This episode will make a great companion for a long drive.
Cancer. Everyone knows at least one person who did everything right and still got afflicted. So what can we do? Peter Attia has some thoughts...
Here's what he covers:
Rhonda: But you know we're talking about metabolic health obviously you've talked endlessly about the importance of metabolic health for cancer um certainly you know cancer prevention but um looking at like so the biggest risk factor for cancer is age , right?
Peter: Uh yes if you yeah unless you include yeah if you don't include modifiable risk so yeah we generally talk about modifiable risk okay yes age age is the greatest risk for all disease including cardiovascular disease
Rhonda: The biggest modifiable risk factor so let's talk about modifiable risk factors like obesity being smoking is number one smoking okay still number one of course smoking i always it's easy to forget you should not be smoking but it is easy to forget it's like oh yeah people do still smoke it's it's it's hard to fathom that but addiction is addiction so smoking is the number
Peter: Smoking is still the number one um modifiable risk
Rhonda: What's after that
Peter: obesity
Rhonda: obesity so why do you think obesity what if you were to speculate why do you think
Peter: yeah i i feel pretty strongly about this i mean i i'm happy to speculate on things and i'm happy to acknowledge when i have no idea here i think we have a pretty good idea first of all i don't think it's the excess adiposity right like i don't i don't think it's the extra two pounds i have on my waist that i wish i didn't have for vanity purposes um it is the environment of growth factors that comes with obesity namely the hyperinsulinemia but also the chronically elevated igf and things of that nature and it is the inflammatory environment that comes rife with obesity and again that's not due to the excess energy that's stored within the confines of the subcutaneous storage depot it's due to the um excess fat that spills over from that into these other areas where fat accumulation is very harmful so fat accumulation is not problematic believe it or not despite our aesthetic preferences when it occurs in areas that we are designed to store excess energy it becomes problematic when it escapes those areas and gets around the viscera gets around our organs enters the muscle itself by the way that's how it directly contributes to insulin resistance uh when it accumulates in the liver accumulates around the heart within the pancreas itself where it serves the double role of not just creating an inflammatory environment but also reducing the amount of insulin that the beta cell can release um and also around the kidneys so those are the main places where even a small amount of fat i.e if just 10 percent of your total body fat were in those places you would be at enormous risk for cardiometabolic disease
Rhonda: yeah i remember i've seen a few studies where it's like visceral fat so you're talking about the fat that's you know covering surrounding your organs you know that was highly correlated with an increased cancer risk and there was like there was also another correlation with like there's some specific inflammatory cytokines that were being generated or you know associated with i guess i would say with the visceral fat and the cancer incidence which again it's like the inflammatory environment like you're talking about so so the metabolic health being important we talked about you know the best like exercise being at the top right i mean that's one of the best ways to
Peter: exercise energy balance sleep and then of course you know management of of of distress right hypercortisolemia will also contribute to this significantly
Rhonda: right which of course even doing things like exercise and getting enough sleep help help balance those right exactly um when it comes to cancer prevention you know you you talk a lot in outlive about cancer screening aggressive cancer screening so can you talk a little bit about weighing the benefits versus the risk of that type you know doing more of an aggressive type of cancer screening
Peter: yeah i mean the reason i think we have to pay attention to cancer screening in such an aggressive way is that unlike cardiovascular disease and even though we didn't really go into the pathogenesis of it today i mean i've covered this on other podcasts i'm sure you have as well it's very well understood doesn't mean we know everything i'll happily spend 20 minutes telling you all the things i don't understand or that we don't understand as a community but we have a pretty good sense of what's going on that's not the case in cancer it is still a really really big black box to try to understand all the different ways in which people get cancer and if you just want proof positive on this i bet you there's not a single person listening to this not one who