This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.drvinayprasad.com/subscribe
Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.
Transcript
From the show. Tap a time to play from there.
SPEAKER_00Multi-cancer early detection tests. Are they ready for prime time? We'll get to the bottom of it in today's Medicine Unpacked. They're all the rage. There's a lot of enthusiasm for a blood-based test that tells you you have a cancer in one of your internal organs. We then find it, we cut it out, and we cure you before it spreads and causes your untimely death. I think conceptually it makes a lot of sense. It's a great idea. The problem is the evidence just isn't there. So the first thing this test needs to do is it needs to find early disease,
SPEAKER_00disease that's localized in one organ or the other. It can't find disease that's already spread. That's too little too late. The problem with most of these tests is that they're much better at finding disease that's spread than disease that's localized. The next problem is if you have an abnormal test result, an early stage cancer, you need to be able to scan a person or provide an endoscopy or somehow look in the body and find out where that cancer is and cut it out before it spreads. It's no use to you to find out you have a small intestinal cancer.
SPEAKER_00So what am I to do with that if I can't localize where it is? You want me to cut out all of my small intestine? You have a colon cancer. You can't figure out where it is. You want me to cut out my entire colon? So it has to be visualizable on radiographic scans or on endoscopy or on some surgery. And then the final step is it has to be amenable to being clipped out or cut out. If you visualize the colon, you find one mass, you think that's the reason the blood test is positive, you cut it out, but the blood test doesn't come back to normal, it's still elevated.
SPEAKER_00That suggests that it was somewhere else in the colon and you didn't quite cut it out. Additionally, this may surprise you, not every part of the body is equally amenable to having something resected. The head of the pancreas, the tail of the pancreas, the mid-body, these are very difficult regions to surgically remove something, and often it requires a huge surgery. So what do we know about these tests?
SPEAKER_00We have some test characteristics. We know the sensitivity and specificity with which it finds early disease. But what do we know about whether or not that disease can be localized or whether or not you can be cured of it before it causes untimely death? We know very little. We have some randomized data. That was powered and designed to show that by running this blood-based screening test on tens of thousands of people, we would reduce the rate of advanced disease, stage three, four disease, but even this modest benchmark fell short. We're not even talking about mortality. We're just talking about reducing advanced disease and even that we haven't quite gotten to.
SPEAKER_00And there's this conception that You know, the more you do these blood tests, the better off people will be. The worst case scenario is you get a false scare or you can't find it right now, but you follow them with scans. Eventually it pops up. That's the worst case scenario. But that's not the worst case scenario. We know from many lines of evidence that there are tumors that actually look exactly like a breast cancer or a colon cancer or a lung cancer that are destined not to cause you any problems in your natural life,
SPEAKER_00not to cause morbidity or mortality. This is the so-called over-diagnosis problem. But what we don't know is whether or not these tests are good at finding the cancers that are going to spread and cause you harm and not the cancers that aren't going to spread and cause you harm, those that are going to remain indolent, those that may spread but not grow tremendously and not be the ultimate cause of death, or those that may even spontaneously regress. We have no idea what we're doing. The other thing to know about this whole class is that many of these tests are already available to you if you pay out of pocket.
SPEAKER_00Any regulatory action is not for you to have access to the test. You already got access. It's so that the insurer, a third party, can pay for it for everyone. And that's what it's all about. It's about getting someone else to pay. Because the moment you get someone else to pay for the test, you're going to get a widespread adoption. You're going to get a big boom in the market. But the moment individuals have to be persuaded to put their own cash on the line, well, of course, the market share is much reduced. You know, recently someone asked me, which do you dislike more?
SPEAKER_00The idea of a whole body MRI every month or week or year or the blood-based cancer screening? Which one do you have more quibbles with? And I must admit, it's actually Especially the blood-based screening test, because the imaging test, if it finds something abnormal, you've got one piece of evidence the blood test doesn't guarantee, which is, you know where the abnormality is. You can follow it with serial imaging, you at least can point at something. What do you do with a blood test that gives you an abnormal result and you can't point at anything? You might think it's a false positive, and it might be.
SPEAKER_00You might not have any cancer, or you might be the kind of person who has a cancer somewhere that is about to become radiographically apparent, but it's not yet apparent, and that might take considerable time. What if the test is correct? You do have a cancer. You just can't see it, and no surgeon knows where to operate. They can't do anything about it. They follow you with scans. By the time they see it, it's already metastasized. It's of no use to you in that case. So what am I to think? Is overall mortality the only thing we're looking for here?
SPEAKER_00Absolutely not. We'd be happy to show a reduction in surgical morbidity and a reduction in advanced disease. We'd even be happy to take a reduction in advanced cancer, stage threes or fours. We don't have any of that, okay? We don't have any of that with this class of intervention. And that's the problem because it's about getting people to pay for something before the evidence is generated that anyone is better off, well, besides the person who makes the test. So multi-cancer early detection tests, are they promising? Absolutely.
SPEAKER_00Is it the sort of thing that I think AI might be able to help and better refine the tests and get better results in the future? Absolutely. But for the time being, is it ready for prime time? I don't know. I struggle to imagine what will happen to the person who gets an abnormal result who goes to the doctor. Is the doctor obligated to do MRIs of the whole body, PET CTs of the whole body? What if you do all that and you still don't find anything? Doesn't mean it doesn't exist. We might just be a month away from seeing it. So are they going to do a PET scan every month until they find something?
SPEAKER_00This is all uncharted waters. We don't know what we're doing here. So we're opening up this box. We hope it's not Pandora's box. We hope it's going to help us lead longer and better lives. But there are lots of reasons to doubt that. And the worst case scenario is you turn someone into a repeat surgical patient and you keep cutting out parts of the body. The test has not reduced its value. It still looks abnormal. And at the end of the day, you still die of the cancer you might have otherwise died of. That's the worst case scenario. And we're not sure how far advanced we are and whether or not we can avert those scenarios.
SPEAKER_00I think it's a tremendous opportunity for third parties to generate information that might be beneficial for different payers to think about what this means. Just because federal payers may be obligated by law to cover any test doesn't mean private payers have to, and there may be some flexibility around this space. This is a great opportunity for trying to educate the public that not everything that looks like cancer under the microscope is something that will kill you. And not every abnormal cancer test is something that can lead to an action
SPEAKER_00that will benefit you. And that's a tough message to get across that many of us have struggled with for our whole careers, including Gil Welch, who I think is a great public communicator of this topic. So if you like this video, follow me on this channel, like, subscribe, comment, leave a message below. And read Gil Welsh's book, Should I Be Tested for Cancer? Maybe Not and Here's Why, or Overdiagnosis. I think those are my two favorite on the topic. Until next time.
Live episode from Youtube This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.drvinayprasad.com/subscribe
Elizabeth Holmes/ AI researchers start smoking/ Moderna vax by Plenary Session This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit...
Pope has cancer; Phase 1 trials for AI; Can AI hack cancer trials__1 by Plenary Session This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit...