrich your cell-free-

    19. MDMax Diehn1:14:30

        But-

    20. PAPeter Attia1:14:30

        Yeah

    21. MDMax Diehn1:14:31

        ... you could enrich for it again. Uh, and that might actually be a future interesting research direction. But the, um, the problem is currently if you were to just take some of that cellular compartment, there's so m- vastly more DNA in the cells than there is in the cell-free space, um, that, um, you, uh, you would basically lose the signal. You would just have all this DNA.

    22. PAPeter Attia1:14:50

        Yeah.

    23. MDMax Diehn1:14:51

        The vast, vast majority is from the cells in that compartment. So we, we take the, the, the aqueous part, the, the plasma part at the top of the tube after you spin it, and there's very low levels. You know, it's only about, um, it's on in a healthy individual, about one to five nanograms per mil, uh, of, of cell-free DNA in, in the plasma.

    24. PAPeter Attia1:15:09

        By the way, you know what's interesting? I have beta thalassemia minor, so, uh, my phenotype is normal. Uh, well, I shouldn't say that. My hemoglobin, hematocrit are normal, but my phenotype is that I have very small RBCs. So my-

    25. MDMax Diehn1:15:22

        Hmm

    26. PAPeter Attia1:15:22

        ... my MCV is, you know, in the 50s or 60s or something like that, right? And my RBC count-

    27. MDMax Diehn1:15:28

        Mm-hmm

    28. PAPeter Attia1:15:28

        ... is very high, so that's how I have a normal hematocrit because I have a lot-

    29. MDMax Diehn1:15:32

        Yeah

    30. PAPeter Attia1:15:32

        ... of small red blood cells.
11. 1:21:41 – 1:34:59

    Zeroing-in on circulating tumor DNA in cell-free DNA

    1. PAPeter Attia1:21:41

       Okay, so let's say we're talking about a normal person. You've taken 10 cc of blood. You, you take the plasma from them. How do you, uh, quantify how much cell-free DNA you would expect to get out of that? And then furthermore, how are you distinguishing what is potentially cell-free DNA from a normal cell, which presumably is the bulk of it, versus cell-free DNA that is from the cell of interest, the cancer cell?

    2. MDMax Diehn1:22:07

       Right. Um, so once we purify the DNA, once we have the plasma, so the non-cellular com- compartment, there is, um, laboratory procedures for isolating DNA that have been, you know, well worked out for over many decades that we use in all different fields of biology. And so we apply some methods there that have been optimized for these low concentrations of DNA. That's one of the unique things about the cell-free DNA work. The concentration is really low, like I said, the, you know, one single-digit nanograms per mil. Um, and often when we do laboratory procedures, we like to work with at least micrograms of DNA. So this nanograms is, it's a small amount, and that was one of the challenges early on in the field to try to overcome that. Um, so we purify the DNA and then we can quantify how much there is. There's a variety of laboratory tools we have. Um, the most commonly used one in our lab is a, um, fluorescent-based method where you basically add a fluorescent dye to, to the solution. It, it binds to DNA and that, uh, gives a, a specific fluorescence, and then you can read that out in a, in a machine that can read that fluorescence. So you can then have a s- and have a standard curve and read off how much DNA you have and again, like a, like what would I expect for a healthy individual.

    3. PAPeter Attia1:23:11

       So in, in situ hybridization, or?

    4. MDMax Diehn1:23:14

       No, it's, uh, this is just purely a dye-based method.

    5. PAPeter Attia1:23:16

       Okay.

    6. MDMax Diehn1:23:16

       It's, it's much simpler than that. You could do, you could do some-- You could do more complicated things. You know, probably the next most, the more complicated thing that we would do next, when we do sometimes in certain situations, is qua- what's called quantitative, uh, PCR.

    7. PAPeter Attia1:23:28

       Mm-hmm.

    8. MDMax Diehn1:23:29

       Where basically you, uh, you use a, a enzymatic method that can amplify DNA, um, and you, again, have a standard curve where you know how much DNA is in there, and you compare how the amplification curves grow, and that, that can let you read out how much DNA was in your sample. But most, most of the time we can use this more simple method, which is just add a dye, put it in the machine, get your result. It's, it's very fast.

    9. PAPeter Attia1:23:49

       And, and just give me some quantity. So in 10 ml of plasma, how many nanograms would you expect? Or how many-

    10. MDMax Diehn1:23:58

        Yeah

    11. PAPeter Attia1:23:58

        ... what's, what, how many nanograms per milliliter in any sample would you expect to, to have of cell-free DNA?

    12. MDMax Diehn1:24:03

        Right. So 10 mls of plasma would be 20 mls of blood we'd collect 'cause the plasma-

    13. PAPeter Attia1:24:08

        Yep

    14. MDMax Diehn1:24:08

        ... in most individuals is about half, right? So the, from 20 mls of blood, we get 10 mls of plasma. Those 10 mls in healthy individuals would have somewhere between one to five nanograms per mil, so that would mean 10 to 50 nanograms of DNA at the end. And but in some conditions, in some, uh, advanced cancer patients, patients who have trauma, have an infection, the levels can be much higher-

    15. PAPeter Attia1:24:31

        Yeah

    16. MDMax Diehn1:24:31

        ... much higher than five nanograms per mil because there's active cell death happening in the body. And so when that happens, we, we sometimes see levels, you know, in the, you know, hundreds of nanograms per mil. That's rare, but we-- it can happen. Um...

    17. PAPeter Attia1:24:44

        So now the big question is: How do you separate the good actor and the bad actor in that pile?

    18. MDMax Diehn1:24:52

        The way that, uh, we and others in the field do that is to focus on a, a unique molecular property of cancer cells that normal cells generally don't have, and that is the mutations that cause the cancer. Of course, as, uh, many of your listeners will know, uh, cancer is a, is a disease that is caused by mutational mistakes in the DNA of a normal cell that accumulate and ultimately lead to that cell not being responsive to cues to stop growing or to, or to, to kill itself and just to grow uncontrollably. And so then, um, those mutations, both the ones that cause the cancer as well as many thousands that are along for the ride in most tumors, um, are ... are serve as really exquisitely specific markers of the cancer cell because these mutations are really only present in the cancer cells, not in the patient's normal cells. So this is a really attractive biomarker and, uh, for a variety of reasons. One is you're tracking the cause of the disease, right? You're really at the molecular cause of disease. It's a mutation. That is what you're tracking. What better biomarker can there be for an illness, right? You're actually, uh, tracking that. And the second thing is the specificity, particularly when you compare it to the proteins, where, as we mentioned, like PSA can be made by normal prostate cells and by prostate cancer cells, but the mutations in a prostate cancer would not be present in the normal prostate cells. They would be present in the cancer cells. So how do we tell then if there's circulating tumor DNA present, uh, is that we, um, look for mutations, uh, in these short DNA molecules. Mu- so we look for mutant molecules. Uh, most of the molecules in the blood come from the healthy cells of the patient, and they don't have mutations. But a small subset, and that can be, you know, range from, you know, an advanced, an advanced lung cancer patient might have 1% of the circulating DNA is from the-

    19. PAPeter Attia1:26:40

        Wow

    20. MDMax Diehn1:26:41

        ... is from the cancer. So still a small amount. It's still, you know, relatively small. That's, and that's high levels. In early-stage patients, it's often well less than, you know, .01 or .001% of the DNA that, that, that's from the cancer. So-

    21. PAPeter Attia1:26:54

        So how, how do you even detect that without... I mean, h- do you have the resolution to detect .1% of the cell-free fraction, which itself is already very small, and know that you're not seeing an artifact?

