Good morning, and welcome to this morning's webinar, sponsored by the American Society for Gastrointestinal Endoscopy, addressing the recent New England Journal article released online on October 9th, entitled, The Effect of Colonoscopy Screening on Risks of Colorectal Cancer and Related Death. I'm Brett Peterson, the current president of the American Society for Gastrointestinal Endoscopy, and I will serve as your moderator today. This study, which we're discussing, employs clinical investigators from the Nordic European Initiative on Colorectal Cancer. It provides the first 10-year results of a pragmatic, randomized trial, first initiated in 2009. Within hours of publication, our American media distributed a flurry of news articles with the headline calling into question whether colonoscopy reduces cancer-related deaths. Today we have four distinguished guests, all experts on screening for colorectal cancer, to discuss the Nordic trial results and the implications for current practice. To begin, we will hear from the primary author, Michael Brethauer, who is tenured professor of medicine at Oslo University Hospitals in Oslo, Norway, as well as Michael Kaminsky, who is professor of medicine at the Maria Soledowska Curie Memorial Cancer Center and Institute of Oncology in Warsaw, Poland. Thereafter we'll hear responses from Douglas Rex, ASGE past president and distinguished professor at Indiana University in Indianapolis, and Jason Dominance, professor of medicine at the University of Washington in Seattle, Washington. Following the remarks and response, we'll have panel discussion for about 10 minutes and further remarks from both authors and respondents. And so I'm happy to introduce Dr. Michael Brethauer to begin our discussion this morning. Thank you, Brad. Dear colleagues, everybody, it's a pleasure to be here and thank you for the invitation and the opportunity to discuss our recent paper in the New England Journal with you at the ASGE. So as Brad was elaborating on, this is about the recent paper from our randomized controlled trial of colonoscopy screening versus no screening, which was performed in Europe. So the Nordic trial is a randomized population-based trial that is important to remember for later discussions of a little less than 100,000 people, all presumptively healthy, around 60 years of age, 55 to 64, who either got randomized to one colonoscopy screening or to usual care, which meant no screening in the trial areas. So it's important to remember that this is a trial that invited everybody who lived in the trial areas, with no prior question if people would be interested. That is important. The trial was in four countries, as you can see on this slide, and the primary endpoints are colorectal cancer incidence of mortality after 10 to 15 years, or finalizes after 15 years, and our first after 10 years, which is what this paper is about. Now for this paper, we were able to retrieve data from three of the four countries. The fourth country, the Netherlands, unfortunately, although we have all the approvals and everything, unfortunately, due to the new European data regulations, it was impossible for us for this analysis to retrieve data from Dutch patients. So that leaves us with about 85,000 patients that were in this analysis that we have presented in the New England Journal last week. Here is the baseline characteristics of the trial participants. As you can see here, most of the patients are from Poland and Norway. Very few were recruited in Sweden, less than 5%. So with regard to screening participation, as has been broadcasted, the general participation all over was 42% out of all those folks who were invited and lived in the areas. It was very different in the different countries, with 33% in Poland and 60, almost 61% in Norway. Please keep that in mind when we come to the results. Now the screening performance we think was very good, good or very good bowel preparation in above 90% of all the people who attended, and cecum intubation rate above the general threshold for good performance. With regard to adenoma detection, it was generally high, we believe. Although it was lower in Sweden than in the other two countries. And of course, there was a variation in ADR between endoscopists, as you would expect, and as has been observed in many other settings as well. Now with regard to the Swedish ADR, we have to remember that colorectal cancer risk in Sweden is lower than in the two other countries. And this is just a reflection of how the incidence is in the different countries. And therefore, you would also expect a somewhat lower ADR in that country as compared to the other ones. Not many screening-related adverse events, as you can see. This is entirely predicted. Also, just for everybody to remember, the contamination in the usual care group was very low. Below 2% of the folks who were in the usual care group actually received a screening colonoscopy, which is minimal as we consider it. These are the main results. This is the intention-to-screen or intention-to-treat analysis, which always is the primary analysis in randomized trial for a good reason, because this is the analysis which is not biased, because it is the folks who are randomized to one group versus the other, which is textbook analysis for all randomized trials in medicine. As you can see, there is an 18% decrease in the relative risk of getting colorectal cancer in the people invited to screening as compared to the usual care group. And the overall risk is at about 1% at 10 years. Now, also, of course, important because everybody realized 42% of the people showed up who were invited. It's interesting to see how did these people fare