can't tell you of at least one person they know who's been afflicted with cancer who otherwise did everything right they didn't smoke they weren't obese they didn't have you know huge chemical carcinogen exposures they lived a perfectly healthy life and they still got breast cancer or they still got leukemia or they still got some god-awful cancer so the truth of it is in cardiovascular disease when we sit here and talk about modifiable risk factors like lipids smoking blood pressure all these things that virtually accounts for the entirety of the disease in cancer when we talk about the modicum modifiable risk factors it doesn't even account for half of it so it's free money don't leave it on the table don't make unforced errors don't smoke and be metabolically healthy but you don't want to leave it at that there's still way too great a chance that you're going to end up getting cancer relative to you know if you just take the approach of well
I've taken care of those things. Therefore I've done everything I can. So the missing link, how we bridge that gap has to be through aggressive screening because about the only thing you can say about cancer that is capital T true is when you treat a cancer in an early stage, you will have a better outcome than if you treat that cancer at a later stage. In the book, I talk about a couple of very specific examples of this where we have just overwhelming data. I use breast and colon cancer as an example. So when a person has a stage three colon cancer, that's still a big cancer, right? And it's by definition, because it's stage three, it has spread to the lymph nodes, but it has not spread visibly beyond the lymph nodes. So when you do a CT or an MRI on that patient, you'll see that there is no other evidence of cancer, uh, outside of the region of the resection, which is the colon and lymph nodes. Now, you know that there's microscopically cancer elsewhere. So there are still millions to billions of cancer cells throughout that patient's body, almost assuredly in their liver. Um, but they're not in, you know, you can't see them. If you give that patient the full Fox regimen, which is the standard chemotherapy regimen, that's three drugs. Um, 65% of those patients will be alive in five years. So a third of them will still die, but two thirds of them will live. If that exact same patient, when you go in and you take their colon out and you take their lymph nodes out, also has visible metabolic disease in the liver, they're now stage four. After surgery, they will go on to get the same chemotherapy. None of those people will be alive in five years. There is a fundamental. Why, why that difference? Same is true with breast cancer. Same is true with every cancer. The reason is the more cancer cells you have, the more heterogeneity you have around the burden of mutations in that cancer, the more capable that cancer is to mutate its way out of treatment, evade the immune system, a whole bunch of other things. So if step number one is don't get cancer, which it should be, and we want to do everything we can to not get cancer. Step number two is if you do get cancer, you want to be able to catch it as soon as possible so that you have the smallest possible burden of this disease to treat. And by the way, you know, there's an entire argument that says, well, screening is too expensive. It's a lot cheaper than treating late stage cancer with very expensive drugs that do very little.
Rhonda: So you brought up a lot of good points, Peter. I mean, I, I really liked the way, like, you can do everything you can. And, you know, like my, one of my favorite Peloton instructors, Leanne Hainsby, you know, she's out, she's like doing physical activity every day. I mean, she looks amazing. I'm sure she's, you know, not eating a terrible diet. And she came down with breast cancer, was being treated, and was still doing Peloton classes while she was being treated. I mean, amazing. But the reality is, is that they're, like, over a lifetime, you know, you do, like, there's random amount of, like, things that can happen. Let's say you're metabolically healthy and everything, like, your cells are dividing, you can get a mutation. Your immune cells will take care of it most of the time as, you know, we're progressing through life until we start to get, you know, into our, what, fifth, sixth, seventh decade. Maybe the immune system's not working as well. I mean, there's things that you just can't control. Like, there's that, like you mentioned. So, with cancer screening, what, let's say you don't have any known genetic risk factors, and there's no, like, family history, right? What age would you say, or what decade of life around where would you think that, or how do you treat it in your clinical practice with respect to cancer screenings? What are the major ones, you know, to do? You said colon and breast. Are there any others?