    22. MDMax Diehn1:27:10

        Yeah, so th- and that's, and that h- was the challenge, uh, in, in, you know, in, in the field when we started is that's not trivial. It's actually extremely difficult, a- as, as you suggest. Um, the reason is that, um, uh, the methods we use to try to detect mutations, and there's a variety of them, um, uh, but the one that we chose to use in which the, basically the, the vast majority of the field has moved to since as well is next-generation sequencing, which is these high-throughput molecular methods for sequencing DNA, um, where you can identify the actual sequence of the bases, the AGTC, uh, bases that make up the DNA. Uh, you can get the exact sequence for hundreds of millions of molecules or billions of molecules even in parallel in one experiment. Um, so we use those tools because we can just read the d- it's like we could, like we have a-- It's as if we had a microscope that's so good that we can just read off the bases. It's not a microscope. It's a complex molecular biology procedure, but the end result is the same. We get, we get a data set of millions of DNA sequences that are all about 170 bases in length, and we can then compare those sequences to the patient's germline, uh, DNA, let's say, that we purify from a buccal swab or from the cellular compartment, compartment of the blood. And we can say the vast majority of the molecules look identical to the patient's germline, meaning their, their normal healthy cell DNA. But some small proportion of this, of the molecules have mutations. And in many of our experiments, the way that we can be very certain that's the case is we, 'cause we start by actually sequencing the tumor. So let's say particularly for this initial problem we had of telling who is cured or not, um, if you sequence... Those patients all had treatment. They're all known to have cancer. So we have a piece of their tumor. We can sequence that and identify what mutations that tumor has compared to the patient's healthy cells. Let's say we find, you know, in a, in a particular panel we might be using, let's say we find 10 just for the sake of example, so 10 mutations the patient's tumor has that the healthy cells don't. We can then go look in the blood in the sequencing data set we have for those mu- mutant molecules that carry tho- one of those 10 mutations. And that turns out to be exquisitely specific and sensitive for the presence of the cancer. If we see even one or a handful of molecules that carry the mutation, one of these mutations, has borne out that that basically means there's still cancer cells in the body.

    23. PAPeter Attia1:29:25

        Now, the other thing that I wanna make sure I understand here is we're talking about relatively short fragments of DNA. You said a hund- call it 150 bases, correct?

    24. MDMax Diehn1:29:36

        Uh, yeah. I'd say 170 maybe. Close to 170.

    25. PAPeter Attia1:29:38

        99.9% of all of that cell-free DNA lines up to some segment of their germline. But no two of those are the same either, right? So it's like if you could lay their entire germline genome out, it would, you know, span miles and miles and miles. You probably have no two segments that are the exact same. So you have a whole bunch of 170-base segments that line up somewhere on that problem. That's a pretty interesting computational problem to me because 170 is not that big. Presumably-

    26. MDMax Diehn1:30:13

        That's true

    27. PAPeter Attia1:30:14

        ... some of that 170 also is present in the cancerous, in the mutated-

    28. MDMax Diehn1:30:20

        Yeah, the vast majority. The vast majority, right? The, a c- a cancer-

    29. PAPeter Attia1:30:23

        A cancer is only gonna have maybe 100 mutations.

    30. MDMax Diehn1:30:27

        Yeah, so 100 mutation in the coding regions, exactly.
12. 1:34:59 – 1:37:41

    Cell-free RNA and Max’s vision for cancer detection from a blood sample

    1. PAPeter Attia1:34:59

       So, so let's now talk about, is there such a thing as cell-free RNA, or is that just so unstable that it's not even worth thinking about? And is, is cell-free DNA the only place to be thinking about, uh, this?

    2. MDMax Diehn1:35:13

       No, absolutely not. There, there is such a thing as cell-free RNA. That field is much m- uh, more nascent than the, than the DNA field.

    3. PAPeter Attia1:35:21

       What advantages would it offer given the instability of the molecule?

    4. MDMax Diehn1:35:26

       I think there are some critical questions initially about what is the stability of the molecule, you know, how much of it is there, uh, and how stable is it.

    5. PAPeter Attia1:35:33

       And then you miss the non-coding mutations as well, right?

    6. MDMax Diehn1:35:37

       Well, yeah. You would-- There you can-- If, if you're looking for mutations, I don't think it has advantages-

    7. PAPeter Attia1:35:41

       Yeah

    8. MDMax Diehn1:35:41

       ... really. Um, it's-- But it, you can do other things, right? So, like, the vi-- My vision is that ultimately we wanna get to the point where we can, from a blood draw, determine as much as possible about a patient's cancer, ideally, eventually to the point where we don't even need to biopsy it. Now, that is science fiction, uh, currently, right? Um, but if you ex-extend the line of work, uh, you know, decades into the future, that is where we're trying to get. Uh, and, um, that means we have to be able-- If we wanna get there, we have to be able to, uh, measure things other than mutations. Mutations are important, they're critical, but they're only a small part of the puzzle. Uh, measuring RNA could tell you about which genes are on and off in the cancer-

    9. PAPeter Attia1:36:26

       Yeah

    10. MDMax Diehn1:36:26

        ... theoretically. And so that is critically important. For example, for in the immunotherapy field that you res-you previously-

    11. PAPeter Attia1:36:32

        Yeah

    12. MDMax Diehn1:36:32

        ... mentioned. You know, we know there's some markers like, uh, PD-L1 that can be expressed in the, uh, tumor cells that can basically hide the cancer from the immune system. That marker is, that's not a mutational process in the vast majority of patients. It is, it's expressed, uh, from the promoter, like, you know, just turned on for sig- you know, reasons having to do with signaling, epigenetic reasons, et cetera. And so you can't tell by looking at the sequence in DNA of PD-L1 whether the tumor is, has this marker. It's not a mutation marker. So we'd like to be able to measure something like that from the blood. We'd like to be able to tell the difference between an adenocarcinoma and a squamous cell carcinoma or between a lung cancer and a breast cancer from the blood. Um, and, uh, some of those things you can do with DNA potentially, but a lot of them you can't.