as compared to the usual care group. This is an analysis which, of course, is biased by self-selection. However, we tried to adjust for that as best as we could. And here you see the results. The usual care arm is 1.22% absolute risk, and in the screen group, 0.84, which gives you a 31% relative risk reduction of colorectal cancer incidence for those people who actually attended the intervention. What about mortality, which has been broadcast widely? We believe at this time that incidence is the more interesting input. However, we also give you the mortality results. There's no reduction in colorectal cancer mortality for the intention to screen analysis. There is a reduction in the per-protocol analysis. We applied two different methodologies because this is difficult to do, and it's difficult to do per-protocol analysis with a low risk of bias. Therefore, we ran two different methodologies that gave us relative risk reductions for mortality of anything between 28% and 50%. Now, just one slide about the two different countries with that participation rate, which is very different to each other. And these are the two countries that contribute with more than 95% of all the patients in the trial. Participation in Norway, 60%, as I said. In Poland, 33%. If you look at the effects of the screening, for intention to treat, not much different between the two different countries. For the per-protocol, the effect in Norway is higher than in Poland. So you could take away two things from that. Number one is that, of course, it's important that more people show up. But number two is that there is no linear relationship between participation and screening effect. And that is important to remember. Now, to conclude, there is a reduction in colorectal cancer incidence, which is anywhere from 18% to 31%, with the absolute risk as displayed here, around 1%. For mortality, for colorectal cancer mortality, that is, it may be up to 50%, but it may also be zero. The point here is that the absolute risk for dying of colorectal cancer is actually very low, both in the screening group and in the control group. And finally, just one word with regard to all-cause mortality, colonoscopy screening does not reduce all-cause mortality, at least not in this study. Michael, over to you. So then we would like also to explain two things, or our understanding of the effect on mortality, colorectal cancer mortality and colorectal cancer incidence. This first slide shows, we think, an interesting view on epidemiological background. So this slide presents expected and observed colorectal cancer incidence and mortality at 10 years follow-up in the usual care group. Obviously, when we did our calculation, power calculation, we based it on incidence-based colorectal cancer mortality. And what we did expect at 10-year observation was around 0.54% mortality in the usual care group. Obviously, it will be powered enough at 15 years of follow-up when the expected colorectal cancer mortality, based on the data from cancer registries, is supposed to be 1%. However, I want to draw your attention to the fact that what we actually observed in the trial after 10 years follow-up in the usual care group was colorectal cancer mortality of 0.31%. It means that also in the usual care group, the colorectal cancer mortality, as compared to what was expected, was reduced by 40%. And this is likely due to the improvement in diagnostics and also treatment of colorectal cancer, which means that probably colorectal cancer becomes a less lethal disease than we expected 15 years back when we designed the trial. When you look at the incidence, the expected and observed values are pretty much similar. For the incidence, there is not so much difference, and obviously some of the difference comes here from shorter observation period in Norway, which has a pretty high colorectal cancer incidence. So over time, it will be probably around, the observed incidence will be similar to the expected incidence. So the first bottom line, lack of effect on colorectal cancer mortality in the trial at 10 years may also be related to the fact that the colorectal cancer mortality observed in the usual care group in general is lower as compared to what we expected. So the second thing that I would like to focus our attention to is also the effect on colorectal cancer incidence, because when we analyzed observational trials, we were seeing amount decrease in incidence of colorectal cancer of, let's say, 60 to 90 percent. But most of these studies were observational, and this design limits the value of the trial, of the studies. So what we observed, and it's shown on this slide, in a moment, I will click on, okay, is that our, in the per-protocol analysis, when we compare screening attenders and the usual care group, the reduction is around 30 percent. But what was previously done in many other observational trials is this comparison. So this was the comparison between, let's say, usual care group and those who are examined with colonoscopy had no cancer, and it was comparison only of incidence of post-colonoscopy cancer as compared to what was expected in the population. And if you just look at these numbers, this corresponds to around 70 percent reduction if you take it like that. But what you really need to take into account is that we detected colorectal cancer in 0.5 percent of population already at baseline. It means that if you take into account prevalent cancers and the risk at 10 years, the maximum the maximum expected colorectal cancer incidence reduction would be no more than 58 percent, even if you would protect every single cancer after colonoscopy. So no post-colonoscopy colorectal cancer