Peter: Yeah. So, you know, a discussion like this always begins with our patients by saying, you know, you have to understand your risk appetite as an individual, and you have to understand the price you're going to pay for screening. Because there's a couple of prices you pay. The first is economic. Everything we're about to talk about is going to be outside of the standard of care. Not everything. I mean, if you're at a certain age, your breast, you know, your mammography and your colonoscopy will be covered, but your colonoscopy won't be covered at the frequency that we're going to recommend you do it. And even if your mammography is covered, they probably won't cover the MRI or the ultrasound that we're going to recommend, because we never recommend mammography in isolation ever. If we're doing a PSA on you and any of our metrics show more care is warranted, they're not going to cover the follow-up study, like a 4K test or a, you know, multi-parametric MRI, unless your PSA is very high. So understand there's a cost that has to go into this. But I think there's an even bigger cost that you have to be willing to tolerate if you go down this rabbit hole, which is the cost of the false positive, the emotional cost of the false positive. So we always kind of start by explaining how sensitivity and specificity work. And I know a lot of people's eyes kind of glaze over and they're like, oh my god, like I don't want to hear the stats on this. But if you don't understand what sensitivity means and you don't understand what specificity means, you can never understand the things that really do matter to anybody who gets a test, which is positive and negative predictive value. Positive predictive value means if this test comes out positive, how likely is it that I actually have the thing it says? Conversely, if this test comes out negative, how likely is it that I'm truly negative? You want very high positive predictive value and very high negative predictive value. And that's a function of three things. The specificity of a test, which is the ability of a test to detect a condition being present if it is indeed present. The specificity of a test, the ability of a test to conclude that something is absent if it is indeed absent. And the prevalence of the condition being tested, meaning how likely is it that you have this before I test you? So you can call that prevalence if you're screening. You can call it pre-test probability. But the point is this is all a Bayesian process. So I really spend a lot of time going through this with people. And let's just start with something as simple as mammography, right? So Peter, why are you saying you're not satisfied just doing mammography? Well, here's why. Mammography has a sensitivity of about 90% and a specificity of about 85%, which is fine, except if I'm going to do a mammography on you at this moment in time, your pre-test probability for having breast cancer is pretty low, like a couple percent. That means the positive and negative predictive value of this test in isolation are very poor, like less than 20%. Furthermore, there are features about you personally that might make you a bad candidate for MRI in isolation. One is you're very young. You're not in menopause yet. Your breast tissue is very glandular. Now in 40 years on a mammogram, your breasts are going to look totally different. The mammogram will actually have an easier time seeing what's going on in your breast because there's going to be less dense glandular tissue. The mammogram, because it's an x-ray, is really good at seeing calcified lesions. It's really bad at seeing non-calcified lesions. Conversely, an MRI really has no issue with glandular tissue but can't see calcified lesions very well. So we go through this analysis and you realize there's actually no perfect test for screening. You have to stack tests on top of each other if you want to increase positive and negative predictive value. If you rely on any one test by itself, you're always going to have a blind spot. The one exception to that, by the way, is a colonoscopy. A colonoscopy is a test that has 100% sensitivity and very high specificity. With colonoscopy, you have a whole different risk, which is a physical risk. There's actually a risk of harm from a colonoscopy, basically three big risks. There's the risk of dehydration, electrolyte imbalance, hypotension that comes from the bowel prep. There's the risk of the sedation. And then there's the risk of a perforation or bleeding, actual procedural risks. Now, if you look at the largest study that came out on this, which was last summer in the New England Journal of Medicine, this was actually a study that was meant to show that colonoscopy wasn't worth it, actually showed something totally different in my mind, which showed how safe it was. It was a study of, I think, over 20,000 people and had not a single incident. It showed that in good hands, a colonoscopy is a very safe procedure, but I always want to make sure people understand we don't take this stuff lightly. And there's a reason you don't do a colonoscopy three times a year, which if you did colonoscopy three times a year, you'd never get colon cancer because colon cancer always has to come from a polyp. So if you were checking somebody three times a year, they would never be able to develop a polyp that you wouldn't catch. But at that point, the risk would be just too high that something else would go wrong. So standard recommendations used to be every 10 years, starting at 50. Current recommendations are starting at 45. And there's some controversy about whether you would do it every five to 10 years. We typically say with no family history or risk factors, meaning you don't have inflammatory bowel disease or Crohn's disease or things like that, we would typically say 40 and then about every three years, depending on the findings. So sometimes the findings on a given colonoscopy will make you want to actually do a more frequent surveillance. If you find a sessile polyp, for example, or if a patient has an incomplete bowel prep, you might decide, actually, we need to do this a little more urgently and do it in a year again, as opposed to wait three.
Rhonda: Great information. And with respect to the combined, you know, especially for younger individuals, like younger like myself, the mammogram starting.
Peter: So I might say like, you know, at 40, I would start doing a mammo and an ultrasound every other year. Sorry, every six months. So you would do a mammo every year, you would an ultrasound every six months, every year, but stagger them by six months. So if there was if there was a high enough risk, that's probably an approach I would take now.
Rhonda: Is that because there's a lifetime risk of one in ages for on average? Forget about all that. OK.
Peter: And again, breast cancer is one of those cancers where if you treat it early, like it's it's it's absolutely a disease that that can be treated early. If you catch this in a stage one, it's a nonfatal disease. A stage four disease is a uniformly fatal disease.
Rhonda: What's the positive predictive value of catching it in stage one with the combination? Well, so.