    13. PAPeter Attia1:37:17

        Mm-hmm.

    14. MDMax Diehn1:37:18

        Uh, and they might be better done with RNA. So I-- It's absolutely, it's a fascinating area. My group is working in this area. Um, we haven't, uh, published, uh, our work, but, um, we will hopefully soon. Uh, we have some exciting preliminary results. I think, I think in the future, RNA will be A big part of the liquid biopsy puzzle. That doesn't replace DNA, it just, it has, it offers a complementary, um, pieces of, of information that

13. 1:37:41 – 1:40:15

    Methylation patterns and other informative signatures found in DNA

    1. MDMax Diehn1:37:41

       you can't get from DNA.

    2. PAPeter Attia1:37:42

       Now, is there a difference between cell-free DNA and circulating tumor DNA? Sometimes I've heard people use those interchangeably. Are they, are they interchangeable-

    3. MDMax Diehn1:37:49

       Good question

    4. PAPeter Attia1:37:49

       ... or is there a distinction?

    5. MDMax Diehn1:37:50

       They are different. Generally, cell-free DNA refers to the total amount, the total DNA in the circulation. That includes both the healthy cell D- derived DNA and the cancer cell-derived DNA. And so circulating tumor DNA-

    6. PAPeter Attia1:38:01

       And circulating tumor is just the subfraction of that that's the tumor

    7. MDMax Diehn1:38:04

       Just the portion.

    8. PAPeter Attia1:38:05

       Yeah.

    9. MDMax Diehn1:38:05

       That 1% that's from the, the, or whatever that's from the tumor, exactly. So, um, that, that, that, that's the difference in the terminology.

    10. PAPeter Attia1:38:12

        And then are there any other signatures that can be helpful here, uh, for example, looking at methylation patterns on the DNA? Can that be informative at all?

    11. MDMax Diehn1:38:22

        Absolutely. Yep. So meth- DNA methylation, um, is, uh, refers to a chemical modification of the DNA molecules. So ag- again, we have these about 170 base pair molecules in the circulation, and at certain of the, the, b- uh, nucleotides, there can be chemical modifications that are put on by enzymes, particularly a methylation, uh, a methyl group, which is a chemo- chemical group. Um, and that, uh, is put on not the same in every cell. Different cells based on u- largely what tissue they're from, have different patterns of this methylation, and it has to do because these methylation marks, these chemical modifications can, uh, influence which genes are turned on and off. So you could envision that in a lung cancer that needs, you know, the genes to, uh, or the lung cell that needs genes for surfactant production or making alveoli or whatever, um, needs those genes on, but it doesn't need the genes on that, you know, of a, of a, uh, adipose, a fat cell, to make fat 'cause it's not doing that. So they might be methylated, uh, some, some, some genes are methylated and turned on or off in different, uh, cells in different ways. So they can be m- exquisite marks of the, uh, origin of the DNA. And so there's a l- uh, been a lot of, uh, work in the last couple years looking at, uh, methylation marks, um, to, to do that. Um, and it has some potential advantages because, uh, unlike mutations, um, you know, they're mu- it's actually much more stereotyped. So a lung, the methylation profile of a lung cancer, of two lung cancers is more similar than their mutation profile.

    12. PAPeter Attia1:39:52

        So in other words, you can use the mutations to identify and distinguish between tumor and non-tumor, and you use the methylation to hopefully zero in on tissue of origin.

    13. MDMax Diehn1:40:02

        For, you could, for example, yes, exactly.

    14. PAPeter Attia1:40:04

        If, if you were using it as a pan screen.

    15. MDMax Diehn1:40:06

        Or maybe you, uh, for screening purposes, exactly, which is a lit- it's a different, uh, scenario than this initial scenario of saying, "I have a patient who I've treated with surgery or radiation. I

14. 1:40:15 – 1:45:27

    Mutation-based methods of liquid biopsies

    1. MDMax Diehn1:40:15

       wanna know if they're cured." Now, I think an important, uh, thing to recognize is that the sensitivity of the methylation-based approaches is significantly inferior to that of the mutation-based approaches. Um, which, but from a practical standpoint, the methylation approaches have some advantages for the screening question, but the, the sensitivity and the practicality aren't aligned 'cause you'd, uh, really [laughs] you'd want the most sensitive assay possible for screening, right?

    2. PAPeter Attia1:40:40

       Yes.

    3. MDMax Diehn1:40:41

       Um, 'cause the tumors are tiny. But that is not the current state in the field, is that the mu- mutation-based methods are much more sensitive. So we recently published a paper, a, a new meth- a new method that's bit of our third generation mutation-based method, that's about 100 times more sensitive than our prior methods as well as other, you know, methods in the field. We c- we, that can now get down to, uh, one part in a million. Uh, so we can detect, you know, maybe where the prior methods were kind of bottoming out at about .01%, the, the prior mutation-based method around .01%, so one in 10,000. This new method can get to one in a million. So ex- you know, that's, you know-

    4. PAPeter Attia1:41:22

       So a two log improvement

    5. MDMax Diehn1:41:23

       ... two orders of magnitude.

    6. PAPeter Attia1:41:24

       Yeah.

    7. MDMax Diehn1:41:24

       Two log improvement. So that's, you know, a log improvement is a dramatic, of course, uh, improvement in a diagnostic assay. And so, um, with that assay, we have, you know, now with lots of patients where we had sample left over so we could use our first gen assay and now come to this new assay, and we had false negatives with the first gen assay that now we can pick up with this newer method saying that it's not that they didn't have any, it wasn't that there was no ctDNA. There was circulating tumor DNA. It just was not at the level that the prior generation could detect. So that one in, you know, we're about a one in a million. With the methylation-based assays, it's actually the data is still not very mature to know exactly what their sensitivity is, but as the put, you know, putting data, the data that's been in the literature as well as some of our own experiments, it's probably closer to one, one in 1,000. So .1%.

    8. PAPeter Attia1:42:11

       Now, of course, to be clear, with the mutation, the advantage, what makes the mutation-based method so much more sensitive is you know what you're looking for. You have the mutation a priori. In the-

    9. MDMax Diehn1:42:24

       Correct

    10. PAPeter Attia1:42:24

        ... in the methylation method, you don't, right? So you, you, if you're doing a pan screen for someone who doesn't have cancer, by definition, you can't know what mutations to look for. Um-

    11. MDMax Diehn1:42:39

        That's right.