would translate into no more than 58 percent reduction. So we think that this is important to remember and to take into account also prevalent cancer detected at baseline colonoscopy. So this is the difference as compared to observational science. Thank you so much. So first of all, I want to congratulate all the Nordic investigators and Michael and say how much we appreciate your being here to talk with us about the really American reaction. I am only sort of aware of the American reaction. I don't know what the other international reaction is to the study, but all of us are aware of what an amazing undertaking, an enormous undertaking this is and that the study is beautifully reported and was beautifully executed. It's the first randomized comparison of colonoscopy to no screening. And we want to say congratulations to you and particularly thank you for joining us this morning. So I think one of the issues really that's come up for us is just trying to cope with the reaction in the United States from the media. And in the abstract of the study, the result that was presented was the intention to treat or intention to screen result, which Michael explained the rationale for, which showed this 18% reduction in incidence of colorectal cancer, a 10% reduction in mortality that did not reach significance. So I think the reaction in the United States was, you know, so perhaps colonoscopy doesn't work. And this was created a lot of challenges for us because we are totally invested in colonoscopy as a screening test. Colonoscopy was approved by the federal government as a screening test in July of 2001. So we're more than two decades into the process. We have about two thirds adherence to screening in the US, and that has been largely by colonoscopy. So this was a very big deal, not only because colonoscopy plays a role in screening, but of course, it is used to essentially evaluate all patients who are symptomatic. It's pretty much used universally for surveillance. And it's also used to evaluate other diagnostic tests. So I think it raised a lot of questions about whether it works in general. And we heard lots of questions from patients. We heard questions from nurses about family members who were, you know, not wanting to have a colonoscopy and then sort of saying, well, I guess I don't need to have one anyway, since it doesn't work. And I think that although the intent to screen analysis is very important from a scientific standpoint, that in a country like the US, where the use of colonoscopy for screening is so well developed, that the per protocol analysis is arguably the most important. And the per protocol analysis, and I realize that there are some potential biases there that have to be dealt with, and which you guys did your very best to deal with, you know, the incidence reduction of 31% and the mortality reduction of 50% are sort of on the order of what we might expect based on previous observational studies. And so in our discussion with the media, we have very much tried to emphasize the per protocol analysis, because I think in a country where the use of colonoscopy screening is so mature, that the question that is probably most appropriate for most patients is, you know, if I undergo colonoscopy, what is the benefit that I will accrue from taking that step, we already know that a mailed out invitation is not going to work, because we have about a third of the population that won't undergo screening, even when they're invited to screen in the office by a physician, which is the main way that screening occurs in the United States opportunistic screening. So on the other hand, or in addition to that, the other element I think that has come up is whether the overall result of the Nordic trial is is disappointing this 18% reduction in incidence, because it basically puts the absolute reduction in incidence about on par with flexible sigmoidoscopy. And this is the trial that was just or the study just reported in the annals of internal medicine for flexing the flexing screening trials had slightly higher adherence rates, but yet the overall incidence and mortality reduction about 20%. And that looks about the same as colonoscopy. And so another question that we were commonly asked and had been discussed is, why bother with colonoscopy if it's no better than sigmoidoscopy as a screening test. And so the issue has come up, does it add something to detection by examining the entire colon and I think there are a number of lines of evidence that it in fact does. There are observational studies and again, these are subject to some bias, but I'm going to refer briefly to those. We have a US sigmoidoscopy randomized controlled trial, where the rules for which findings that sigmoidoscopy led to a colonoscopy were more liberal than some of the other trials and a greater number of patients underwent colonoscopy, where there was a reduction in cancer in the right side of the colon. And then as another line of evidence, we have seen that higher ADRs are associated with protection throughout the colon. We often hear this rule from the Northern Kaiser studies of a 3% reduction in incidence for 1% increase in ADR. And the risk or the increase in protection seems to be higher in the distal colon at 4%, but there is a significant reduction in incidence with higher ADR in the proximal colon. So there is quite a bit of evidence that indicates that colonoscopy does add to sigmoidoscopy. So this is Herman Brenner's meta-analysis of observational screening studies. You can see that there aren't a huge number of these and in general, the incidence and mortality reduction are lower in the distal colon, but there are reductions that occur in