Peter: OK, so the way to think about it is you think about it as what's the positive predictive value of the combined modalities and and and here it's a little more complicated because it depends on the hormone status. So I'll give you an example. Another thing that we use that we haven't talked about are liquid biopsies. Yeah. So we incorporate liquid biopsies into our testing. Let's talk about them, yeah. Yeah, yeah. So so so have you talked about them on the podcast? Do your listeners know what they are?
Rhonda: I know, there's a question I was going to ask you about with, you know, the the grail by the gallery by grail. Yeah, yeah.
Peter: OK, so what does this test do? So the there are basically three things that you can figure out by looking at strands of DNA in the blood that can give you a clue as to whether or not a patient has cancer. So let's say you collect a bunch of you connect, you know, the grail test uses 10 cc of blood, relatively paltry sum of blood, and they look at all of the cell free DNA. So again, they separate the DNA that's in cells. They don't want that right from the cell free DNA and. Determine so so basically there could be known mutations that we know are cancer genes like a mutation or mutation where you might say, oh, well, if you see that mutation, like there's cancer somewhere in the body. The second thing that gives you a clue that there could be cancer in the body is the length of the DNA fragments that you see. So there's, you know, this is not what grail does, by the way, but there are other technologies that are looking at fragment length and using fragment length to impute probability of cancer. What grail does is they look at a third thing, which is methylation. So they say, OK, well, all of this DNA is yours. We're not going to worry about what the mutations are, what the fragment lengths are. But what we do know is certain methylation patterns are indicative of cancer and tissue of origin. That's a very big deal. So now you are doing a screen for not just does this patient likely have cancer or not, but if they do, can you tell me where that's coming from? So we can now go and look more closely there. Now, there's something really interesting about how this works because it's different from any other type of screening test. See, that MRI that we talked about or the ultrasound or the mammogram or the colonoscopy, for that matter, are basically morphology tests. You're looking visually either directly in the case of colonoscopy or indirectly in the form of a mammogram where you have to look through the tissue. You're looking at the morphology of a cancer. The grail test says nothing about that. It's simply telling you, is this a cancer that is leaving its site of origin or shedding its DNA in sufficient enough quantities outside its site of origin? So something very interesting emerges when you take a closer look at the grail data. And this is why we use the test. Again, I have no affiliation with grail. So this is just my clinical experience and observation. At first glance, the sensitivity of the grail test for breast cancer is quite low. The specificity is very high for grail, by the way, meaning if you don't have cancer, it is very likely to tell you you don't have cancer. The sensitivity is quite low, meaning if you have cancer, it could miss it. And it's been tuned that way. So the algorithm has been tuned for a very high specificity, a low sensitivity. But if you look at breast cancer overall sensitivity, it's about 20% for stage one, stage two, which seems kind of abysmal, meaning if you have a breast cancer that's early stage, stage one, stage two, there's only like a 20% chance it'll show up on the grail test. And many people, myself included at one point, thought that doesn't justify doing the test. I don't need a liquid biopsy to tell me I've got a stage three breast cancer. I'm going to figure that out falling off a log. So I need something to tell me when there's a stage one breast cancer. But a closer look at the data showed that if you looked at ERPR negative breast cancers, stage one, stage two sensitivity was 75% to 80%. It was only in the triple positive, ERPR positive, HER2-neu positive, that the sensitivity specificity are so low. And since that's the majority of breast cancers, it brings it down. What does this mean? It means that the more indolent a breast cancer is, the less likely the grail test picks it up at an early stage. But the more aggressive it is, the more likely it
is to pick it up at an early stage. The implication might be here that it's catching the cancers that matter. And I think that's a very interesting way to combine liquid biopsies with morphologic studies.
Rhonda: Do you ever not combine, like do you think doing just a liquid biopsy by itself would be a useful thing, or do you think really it's better with, you know, in combination with other morphology types of screening?
Peter: Yeah, that's a great question. I mean, we don't do them in isolation because I still think we're in really early days, and I just think a little bit of a belt and suspenders approach makes sense. But it'll be wonderful if the day comes when all you need to do is the liquid biopsy, and only if it comes up positive do you need to go and do a morphologic survey.
Rhonda: A couple of questions. So, you know, talking about some of the major screenings, the colonoscopy, the mammogram, you mentioned PSA, so with like some of these, you know, types of morphology screenings, like the mammogram, for example, people are concerned, like there's a whole group of people that are very concerned about the potential, the mutagenic potential of, you know, these types of screening methods, you know, potentially causing cancer, right? So CT scans, the X-rays.