    12. PAPeter Attia1:42:39

        So how does that gap get closed? Because if you look at a test like GRAIL, for example. So GRAIL is, I believe, looking at exactly this method, right? They're using, uh, I think they're looking at methylation patterns of cell-free DNA. But they're doing it as a pan screen. So it's give me 10, 20 mL of blood, and we're just going to look for all the cell-free DNA. We're gonna identify if there's cancer or non-cancer in there. If there is cancer in there, we're gonna try to predict the tissue of origin.

    13. MDMax Diehn1:43:10

        Yep. Correct.

    14. PAPeter Attia1:43:10

        How, how good can that get? Because right now, the sensitivity for all stages is probably 50% s- with a specificity of about 99%. It's above 99%. But if you look at stage one, right, if you're really trying to catch an early cancer, we're talking about a sensitivity of maybe 20%, right?

    15. MDMax Diehn1:43:30

        Right. So th- th- this gets into a very interesting part of this, this f- the diagnostic field in general, and particularly this liquid biopsy field with regards to early detection, which is how results are Uh, are presented and you put your finger on it. We care about sensitivity and specificity. We want specificity very high, 95, 99%, something in that ballpark, and we want high sensitivity. Now, um, when, uh, studies report sensitivity across stages, I have an issue with the way that when that is done because it is v- you can easily bias that by just including more stage four patients, right?

    16. PAPeter Attia1:44:05

        Mm-hmm.

    17. MDMax Diehn1:44:05

        You can include one stage one patient and nine hundred ninety-nine stage four patients and say, "I have sensitivity of blah from stage one to four."

    18. PAPeter Attia1:44:12

        Yep.

    19. MDMax Diehn1:44:12

        Uh, so I, it's, I think what you did in this, as you laid it out, is it's important to break it down by stage, and actually even within stage one, it should be broken down because when you, for like a lung cancer, let's say, we break stage one into, uh, stage 1A, stage 1B, uh, and even stage 1A1, stage 1A2, stage one. So it gets very gr- uh, granular.

    20. PAPeter Attia1:44:30

        Yep.

    21. MDMax Diehn1:44:30

        Uh, and that all relates to smaller and smaller tumors that have better and better outcome that are the ones you wanna catch. And so in lung cancer, actually the sensitivity for stage one lung cancer, uh, uh, in the real data that's been presented, is actually when the, you know, 5% or less-

    22. PAPeter Attia1:44:46

        Okay. So let-

    23. MDMax Diehn1:44:47

        ... for stage one lung cancer

    24. PAPeter Attia1:44:47

        ... let's make sure people understand the math on that. If it's bigger than stage one, you don't need a liquid biopsy. You can do a low-dose CT and you'll see it, right? The, in theory, the only reason you'd care about a liquid biopsy is if we're talking about fewer than a billion cells.

    25. MDMax Diehn1:45:05

        That's true. I, I mean, if, if you, if you had a patient and you could do either test, that's true. There are some potential advantages of the blood-based test as a way of getting more patients screened. I think there are some practical reasons why you could argue, uh, let's say they were equ- um, you know, uh, say they were equivalent-

    26. PAPeter Attia1:45:20

        Yeah. If, if people have less access to care and, and things like that.

    27. MDMax Diehn1:45:23

        Access or they're-

    28. PAPeter Attia1:45:23

        Right. It's like, it's the advantage of-

    29. MDMax Diehn1:45:24

        ... they're hesitant to get the scan.

    30. PAPeter Attia1:45:25

        Yep, yeah. Fair enough. Okay.
15. 1:45:27 – 1:53:57

    Understanding the sensitivity and specificity of a diagnostic test

    1. PAPeter Attia1:45:27

       but let's just say, let's, I wanna make sure people understand the following. If a test has, and this is gonna be, I think, amazing for people, 50% sensitivity, 99.5% specificity, if you said that that was your stage one performance in an unbiased sample, I think people would be like, "Wow, that's pretty good." I mean, for people who don't have cancer, for, for, or who, for whom we believe don't have cancer and whom we don't have any mutational information, I have a test that is 50% sensitive and 99.5% specific. Do the test. Okay. Here's how I explain this to my patients. What is the pre-test probability that you have cancer? So if you're one of my patients and you're 45 years old and you're not a smoker and your family history is relatively normal, and oh, by the way, your colonoscopy is negative. Uh, in other words, your pre-test probability is 1%. So you go... And, and I encourage everybody to do this, and for the show notes, we're going to include my calculator. So I, I built a little calculator, but there are apps that do this. I just do it in Excel. Um, but you can download an app that will allow you to do this. You plug in sensitivity, specificity, and prevalence, and it will spit out positive and negative predictive value. Well, if you use the numbers I just gave you, [lip smacks] 55%, 99 and a half, s- uh, specificity, 1% on the prevalence, if my memory serves me correctly, your negative predictive value is like, [clears throat] you know, goes to 99.7% or something. Or it goes from, you know, let's just say you put in 99% specificity.

    2. MDMax Diehn1:47:15

       Yeah.

    3. PAPeter Attia1:47:15

       It goes to 99.5% negative predictive value. And they say-

    4. MDMax Diehn1:47:19

       Right

    5. PAPeter Attia1:47:19

       ... "Well, that's really good, isn't it?" And I say, "Yeah, but it's only a little bit better than what it was at the outset. Your pre-test probability was 99% negative. I just took you from 99 negative to 99.5 negative, and by the way, your positive predictive value is 40%."

    6. MDMax Diehn1:47:35

       Yep.

    7. PAPeter Attia1:47:35

       So, so if, if I get a positive test in you, it's more likely than not a false positive.

    8. MDMax Diehn1:47:41

       Yep.

    9. PAPeter Attia1:47:41

       And so [clears throat] I don't wanna be the guy that's raining on the parade because we are using these tests in our patients, but never in isolation. And I just, I guess the question, that's a, that's a lot I loaded into that, so feel free to comment on any and all of it, but where I wanna go with this, Max, is without the mutational information that makes the type of testing you're doing in patients with known cancers so sensitive, how can we make these things more than a parlor trick in, um, patients who don't have cancer? At, at least-

    10. MDMax Diehn1:48:16

        Yeah

    11. PAPeter Attia1:48:16

        ... at the surface.