the proximal colon. And there's been a lot of variation across these different screening studies, but overall another line of evidence that we would expect there to be an improvement with colonoscopy. So another question that's come up is could potentially there be variations between countries in the results of colonoscopy? And looking at the Nordic trial, I would say that the trial itself provides some evidence that there is variability between countries because of the differences that we see in the participants. There are differences in adherence rates, different substantial differences in the use of sedation, a threefold difference in adherence, a ninefold difference in the use of sedation. And I think, you know, there this reduction, a difference in reduction in incidence in the PER protocol analysis between Norway and Poland, that's a threefold difference. Those are really substantial differences. And so this has raised questions about whether or not the results can be extrapolated to all other countries. We also have differences with regard to sedation. We sedate almost everybody in the United States. I realized that the literature on whether or not sedation improves detection is overall mixed and I would say overall not strongly suggested that it makes a difference. But it may make some difference with regard to willingness to follow up. It's certainly a difference in practice. The overall ADR in the Nordic trial, 31%. We may have been at about that level, but we currently, according to registries, are at about 40% in the United States. Probably all around the world, we are currently improving with regard to detection. So this has raised the question and the discussion of whether or not colonoscopy could have, the benefit of it could have been underestimated by the Nordic study. And I think you guys very clearly stated in the trial that there was some potential that the benefit was underestimated. But of course, in dealing with the media, again, the details of the study and what is said in the paper are overshadowed by what the central result is. So some of the questions that we have had, first of all, the duration of follow-up was 10 years. In the original report of the results of the colonoscopy, it was said that the main results would be expected in 15 years. And as we all know, it takes time for polyps to turn into cancer. So it's possible that the separation between the colonoscopy groups and the usual care groups could increase with time. Secondly, with regard to quality issues, 29% of the endoscopists in the study had ADRs below the current recommended threshold of 25%. I think there are a lot of units in the United States that don't see that anymore, and perhaps would not be comfortable with endoscopists continuing to do colonoscopy without remediation, without correction for lower ADRs. 17% had sequel intubation rates that were below 95%. This is something that we also, at the current time, especially with sedated colonoscopy, simply don't see. I think you guys presented the protocol for surveillance, but we don't know the percentage of patients that actually underwent surveillance. There is some evidence that surveillance contributes to incidence and mortality reductions during follow-up. There is not an apparent stage shift in the screen detected cancer. And my understanding from looking at a lot of trials is that when we do colonoscopy in asymptomatic patients, we expect to see a stage shift. The asymptomatic patients, when cancers are detected, have earlier Duke stages than cancers that are present in symptomatic patients. Michael commented about contamination in the usual care groups. When I had read the paper, I had seen comments that were primarily qualitative, so it's useful to know that the contamination rate was less than 2%, the contamination rate of lower bowel endoscopy. Overall, it seems, and I think with any trial that is as big as this, that's operated in multiple countries, that there are a lot of moving parts that could potentially affect the results. It appeared from the study that the invited patients were at higher risk in colon and the usual care group at higher risk in Norway, which may account somewhat for some of these differences. So in summary, with regard to the U.S. interpretation of this, and I think a lot of the concern has to do that we are fully invested and fully committed to screening. We feel that the per protocol analysis is of greatest interest to us, and that's what we've tried to emphasize with media and with patients. I think that based on some of these factors, it's reasonable to suspect that the intention to treat and per protocol analyses underestimate the benefits of colonoscopy. We don't feel that the data are currently such that from one study, it should derail the approach that we take to screening. And part of that approach in opportunistic screening still involves what we call sequential offers of screening, where colonoscopy is offered first, and then if patients decline that, they're offered non-invasive testing. That's not the only approach that's used. There's also multiple options, but we think that this is still an acceptable option. So I think a lot of this, you know, that has happened has had to do with the interpretation in the media, and us trying to provide a level of confidence to those people who are undergoing colonoscopy, and I appreciate very much what you guys have done with the study. We appreciate your being here to discuss this with us and have an exchange about it. So thank you. Jason, I want to ask you, because you don't have slides, do you have any comments that you would like to make before we start the back and forth? Thanks, Doug. As