Blood based Colon Cancer Screening - NEJM papers - ECLIPSE - Guardant

welcome back to the Channel people have been asking me to talk about the two recent papers in the New England Journal of Medicine about coloral cancer screening we've got the blood-based cancer screening test by gardant we've got the stool DNA test by exact biosciences and these are poised to become lucrative tests that doctors and hospitals administer to patients or I mean healthy people we'll see if they make anyone better off we'll talk about that in this video so first of all we have to be clear screening is about healthy people it's not about patients so I may occasionally make that mistake as I talk through this video but we're taking healthy people we're extoling them to come in and participate in some preventive screening program in the hopes that we're going to make them better off so let's talk about the history of coloral cancer screening let's talk about what we know about these two tests and let's talk about why I have some reservations about them so first the history of coloral cancer screening I mean we have a number of modalities that have Direct randomized Control trial evidence evaluating whether or not that screening low lers the rate of coloral cancer death and ultimately all cause mortality that's the thing we care about living longer as a result of screening when it comes to feal occult blood testing fobt we have a few randomized control trials that show very clearly that fobt cards when used when offered to patients can reduce the risk of coloral cancer death but we also know that even in pulled analysis that reduction in color rectal cancer death does not translate into an improvement in all cause mortal the other thing to know is the reduction of colal cancer death is not 100% from fobt it's roughly for all cancer screening tests between 20% and 40% in other words even the person who participates in the screening program perfectly still has 60 to 80% of the risk of dying of that Target cancer this is something people Miss screening does not prevent you from dying from colal cancer there are plenty of people who died of colal cancer who got all the appropriate screenings it may reduce the risk of that cancer specific death a little bit with fbt we have data does that we also don't have any credible data that those reductions translate into all cause mortality benefits this was also shown in a nice paper by Bret how looking at the average life expectancy gain from cancer screening tests when it comes to colonoscopy we have one randomized control Tri Nordic which has thus far failed to find even a reduction in colar rectal cancer death it might find that in the next analysis but thus far it's failed to show that so we it does however have an increased incidence in col cancer sub followed by a subsequent Fallen coloro cancer incident and we've talked about that on this channel I've interviewed Michael bretow you can go check that video out I've had a debate with Adam seu about this check that debate out let's turn to the modality that has the best evidence flexible sigmoidoscopy it's not a colonoscopy you don't have to spend all night in the bathroom you don't have to drink the bucket of glp and poop yourself to near death you can just get a couple of Animas and have a flexible sigmoidoscopy it evaluates the left side of the colon not the right side and it has shown very conclusively in multiple randomized control trials a reduction in Death from coloral cancer and even in a pulled analysis a reduction in in Death from any reason an improvement in all cause mortality it actually we can say makes you live longer in my mind Flex Sig which is a cheap test which is a test that doesn't require all the preparation of colonoscopy should be the de facto standard of care it has the best evidence it's bizarre that we do colonoscopy why do we do it you may ask well of course at the end of the colon is a $1,000 bill and every GI doctor wants to go sag that no I'm just kidding but that's partly the case there's a big Financial incentive to do a colonoscopy the other reason is I think people genuinely believe that if it works for the left side of the colon why wouldn't it work for the right side of the colon there are a lot of reasons to think it might not work the same on the right side in part because the biology is different on the right side than the left side and in part that we don't have direct evidence that it works on the right side we don't have direct evidence that colonoscopy reduces cause specific mortality yet but be that as it may for whatever the reasons flexible sigmoidoscopy often a quarter the price of colonoscopy depending on the place maybe even less it's done much quicker it is in my opinion the best screening modality is the best evidence that's not to say that everyone has to do it if you offer it to someone and they say they are willing to tolerate the slightly higher risk of mortality because they don't want to go through the procedure I think that's perfectly acceptable people have the right to take risk in their lives so the job of a cancer screening program is merely to offer the cost-effective evidence-based solution to the majority of people but not to co them or compel them to do it in my opinion so I think flexx Wings wins now let's talk about the advances since then we moved from fobt to feal imunohistochemical testing because it offered what we felt was better sensitivity and specificity than the original test and fit was also meant to be done yearly okay it's done yearly and over years the sensitivity will improve with each subsequent year this is an important Point as it relates to these two recent papers so we move to fit for the most part but again fit does not have any Direct randomized Control trial evidence that improves cause specific death or even all cause mortality now let's talk about a few years ago they made a blood-based