    12. MDMax Diehn1:48:18

        Excellent point. Yeah, you, you, you, uh, laid out some things that I often do in my, my talks as I, as I talk, uh, about, about these things. Um, this is something that's even not well understood, not just by patients, but by a lot of, uh, researchers and physicians, this issue that, um, something that, you know, sounds great, 50% sensitivity, 99% specificity. Even ignoring for the fact that the stage one sensitivity was actually less than 5%, right? Just, let's just say on the surface the 50% where it's, where we're at. Um, that if you already basically have no risk of having cancer, you're, you know, a less than 1% chance of having cancer, the test isn't moving, really moving the needle for you significantly, and, uh, s- has a high risk of catching, um, false positives, as you mentioned, 'cause the positive predictive value isn't so good. Um, and that leads to lots of further testing, anxiety. Often we can't find anything, but then patients are worried for years that is there something brewing? Is it just the next scan will show it or the next time? And, um, so I have lots of concerns about that. I am, I'm one of the researchers that's really focusing on this field and, and I very much see these issues and, um, am concerned about them. Um, I, this is why I really strongly feel that we as a field need to ultimately really do studies that prove cancer-specific survival benefits of these tests, and that is currently not really on the roadmap. The, you know, the, the commercial efforts, uh, which there are, of which there are many uh, are largely n- they're not planning on doing this large, these large randomized trials like the, the lu- LDC, low-dose CT s- study we showed, we talked about earlier that proved that, the, that you save lives from lung cancer. Um, why not? Well, they're expensive and take years, right? And, um, so it's not attractive from an investment standpoint. But if we really wanna be sure that we are helping our patients and we're not just adding costs to the healthcare system, th- that's what we have to do. And it m- we don't know if the sensitivity as, of the best tests as they currently are, are good enough to save lives, and there's multiple reasons for that. But an important reason is that, uh, it gets very complicated, but even beyond ... Okay, we already said you gotta look at by stage 'cause by saying across all ca- stages doesn't, is meaningless because you can game that. Even within stage I, you gotta look by stage IA1, IA2, IA3. But even once you're at that point, it gets even more complicated because there's actually really ... Let's say we have stage I lung cancer patients. There's two types of stage I lung cancer patients. There's stage I lung cancer patients where the cancer cells have not left the lung. There's, that's where they are. That's the only place they are, and the surgeon removes the tumor, and the patient is cured. That's one type of stage I lung cancer patient. The second type of stage I lung cancer patient looks the same, has the exact same CT scan, has the same surgery, but already has microscopic cells in the liver and the brain that we don't know about. They are stage I. We call them stage I because that's all we can see, but ... So stage I is actually heterogeneous. There's two subtypes of it, and in order for a screening test to be useful, it has to catch a significant fraction of the, the first type of stage I, the stage I that doesn't have micrometastases because which the surgeon can cure.

    13. PAPeter Attia1:51:27

        Uh, because I think what's implicit in your comment, Max, is that the surgeon's not gonna cure the sec- the second one.

    14. MDMax Diehn1:51:32

        Good point. Yes. So the surgeon can cure what he can see or she can see, right? What, what they can cut out. But they, the, if there's micrometastatic metastases in the liver and the brain, the surgeon isn't cutting those areas because they don't know about them. Um, and so those will eventually grow back. And so that is why simply saying a test, um, increases the number of patients diagnosed at earlier stages doesn't automatically prove that that test will save lives because it, when you look at the very granular level, you, and you consider this issue about there there being different kinds of st- uh, stage I patients, um, you know, you can envision a test that mainly catches the patient with micrometastatic disease, and actually, we have hints that the l- the l- the ctDNA assays preferentially catch those patients because of the things y- we talked about at the beginning with there being now, you know, lots of microscopic deposits spread throughout the body. If you were to put those all into one place and add them up, you would have many billions of cells, therefore, overall, you have more d-, uh, circulating tumor DNA. But, and you can detect it, but you think they are stage I because that's all you can see on the scan. So how do we ever, how do we prove that the tests are not, you know, falling into that, uh, pitfall is we have to do randomized studies. That's the only way we can, we can prove it. And so that, that's something t- to me, that's a, a major concern. And there's this push and pull about, well, these are great tests. They're very promising, and obviously I wouldn't work on it if I didn't think ultimately these will save lives. But, um, so can we, do, should we withhold them from patients, right? If they are, you know, we have to wait.

    15. PAPeter Attia1:53:03

        From non-cancer patients especially. Yeah.

    16. MDMax Diehn1:53:05

        From non-cancer patient, yeah, non-cancer patient. Should we withhold ... If we have the test, okay, we don't know yet if they're gonna save lives, but isn't it unethical to not give them? Because, you know, if we later find out, if we 10 years from now or five years from now find out they save lives, now there's, uh, five years of patients who didn't have the benefit. That is an argument that's often put forward. You know, the problem with that argument, which is why ethically it sounds, of course, very reasonable, is that if by that argument, we, we would never test anything [laughs] new that we develop in medicine 'cause there's always a possibility that that thing could improve-

    17. PAPeter Attia1:53:32

        Yeah

    18. MDMax Diehn1:53:32

        ... right, things and-

    19. PAPeter Attia1:53:33

        Well, the, uh, and that's-

    20. MDMax Diehn1:53:34

        Yeah

    21. PAPeter Attia1:53:34

        ... I think why the FDA has taken the posture, because right now there are only four approved liquid biopsies, right? Uh, Guardant and three others.

    22. MDMax Diehn1:53:42

        Correct, and those-

    23. PAPeter Attia1:53:43

        Yeah.

    24. MDMax Diehn1:53:44

        Yeah. Guardant. There's some a- some, some academic, uh, assays as well, like the Morris Long can, getting path on. But those are for very different things, right? Those are for genotype. That's something we haven't really talked about. Those are for identifying what mutations are present in patients with advanced disease. That's what those tests

16. 1:53:57 – 2:01:22

    Existing clinical liquid biopsy tests and their limitations

    1. MDMax Diehn1:53:57

       are.

    2. PAPeter Attia1:53:57

       Yeah, so let, let's actually talk about that. Let, let's talk, w- tell, tell, tell folks what Guardant does and, and w- you know-

    3. MDMax Diehn1:54:02

       Yeah

    4. PAPeter Attia1:54:02

       ... it's an FDA-approved, uh, diagnostic, but explain to people what it's used for 'cause I get asked this question all the time.

    5. MDMax Diehn1:54:08

       So the Guardant assay, uh, and they are one of the first companies in this field, they use a very similar method to what we use, a next-generation sequencing to measure circling tumor DNA. But the test that they, their initial test, the one that's approved, was developed for a very specific clinical problem. You have a patient with metastatic cancer, let's say lung cancer, you know, multiple s- parts of the body, and you n- want to identify what mutations that patient has because in lung cancer, we have s- drugs that work for patients with certain s-

    6. PAPeter Attia1:54:36

       Treatments based on mutations. Yeah.