Peter: Well, CT scans would be a very lousy way to screen for that reason, right? The CT scan has a lot of radiation. With the exception, the only time we justify the use of a CT scan is in a former smoker or a current smoker. We don't have any current smokers in our practice, but we do have former smokers. We do still use a low-dose CT for lung screening. Remember, lung cancer risk is, lung cancer is the leading cause of cancer death globally and in the U.S. for both men and women, and 85% of lung cancers occur in former smokers or current smokers. So in those people, you have to ask the question, what kind of cancers do they get? And you basically have small cell, large cell, and squamous cell are the dominant cancers that occur in smokers, and those are best detected on a low-dose CT scan. Adenocarcinoma of the lung is the dominant cause of lung cancer in a non-smoker, and we can detect that equally well with an MRI. So we don't expose a never smoker to that risk, whereas to a smoker, or a past smoker or current smoker, the risk-reward trade-off is worth it, and that's been documented really clearly in clinical trials. Mammography has incredibly low radiation, not as low as like a DEXA scan or something like that, but it's still really, really low.
Rhonda: There's a lot of women that avoid them. I'm sure there are. I don't know, maybe the radiation has lessened over the years.
Peter: It always has. I mean, radiation is constantly going down. I mean, just going back to something we spoke about earlier, 20 years ago, a C, so just let's explain what the numbers mean. Radiation is measured in units called millisieverts, and it's generally established that exposure to more than 50 millisieverts a year will increase your risk of mutagenesis. So now let's put that in the context of certain things. So living at sea level here in San Diego, just the exposure you get to the environment is about one to two millisieverts a year, so that's two to 4% of your annual allotment. If you live in Denver, you're doubling that. So being one mile in the sky doubles your exposure, but you're still, you know, you're at four to 8% of your annual allotment. A CT angiogram 20 years ago was 20 millisieverts, 40% of your annual radiation allotment on one test. The last patient I sent for a CTA last week, because when we get the report, it also shows the radiation, less than one millisievert.
Rhonda: So mammograms are even less than that. Yeah, yeah, yeah, they're a fraction of that. So it really is, it makes zero sense for a woman who has a lifetime risk of one in eight, and perhaps even higher if she's obese and drinks alcohol, right, to avoid doing mammograms. Correct. Okay.
Peter: But again, I would never rely on a mammogram exclusively. I would combine it with an ultrasound or the MRI.
Rhonda: But they're not concerned about, people aren't really scared of the ultrasounds. They're scared of mammograms.
Peter: That's right, yeah, and MRI, of course, has irradiation. So, but again, everyone has to, you know, you just have to, unfortunately, there's a lot of fear mongering that goes on, but you just have to look at the numbers. I mean, it's crystal clear that a mammogram has a very, very, they might be confusing it with, there was another test, I'm blanking on what it's called now because it's never done anymore. It's called, I think it was called molecular breast imaging. It was another high, high-intensity mammogram. It's, again, I've never seen one done. I don't think they've been done in years. But pre-MRI, like pre-utility for other tests, it was done. It was also about a 20 to 30 millisievert. Maybe this is where this is all stemming from. I'm sure there's a complete misinformation and misunderstanding where people are confusing mammogram from what's called an MBI, is what the test was called. Well, this
Rhonda: is good to clear up. Cause I mean, I'm not just, I'm not kidding. Like I know people, I know women that have this fear. So, you know, I think stepping, sort of stepping back, just one more thing I wanna ask you about is like blood cancers. Is there any, like what, is that a-
Peter: The liquid biopsies are very good on blood cancers, actually, because you have the highest proportion of those cells. Like you're gonna get a much higher concentration of cell-free DNA. So yeah, we actually, that's actually one of the areas where I'm most excited about the liquid biopsies is on leukemias and, you know, other sort of hematologic issues, such as myeloma and things like that.
Rhonda: And for people listening, wondering about the cost of it's typically, it's like 900, like close to 1,000. Directionally, $1,000.
Peter: About $1,000, right. And I don't think it's D to C. So meaning, I think you have to go through your doctor to do it. I don't think you can just do the test willy-nilly.
Rhonda: I don't think you can, yeah. But I don't know for sure.
Peter: I'd be surprised if you could.
Learn more about the advantages of a premium membership by clicking below.
Every other week premium members receive a special edition newsletter that summarizes all of the latest healthspan research.