cancer screening test and with rby peri and I wrote an article in jamama where we were critical of the blood-based cancer screening test because it had very poor test characteristics enter the more recent New England Journal study this is the gardant blood-based cancer screening test and it's still in my opinion terrible it has terrible test characteristics it has a 100% sensitivity at finding stage four cancer you do understand that that's not a useful thing to find okay if the cancer is already metastatic you finding it is very unlikely to change the natural history of that tumor you're never going to cure that person you're unlikely to cure that person should say and you're certainly unlikely to make them live substantively longer screen screening for stage four disease is a Fool's errand what you want to screen for why we do screening is ideally you would find early stage cancer and intervene upon it or pre-cancer and remove it before it would otherwise become a lethal cancer when it comes to flex Z not only can you clip a stage one cancer can identify it and reect it or remove it D at the moment you can also clip Advanced polyps preneoplastic polyps very highrisk polyps and that can avert the natur history of the disease the same is true for colonoscopy when it comes to this intervention they're very happy to say they have something like an what is it 83.1% sensitivity for any coloral cancer but they have a 13.2% sensitivity for advanced precancerous lesions that's so piss poor in fact that's worse than every other modality that's worse than the methylated sept9 test that old test from a few years ago that's worse than fit that's wor wor than multi-target stool DNA that is the absolute worst Advanced precancerous lesion detection of anything it's piss poor it's absolute garbage at finding those pre-cancerous lesions so in other words it might have some sensitivity for cancer it has great sensitivity for stage four cancer it has very poor sensitivity for advanced pre-cancerous lesions and it very well could be that finding and removing those Legions is the thing that gives flexx Sig both the cause specific mortality reduction the all cause benefit okay so in my opinion the the mere fact the FDA will approve this product without Direct randomized Control trial evidence that it lowers cancer specific mortality is yet another failure of the FDA they just want to let people sell their costly products to the US market we will have no idea I'm going to say many more reasons why we have no idea if by introducing this product to the marketplace we're going to make people better off but we certainly don't we certainly see already it's test characteristics I think are quite pardon my French okay it's got shitty test characteristics okay I'm not so interested in finding stage four disease okay that to me is also one of the problems here one of the very interesting things is in the paper they don't present the sensitivity by stage it's only the supplement which I find interesting they lump stage 1 two 3 it sensitivity is 87% for stage 1 through three but the precancerous lesions Advanced precancerous Legion sensitiv is very poor it also has you know not terrific specificity so there's a lot of false positive results getting colonoscopy the next point blood-based cancer screening test is only going to be useful if the person comes back and gets a colonoscopy in this study to validate the test characteristics everyone had to have both but in the real world when you deploy the test the authors of this paper and some of the commenters online are saying that what's going to happen is the beauty of this is you don't even have to poop on the card when you come to the primary care doctor we can just draw your blood and we'll just send this test on anybody well the problem with sending the test on anybody is that person might not ever come back to get the followup procedure at least pooping on the card is a compliance device and make sure you're likely to come back if you start sending this test on everyone you're going to find a lot of people who test positive by this blood test who will say I don't want to have a colonoscopy no I'll just keep an eye on it can we just follow it could do I even have to return to Clinic they may just blow you off and ghost you you know so we have no idea if you deploy the test in the real world if people fall up for the colonoscopy we have no idea if the test is actually finding the precancerous lesions that result in improved mortality in other words in the cancer screening model there are three types of things you can find you could find the birds which are disease states that even if you find and catch and and perform surgery on the hes left the barn they're going so fast they already have microscopic disease elsewhere you're not going to change the Natural History stage four disease is one of those things okay finding stage four disease is not so useful we don't know if you're finding Turtles finding the disease that if you otherwise didn't find that would have just gotten along indolently stay a stage one colon cancer stage two and then maybe even declared itself with symptoms in a point where it was still completely treatable by surgery okay so by catching it early you may have done a slightly smaller surgery but it was just going to be cured with a surgery later what you really want to find are those cancers that if you find and clip it clip the pup and remove the cancer in the absence of you doing that it would have become a metastatic cancer and we have no idea from test characteristics alone if we're finding those tumors and that's yet another reason you need direct randomized control evidence that by in that by De debuting or offering this test you reduce cause specific and ultimately all cause mortality for people so that's a failure of the FDA Paradigm they say in the paper that all they have to show is that the