    7. MDMax Diehn1:54:39

       Exactly. It's a- it's what we call actionable. If we know what the mutation is, and if we have a drug for it, that is the first thing we should treat the patient with. Historically, we always had to do a biopsy or a surgery to get the piece of the tissue to, to identify the mutations. What we've, we, you know, the field, Guardant, we others have shown is that you don't, m- many patients with advanced disease, because like, say 1% of the circling tumor DNA is from the cancer, it, that's enough that you can identify the mutations without needing to do the biopsy. So that's potentially very useful, particularly in patients who can't have a biopsy or where they had a biopsy, but the whole, all the tissue is used up. There's lots of cases like this. Um, and, um, or where the biopsy missed the tumor. That happens a l- lot, as you, as you know, um, or a, a certain subset of the time. Um, so, uh, in those cases it can be super useful to have a blood test to say, "Oh, sh- yeah, this patient does have an EGFR mutation, so they, they should get on the EGFR tyrosine kinase inhibitor." That's what those tests were developed for. Now, that is the easiest problem in the liquid biopsy space 'cause you're in stage IV patients that are known-

    8. PAPeter Attia1:55:37

       Yeah

    9. MDMax Diehn1:55:37

       ... to have cancer. They have lots of disease in the-

    10. PAPeter Attia1:55:38

        High tumor burden. Yep.

    11. MDMax Diehn1:55:39

        High tumor burden. And so those tests are optimized for that. They work well for that. They have good agreement, 80, a high 80% agreement with biopsy. They can sometimes find mutations that the biopsy misses 'cause of technical issues with the biopsy, as I mentioned. So they have... They're a good test for that, for that problem. They are not designed for, number one, early cancer screening or detection, where the levels of circulating tumor DNA are many, uh, logs lower, and they're not designed for this question of is the patient cured or not after treatment, again, where the levels are much lower. They are designed for this one thing and, um, and that's very, yeah, important to recognize. So we-- You can't-- That, that test, if, you know, if you were to go to the doctor and say, you know, uh, I, I want a test to see whether I'm cured or not, that, those FDA-approved tests don't do that.

    12. PAPeter Attia1:56:26

        Is that true of, is it CellSearch and FoundationOne? There, there are a couple of the others that are approved, right?

    13. MDMax Diehn1:56:31

        Correct. Yeah. Uh, yeah, Foundation has a liquid biopsy test, um, that's, uh, basically very similar. Again, technology similar to what we developed in Guardant and they do the same thing now. Uh, CellSearch is actually a circulating tumor cell assay, uh, that's, is, is FDA approved. Actually was, that was approved before any circulating tumor DNA assay and, you know, we had talked about circulating tumor cells at the beginning. The, they, at the very beginning of the liquid biopsy field, like in the early 2000s, that, that was the focus. They were, everyone's focusing on the cells, and so that's why that was the first test. It is FDA approved. Uh, no one really uses it because it's not, doesn't provide you actionable information. Your doctor runs that test, y- whatever the result is doesn't impact how the patient is me- treated, and so therefore, that has grown out of favor. There's, and, and there is a difference between FDA approval and actually clinical usefulness that, that is important to recognize.

    14. PAPeter Attia1:57:21

        Now, when you look at a test like GRAIL, for example, the FDA has not approved it yet, but it is still permitted in use. Explain that distinction.

    15. MDMax Diehn1:57:30

        Now, this is a complicated area in, in, uh, in the, uh, diagnostic space in the US where there's really two kinds of ways tests can, uh, be regulated by the government such that they can be used on patients. One is the one that's well known is FDA approval, which is of course a, an agency that focuses on develop, on approving drugs and, and diagnostic tests, um, and evaluates them carefully and then, uh, allows companies to give some approval to market them if, if they pass all their, the, the bar of, of what you need to show. The second way is actually a much easier way to get assays, uh, out to patients, and that is, uh, by setting up the assay in a lab that is compliant with what's ca- the CLIA, uh, Act, which is a la- uh, basically a la- an act of Congress that focuses on regulating laboratories that do diagnostic tests that is independent of the FDA. And those, the way that roughly works is that the lab itself gets this designation that, yes, you guys are doing things in an appropriate way. They get inspected every year or two and, uh, uh, to, to, um, make sure they're doing things appropriately. And then those labs have the blessing to develop any test that they want, uh, using the procedures that they're supposed to follow with, you know, running control, those kind of things, and then offer them to patients. So the FDA never review the, reviews those. It's really a much less highly regulated because basically the individual tests aren't regulated. It's the laboratory that's regulated. Um, and that is what, you know, Guardant did before they got their FDA clearance. That's what GRAIL's doing. That's what all the diagnostic companies, uh, they start by building a CLIA laboratory and then building their assays in that frame because they're h- much less regulated, so you can start selling them to patients and providers, um, long before you'd ever get FDA approval.

    16. PAPeter Attia1:59:15

        So from a technology standpoint, I mean, there are like 30 other companies that are out there doing this right now, and not to use the word GRAIL again, but the holy grail is the blood test that's going to take a patient who doesn't think they have cancer, screen them with a high enough sensitivity and specificity that I do the blood test and it says, "Peter, we're afraid that you have, you know, colon cancer." "What? I, I had a colonoscopy two years ago." "Well, it looks like you have it again. Go get a colonoscopy next week." Lo and behold, there's a tiny little adenomatous polyp there that is an early stage I, and gosh, I need just the smallest little partial colectomy and I'm, I'm cured 'cause that is a 100% curable cancer. Or, you know, a woman gets a blood test and it says, "You have breast cancer." And she says, "That can't be. I don't feel a lump. My mammography was negative six months ago." "Understood. Maybe you should go get a diffusion-weighted image MRI and look at that breast a little more closely." And sure enough, there it is. What would really be amazing is if we had the confidence in these things that even if they were never showing up on imaging, you could treat them, right? I mean, that would be-- This is, now we're getting back into science fiction. What's science fiction? Science fiction is [clears throat] in a, in five years, you're going to have a one centimeter pancreatic adenocarcinoma. And you know, you and I, Max, both know that even a one centimeter pancreatic adenocarcinoma has a 25% five-year survival. I mean, that, that is a death sentence of a cancer, unfortunately, even at stage I. So, you know, waiting until you can see a billion cells on the CT or MRI is not gonna be the way to remedy pancreatic cancer. It's either going to be much earlier detection or treatments that actually work. So today we have neither of the above.

17. 2:01:22 – 2:10:48

    The future of liquid biopsies

    1. PAPeter Attia2:01:23

       Is there a path-

    2. MDMax Diehn2:01:24

       Yeah

    3. PAPeter Attia2:01:24

       ... to this sci-fi world that I'm dreaming of, where using the blood alone we can actually start to develop a high enough sense of confidence that that patient does indeed have cancer, but they might be three or four years away from it being clinically detectable?