confidence interval for sensitivity for coloral cancer is above 65% sensitivity and they'll get FDA approval that to me is such a piss poor bar and yet again reflects this FDA which is an abject failure they're proving Alzheimer's drugs that don't work they're proving boosters in babies based on no credible evidence boosters in people whove had covid-19 cancer drugs that have no credible evidence Cancer drugs like olap rib and the polo studies see my prior videos that are probably worse than the Alternatives of continuing chemotherapy they're proving cancer screening tests that don't work they're just an agency that doesn't want to do their job correctly they don't want to actually regulate products in a way that we know they actually benefit the American people then the next part of the argument when you debut the blood-based cancer screening test when you debut that on the market the people who are writing the editorials by the way we'll talk about the editorial it's not a good editorial the people are promoting this test they keep harping on the fact that we want coloral cancer screening rates to be 80% why do we want that because some organization pull that number out of their ass okay I mean it's a madeup number they're currently at 59% okay actually it's probably even lower because some of the statistical errors in how they calculate that percentage but it's it's a lot of people are not compliant and they say quote the best screening test is the one that gets done implying that if we add this Blood based screening test and we screen a few extra people we bump that up to 70% that's going to be a net gain but what they don't know is that the people who are not doing any test are those going to be the people who do the blood-based cancer screening test and everyone else who is going to do colonoscopy and flex Zig and fit they're just going to do the same thing or are you going to take people who were going to do a colonoscopy they were willing to do a colonoscopy but they got seduced by some poorly done New York Times article fluff piece about the blood-based screening and they do that instead and as a result of doing that instead they actually deteriorate the sensitivity of finding these things because here in this study colonoscopy is hail as a gold standard it has a 100% sensitivity by definition in this study but they do this instead of colonoscopy if that's what happens as a result of debuting the test you actually erode the benefits of screening I strongly suspect that this is going to lead to an erosion of the benefits of screening because the test characteristics for advanced pre-cancer lesions are so piss poor you're going to take a lot of people who were going to do something better and they do your piss poor test and as a result they have worse col rectal cancer outcomes so that's yet another failure of the FDA they have no real world data of what the introduction of this test will mean in the US Marketplace I've been going on about the blood test I think that's because it's more interesting to me there's one more thing that they note that as you get older the specificity gets worse because probably we all have a lot of these um cell-free DNA fragments they're looking for are fragments that are more incidentally found in older people as all of the cells undergo aging and mutations and DNA damage Etc and so actually the test characters get poorer and older people it's an interesting finding now let's talk about the stool based DNA test when we had the original stool based DNA test one of the problems with that test was they compare sensitivity and specificity against fit but at a one-time visit and that's again what happens in this New England Journal paper they compare fit and the DNA test and by the way fit is feal imunohistochemical testing which is not a DNA based test but an IM imunohistochemical assay they compare the two at one time but then they offer that test card every 3 years whereas fit is offered every one year but that's pretty stupid isn't it if you want to really compare the two tests you should do it for 9 years or 10 years and then compare the the sensitivity over the course of 10 years where one thing is done every year and the other thing is every three years it's the difference between you know if you're running a marathon and you drink a little bit water at every mile or you drink a big Gatorade every 3 miles and you compare how hydrate you feel after the first mile it's not fair you should at least get nine miles under the belt so we get some benefit of the fact that in one case you're actually stopping a little more frequently and that's the same with fit you're going to get an improved sensitivity by doing it every year versus every three years why is it every three years it's probably every three years because otherwise the cost would be so astronomically bad they wouldn't be able to justify it for most payers and that's probably why it's stuck at 3 years so similarly these one-time comparisons of stool DNA and feal immun hemistry testing are actually pretty useless in my mind they don't even tell you the test characteristics because know the test characteristics I want to see that over a longer time Horizon and again I come back to the core problems which is at least with the stool tests one can assume that if somebody's willing to poop on an fobt card they're going to be willing to poop on a fit card or a stool DNA card and that they have the equal equally to return to Clinic that's an assumption that one cannot extend to a blood-based cancer screening where they' be much less likely to return to clinic and participate in colonoscopy at least you can make one of those assumptions that they will follow up at least at the same frequency as they did in for fobt but you really don't know what you're getting in terms of cancer specific mortality reduction over the Long Haul particularly when you factor in that fit sensitivity is going to be much better done yearly and so I think that's the bigger problem there I don't want to dwell too much about these I think fit you know for better or worse is an option and it's hard for me to believe that any of these stool based DNA options have shown even that they're as good as fit and certainly not that they're better but for me the more interesting story is the blood-based screening because that's really Pandora's Box because that's when you're going to get a lot of doctors just ordering it as part of the routine blood work they check the lipids and the CBC and this screening test and you're just going to get a lot of people who probably get this test done who didn't really understand what they were signing up for they may get a false positive they may no show they may not want to show there may be some other people who are going to get a colonoscopy and the doctor just does this instead and that may erode and all of this may result in actually worse color rectal cancer outcomes the last thing I'd say when you read the editorial it is I don't want to say the worst editorial I read but it's not a good editorial the person writing the editorial is is just just I don't know it it was more of sort of a high school level summary of the field without any of sort of the careful considerations or the many of the facts that I'm laid out for you what will this actually mean in the real world it was more of a cheerleading kind of thing and the editorialist was quoted in the New York Times again cheerleading that this is the future this is where we're headed and something like that that's not what we need we need editorials who actually put pressure on the manufacturers of these potentially multi-billion dollar products to generate evidence that actually informs us whether or not the introduction of those products into the US Marketplace will result in better outcomes for people or no better outcomes or Worse outcomes even at higher prices and the editorial list was so far from that it was laughable the other thing is when you read the news coverage of this they keep asking people what they think about colal cancer in the New York Times story they ask somebody and it says this person is a consultant for exact biosciences you mean to tell me you can't find someone to talk about this topic who is not a consultant for any of the companies involved in this space H I'm not a consultant for any of the companies developing any colar rectal cancer screening test you could ask me and I'll give you some much better comments than what you're getting there so in conclusion in conclusion is is America is a delusional Place America has lost their marbles there is one cancer screening test in the coloral cancer space with the best evidence that is flexible sigmod docop it not only has a reduction in cost specific mortality as an all CA mortality benefit you don't have to drink the bucket of slop the night before you can do two Flex sigs of course Michael bretow has some limitations he talked about in the prior podcast but even most of Europe has been doing Flex sigs okay you could do colonoscopy instead I think there are other compelling reasons to do colonoscopy it can remove polyps it is something that has better control because anesthesia is given it can often be less uncomfortable for patients because anesthesia's G and there's a perforation rate as well okay you could do colonoscopy instead and I wouldn't have had I don't have as strong an objection to that even though Nordic is for the time being a negative study if you're going to do pooping on a card Let It Be Fit it is cheap it's done yearly it has a long track record it has principles that are quite readily extrapolated from the fobt studies the stool DNA again I don't know what sensitivity is over the Long Haul when you do Q3 years versus q1 year this stool DNA again I don't see any benefit to that I also uh yeah particularly when you compare to against yearly but I I I think the stol is already you're already talking about an inferior option and the blood-based cancer screening test I think is a total disaster but most importantly I think setting the goal of 80% is ridiculous how can you set that goal even in the pool analysis of all cause mortality in the an of internal medicine for flexible sigmoidoscopy the hazard ratio is something like and some listeners can check me I think it's 997 it's been a few years since I looked it up but it is almost one in other words even if you did it very Faithfully for the one thing that has the best evidence of all cause mortality benefit the benefit is very slight so it is ultimately a person preference sensitive question do I want someone to go up there and look with the scope and if the answer is you don't want that or yeah it's not worth it for a small increase in life expectancy then to me saying no is a reasonable answer and getting a blood-based test instead thinking that means no one's ever going to go up there and look is ridiculous because the whole point of the test is and the only way it could possibly benefit you is if if it's positive someone going to go up there and look that's the only way could possibly benefit you and if you're not willing to let somebody go up there and look then you shouldn't do any of these testing okay and these programs that try to push it to 80% they're delusional it should be pushed up to as high as people actually want if they proper informed about the topic there's no number some some populations may say screw that it's not worth it to us and other populations might be higher it might be 100% if they're properly counseled the gastron neurologists and these commenters on this topic have never embraced anything other than colonoscopy for many years one I think their financial conflict of interest they make a lot of money by doing colonoscopy but in this case they're finally getting on the