    4. MDMax Diehn2:01:42

       I do think, I am hopeful that we could get there. Um, it's not gonna come in one step, but we are already taking steps towards that, so building, building the, you know, the bricks on top of each other. And the reason I say that is that the easier application again is in the patients who are already known to have cancer where we have these super sensitive tests that we can, uh, then run after and see who's cured or not. So that we now have. We, we have these tests that, that can de- de- detect the state that's called minimal residual disease, meaning microscopic cells that are residual after your treatment, and, uh, that have currently at least a h- very high positive predictive value, meaning that if you detect the ctDNA, the patient is very likely gonna have the cancer grow back.

    5. PAPeter Attia2:02:27

       And, and is that guiding therapy? Is that immediately now sending you to adjuvant therapy right away?

    6. MDMax Diehn2:02:32

       That's, that's exactly where I was going. Great, excellent point. So there's now studies, clinical trials underway to do that. So we've recently launched a couple here at Stanford in lung cancer where we're taking patients with early-stage lung cancer. They can have ... One of the trials, it's, uh, very broad. They can have radiation or surgery, uh, whatever is appropriate. If they need chemo for their standard of care for their stage, they get that too. When they're done with every standard thing, they get the blood test, and we do the, this minimal residual disease test. If it's positive, we s- give those patients immunotherapy. Before we can ... We can't see any cancer anywhere-

    7. PAPeter Attia2:03:08

       Yep

    8. MDMax Diehn2:03:08

       ... in the body, but we know they have microscopic cells l- left behind. And that is the moment in the patient's, uh, l- uh, you know, lifetime with their cancer that they will have the least number of cells they'll ever have.

    9. PAPeter Attia2:03:20

       Yep.

    10. MDMax Diehn2:03:21

        And the l- least-

    11. PAPeter Attia2:03:22

        And the fewest mutations

    12. MDMax Diehn2:03:23

        ... chance of already having-

    13. PAPeter Attia2:03:23

        Yeah.

    14. MDMax Diehn2:03:24

        Exactly. The lowest heterogeneity, right? The lowest res- chance of having resistance.

    15. PAPeter Attia2:03:28

        Yep.

    16. MDMax Diehn2:03:28

        So I'm very hopeful that by doing that, we will be able to cure more patients, and we already have evidence of that 'cause we know adjuvant treatment, even given the brute force way like we've done for decades where we basically give it to everybody with a certain stage-

    17. PAPeter Attia2:03:39

        Yeah, the same drug-

    18. MDMax Diehn2:03:40

        ... for the most part

    19. PAPeter Attia2:03:40

        ... in adjuvant is more effective than the same drug in metastatic. Uh, y- y- you know, I-

    20. MDMax Diehn2:03:45

        Absolutely.

    21. PAPeter Attia2:03:45

        Yeah, I think this is a very important point that gets lost on the part of many people in the field, which is early detection absolutely matters, and you have to look no further than that simple point, which hasn't-- I've never seen it refuted. Less tumor burden-

    22. MDMax Diehn2:04:03

        Yep, and it, it's-

    23. PAPeter Attia2:04:03

        ... less heterogeneity equals better outcomes. It's an axiom.

    24. MDMax Diehn2:04:09

        That's correct. That's absolutely true, and that's because we know the, you know, we now understand much better that why cancers are become, initially respond and become resistance is because of this heterogen- genetic heterogeneity that they already have mutations. They're not all the same, the cancer cells. Actually, all two, no two cells are probably the same. They're all different. They all have slightly different k- handful of mutations that are different between each cell, and some s- the more cancer you have, the higher chance that one of those mutations will make the cancer resistant to whatever therapy you're using. So this idea of guiding adjuvant therapy based on ctDNA MRD, I think is the first step towards what, you know, the, the future that you're, that you're envisioning or dreaming of.

    25. PAPeter Attia2:04:44

        And, and so let's think about which cancers that's gonna be relevant in. So it will not be relevant in pancreatic because everybody's gonna probably need adjuvant, I mean, if you're really being honest. It certainly is relevant in your field of lung cancer. It's certainly going to be relevant in breast cancer. It's certainly going to be relevant in colorectal cancer. M- I mean, you could s- you could potentially argue prostate, although the, the treatment is pretty bad. Um, those, that really takes care of kind of the big five right there.

    26. MDMax Diehn2:05:16

        The, the common ones, yeah.

    27. PAPeter Attia2:05:18

        Yeah.

    28. MDMax Diehn2:05:18

        I'd throw in bladder, which is whatever the sixth-

    29. PAPeter Attia2:05:19

        Yeah

    30. MDMax Diehn2:05:19

        ... or seventh most common cancer. You could of course... Th- some of the more rare ones it'll, would be useful with melanoma of course. It's much more rare of course, but, um-
18. 2:10:48 – 2:26:07

    How we get to the panacea of cancer screening

    1. MDMax Diehn2:10:48

       cancer cells in the body.

    2. PAPeter Attia2:10:49

       So those are three great use cases for liquid biopsies that are all going to benefit from a very important insight, which is you know the mutations of the cancer at the outset. So you get to pair a very important piece of information, which is tumor identification genetically, with cell-free DNA. We now need a different approach if we're gonna move to the, the sort of panacea of cancer screening, which is every person once a year goes to their doctor, gives ten or twenty ML of blood, [clears throat] and a test comes back that says, one, do you have any cancer, yes or no? Two, if you do, it's this organ, let's go work it up. Which either means in the early stages more diagnostics, in the later stages means direct to treatment, right? That would be, again, what we talked about earlier. It seems to me that one way to think about that is if you were gonna try to do that based on mutation analysis, you would have to develop the world's most robust database of mutations for every given tumor. Is that, is that even possible?

    3. MDMax Diehn2:12:00

       Yeah, great question. Um, so we, uh, about a year and a half ago published a paper where we, uh, developed actually a mutation-based lung cancer, uh, screening method using ctDNA. We call that method, uh, lung CLIP for, CLIP is Cancer Likelihood in Plasma, and it's a method that is purely based on mutations. The way it works actually is you sequence the plasma and the white blood cells of the, of this screening patient, the sampling of the patients you're trying to screen. You sequence both the CTD- cfDNA, cell-free DNA, and the leukocyte DNA. The reason we do that is 'cause we've learned that there, in, and particularly older individuals, there are mutations present in the leukocytes often that have been acquired through age, through a process called clonal hematopoiesis. So those mutations that are in the white blood cells also end, end up showing up in the plasma most of the time because the white blood cells die in the circulation and release their DNA, and that's part of the cell-free DNA. So those you wanna get rid of because we don't-- that's not coming from a lung cancer or breast cancer, right? So we do both. We subtract the things that are in the, uh, white blood cells, and then we use a machine learning algorithm that looks at the mutations that are left in the plasma and looks at things like how long is the cell-free DNA molecule. We haven't talked about it, but there is, we have, and others have shown that the cancer-derived cell-free DNA molecules are a little bit shorter than that one seventy bases. They're a tiny bit, a few bases shorter on average.