blood-based cancer screening bandwagon and I wish it were because the evidence is good I suspect strongly it's because these companies have so much money they are detailing doctors at a level that we have never seen before they're giving more payments to doctors and they're having more ongoing research studies and they're really cultivating those relationships and they're engaging in just Brute Force marketing to push these products which are very likely inferior to endoscopy which is the gold standard of care in this space so coloral cancer screening do you have to do it I think the doctor the doctor should give you the full evidence and you should decide and I think target screening levels are crazy and if we do it in America we should be doing the cost-effective thing with the best evidence which Flex say if we're not going to do that we're going to be crazy we should do colonoscopy at least we can find pops and clip them and if we're not going to do that and we should at least do fit and if we're not going to do that maybe we shouldn't do anything rather than offer these very inferior tests to people inferior or costly or costly and inferior like this present example so those are some of my rough thoughts in this space I think cancer screening is one of the toughest things um it's it's easy to feel upset when you hear about a young person dying of untimely coloral cancer I feel upset too it's easy to wish we could go out there and get rid of coloral cancer but the best evidence suggests that even if everybody did it all the time you know even in the positive studies the majority of coloral cancer is I shouldn't say if every did all I even in the positive studies the majority of colal cancer is not affected by these screening tests it only reduces it maybe 20 to 40% it's not a Panacea there's going to be cancers that occur between screening intervals even if even if you do them as frequently as you could possibly want to do them I shouldn't I should be careful because Brian Johnson might be starting to do it every day who knows this guy he's he's got all sorts of wackadoodle ideas about life extension that are entirely unproven in my opinion Society has to be very careful that just because something is bad we don't do something stupid in response but I would say probably one of the great lessons of medicine is that something is bad and in response we do something stupid that's really the lesson of much of the 20th century medicine doing lots of unproven things and that might arguably be the lesson of the covid-19 pandemic something was bad it was really bad that 80-year-olds died at record high numbers but we did something stupid in response which is we closed the schools for the eight-year-olds I mean that could be pretty stupid to do if you this was the problem and this was the stupid thing you did and similarly coloral cancer screening colal cancer is bad very bad when it kills 40y olds lowering the screening age to 45 I think I talked about in some prior videos and some blogs um and and a paper in bmj uh evidence-based medicine that's that might be stupid we have very little we have no evidence direct evidence that lowering the age actually improves outcomes introducing blood-based cancer screening test to improve the overall percent of people get screened might be very stupid it might erode the benefit of colonoscopy and it might lead to a lot of people no showing for colonoscopy or for follow-up for this test might actually increase the the the failure rate of following up and it might actually not change color cancer mortality in the population so we could be doing something very stupid the only way to figure out if you're doing something stupid or smart as better studies and the FDA this FDA is so bad they will not compel the maker to do any good studies they're settling for these terrible test characteristic studies that have a number of limitations that I've described that's bad the real failure is the FDA I I I when I started my career I always felt the solution was to have a better FDA but now you know having been in medicine for 18 years I think I'm starting to sympathize with some of the critics of the FDA who say it's actually might be better to have no FD at all they just give a false label to things of safety and efficacy they provide no real standard of efficacy and they actually just keep prices elevated by their false label and their bureaucratic hurdles and maybe a world without them at all and a complete L Fair Marketplace might even be better because at a minimum these unproven interventions could not charge an arm and a leg so I'm starting to become sympathetic to the idea just because they've been doing such a lousy job for so many years so those are my thoughts I will be back with more videos I am of course a practicing oncologist and professor of epidemiology and a expert in evidence-based medicine you can check out bad.com and go to my publication tab VK Prasad lab.com and look at our Publications you can check out my UCSF profile which lists I think all of our now 500 plus Publications you can check out my book malignant which is about cancer medicine and evidence and uh ending medical reversal a good book about I think it has a part on cancer screening in there um I'm on Instagram now you can follow me there I'm also on Twitter um subscribe to the YouTube channel follow me on Twitter oh and most importantly check out vanai prad's observations and thoughts my substack where you're going to get a lot of really interesting content including a latest essay on Andrew huberman you got that you got another nice essay coming to you soon so on that positive note I'll be back with more videos and more podcasts plenary session sensible medicine we got a lot in store for you until next time

Channel: Vinay Prasad MD MPH

Views: 40,019

Keywords:

Id: yJL-Fkp_5_c

Length: 26min 6sec (1566 seconds)

Published: Thu Mar 28 2024