    4. PAPeter Attia2:13:25

       Hmm.

    5. MDMax Diehn2:13:25

       The reason we think is because it probably, most of the DNA in the blood comes from white blood cells. We know that's true, about eighty, ninety percent of the cell-free DNA is coming from white blood cells, and that the other ten or twenty percent comes from solid organs as well as cancer. And probably the DNA has to traverse longer to get into the blood, right? If you're, die in the tissue and then you, to get into the blood takes a while, and that's probably a long-longer time to-

    6. PAPeter Attia2:13:49

       To be exposed to the enzymes that are-

    7. MDMax Diehn2:13:51

       To be exposed

    8. PAPeter Attia2:13:51

       Yeah.

    9. MDMax Diehn2:13:51

       That's right. So we think that's part of the problem. It may also be that some of the way the histones and the chromatin is configured may contribute to it. But for whatever reason, that is true. So we look at Is a mutation present? What gene is the mutation in? How long is the cell-free DNA fragment? Is this mutation one of the mutations that is caused by smoking? Um, all these different things. And it put that in a machine learning model, and the machine learning model isn't just saying, "Have I seen this mutation before in lung cancer?" But it's looking at these other properties to say, "Do the properties of the mutations, uh, match what you'd expect in a lung cancer?" And then the model ultimately spits out a probability that that blood sample was from a patient with lung cancer or not. And so that is, I think, the way that, uh, I would, you know, envision doing it using mutation-based approach, where you're not actually doing this catalog of mutations because that won't work, unfortunately, because every cancer is unique-

    10. PAPeter Attia2:14:44

        Yeah

    11. MDMax Diehn2:14:44

        ... and you can have a mutation in any, in any gene, in any position.

    12. PAPeter Attia2:14:47

        Right. So but you're using... I, I assume you're, you can use machine learning to basically-

    13. MDMax Diehn2:14:52

        That's right

    14. PAPeter Attia2:14:52

        ... predict what should be a cancer mutation versus not.

    15. MDMax Diehn2:14:56

        That's what we're doing. Exactly right, yes. You, you basically, you try to use as much information as you can get-

    16. PAPeter Attia2:15:00

        Hmm

    17. MDMax Diehn2:15:00

        ... beyond just the mutation being there. And now, in subs- in follow-up work, we are trying to extend that by adding more such features, right? The, the more features you have that, that you know are cancer, that link with cancer, the better it is. Again, the question is, can we combine that with methylation? Maybe the combination of the two, right, uh, you could kind of leverage the best of both worlds. Can we combine it with other analytes, other, other approaches? So those-- That's how I think we're move, gonna move forward, uh, to try to develop the, the ultimately the best, uh, screening test. And it very likely will be a combination of combining, you know, it's not just methylation or just mutation or just, you know, what's called fragmentomics, which is this size of the, size of the molecules or their, you know, their distribution. Um, it likely will be a, a combo method. Um, and so that's still very much in the early research sta- phase. One has to prove which of those things combine nicely and, and usefully. And then once you know that, then you have to think about, "Okay, can I develop an assay that's com- you know, affordable that anyone could imagine doing on hundreds of millions of people?" Um, you know, but we're still at the, at the discovery phase.

    18. PAPeter Attia2:16:05

        Do you have a sense of [clears throat] what the theoretical limits look like? Assuming you can take information from each of those data sets, right? So fragment size or f- features of the fragment length, uh, methylation patterns, predictive mutations, et cetera. And based on what you know about the frequency with which you can identify cell-free DNA, and given that we're now exclusively talking about patients that do not have disease on any other imaging modality. Let's just limit it to pre-stage one, or say at best stage one, early stage one. Realistically, where do you think the sensitivity and specificity could be pushed to?

    19. MDMax Diehn2:16:54

        Yeah, for stage one ca- I think that's a great question, and I, I think we don't really know yet. I think one critical question is, one critical pro- uh, experiment, which is what we're, we're currently doing, is to take the most sensitive method we have, which is this n- this third gen method I mentioned that has the one-in-a-million sensitivity, apply that to a large cohort of stage one patients, and see how many of those patients can we now detect, uh, circulating tumor DNA in, uh, when we know the mutations. Okay, th- this is not the screening question, but it's sort of the preamble to it where we, we have a, we can, uh, de- definitively say these patients have ctDNA and how much in stage one. So that experiment will tell us what is actually the lay of the land for the, the amount of circulating tumor DNA shedding in stage one lung cancer. Once we have that, we will know what sensitivity do we have to get to with a naive approach, meaning we don't know what the, the patient has cancer, uh, with, you know, with this integrated approach. Um, and before I, before I can say this, I need to [chuckles] I need to see that data. But I, no, I'm, I'm hopeful we can, you know, uh, push it ab- above the, the 5% we talked about earlier. So in, in our study from, uh, a year or two ago, we, uh, within stage one, we could get to about 20%, uh, sensitivity for, uh, including in stage 1A, uh, non-small cell lung cancer in a validation cohort. Um, so-

    20. PAPeter Attia2:18:18

        And what was the, uh, specificity?

    21. MDMax Diehn2:18:20

        The s- the specificity was 99, 98, 99, 98%, I think it was tuned in that, in that analysis, 98%.

    22. PAPeter Attia2:18:26

        Which again, probably, again, I know that's early stage, it's still just not going to make sense for people whose incoming pretest probability is 1% to 2%.

    23. MDMax Diehn2:18:36

        It's too low.

    24. PAPeter Attia2:18:37

        It's just still too low.

    25. MDMax Diehn2:18:37

        100% agree.

    26. PAPeter Attia2:18:38

        Yeah.

    27. MDMax Diehn2:18:38

        100% agree. So that's why-

    28. PAPeter Attia2:18:39

        I mean, my, my back of the envelope math, Max, is that if someone has a pretest probability of 1%, y- y-- I mean, it sounds crazy. You're gonna need 80% sensitivity and 99.5% specificity to say, "This is a test that matters. This is a test where, okay, now my positive predictive value is high enough that I can do something with this. And just as importantly, my negative predictive value means I can breathe easy."

    29. MDMax Diehn2:19:07

        Right. Uh, I, I think in the ideal world, you're absolutely right.

    30. PAPeter Attia2:19:10

        I mean, again, this is, uh, are we-- Do you think this is a decade away? Is that, is it p-

Episode duration: 2:26:07