usual, you did a great job summarizing, I think, the American perspective on this study. I just want to add my congratulations to the Nordic investigators. This is an incredible undertaking, and they did a great job reporting the results of this very important study. So, you know, kudos to you all. We do have a further brief question in the Q&A along the lines that Dr. Rex referred to about the stage of baseline cancers and the contribution that would usually make to lowering risk of mortality. Could you comment further, Dr. Breithauer? Yeah. I think, Doug, you summarized it correctly, and we can talk about the interpretation of this and where we would like to put our emphasis on. But with regard to the cancer stages, it's exactly as you say. Well, there is some stage shift, but it's not as pronounced as I think both you and we would have expected, and we are uncertain why this is. There is some more early stage cancers than late stage cancers in the screening group, but our expectation would be that it would be a more pronounced stage shift. So, and we are uncertain, you know, I mean, we are uncertain about why this is. Michael has alluded to the comparison between the screening-deducted cancers and the cancers during follow-up in the people who attended, which is, of course, very interesting, and he showed you a table. So, there may be something about the people in Poland who showed up who were at a higher risk than we expected. So, maybe there was some self-selection, which of course happens in any trial, can happen, which may be a good thing because you detect cancers in people who are at high risk, or it may be a bad thing because from an analytical side, because it needs longer time to show an effect because you have lead time problems there, and the curves are crossing at a later stage than what was expected. And we tried to emphasize that in this discussion of the paper that we said, and I think, Doug, you referred to it. It may well be that the effect on CRC death takes another two, three, five years to be apparent, and I think that may very well happen, and I certainly hope that it will, but nobody knows. We'll see, and we will do our next analysis in two years' time, and then we will do the final one in another two years' time, and we will all see. Let me just emphasize one thing, Jason, with regard to the quality. Number one is we all know that the culture of doing colonoscopies is very different in different countries with regard to sedation. I think, and we can have our own webinar about that. I think what we should look at is the quality of the examinations, so the quality parameters that we all use, which is cecum intubation, which is bowel prep, which is ADR, of course. There, I think, as you alluded to, Doug, we are doing pretty good, and you remember, this is 10 years ago we did all those colonoscopies. So I think sedation or not, I mean, it's culture, and we have asked all the people if they would like to come back, and virtually everybody says, yeah, we are happy, we want to come back if we get an invitation. So I think the quality is okay. Go ahead, Jason. Michael, I wonder if you and the other Michael can expand a little bit on the invitation process, because that was reported by Michael Kaminsky back in the endoscopy paper where, you know, the invitation goes out, and as I understand it, it says, you know, you have an appointment for a colonoscopy on such and such a date, you know, and please call to confirm. And if they don't respond, there's a follow-up reminder. I don't know if there's more phone calls or anything to call to check in on the people who don't respond. But, you know, if I got a letter like that in the mail, that would seem quite shocking to me, you know, especially, and screening colonoscopy is something that's part of the public consciousness in the U.S. You know, we have Katie Couric having it on TV, now Ryan Reynolds is talking about his colonoscopy. But in Poland and Norway 10 years ago, how would people react to a letter saying you have an appointment? And, you know, how much does that contribute? And a follow-up on that, you know, to your point you made earlier about maybe high-risk people in Poland agreeing. You know, if I was sitting at home having some symptoms or I had a brother or a family member had colon cancer, I can imagine that if I got that invitation, I'd be like, well, that's great. You even, you offer free parking. You gave them a note saying you have the day off from work. You gave them free bowel prep. No cost for the exam, right? So they get this letter, and if they have a family history or symptoms, they might jump at the chance. And if they don't, maybe they're like, this is really weird. Why are we getting an invitation for a colonoscopy? Can you expand a little bit on that invitation, perhaps a reaction to that? Yeah, for sure. I think that I can respond to your question. So I think that actually, like, I mean, the situation that you have in the U.S., I mean, with lots of media, I mean, coverage on colorectal cancer screening, I mean, it creates probably more positive, let's say, background to receive an invitation to colonoscopy. But what we think is really, really nice with this trial, it also shows how, I mean, different societies behave in or respond to different offers. Because in fact, I mean, the point is that this is the first trial in which the invitation system was exactly the same in different countries. And it shows the difference that you observe in Norway and in Poland reflects, I mean, probably the general, like, attitude of society toward, like, medical care and so on. On the other hand, I mean, the invitation process was also really thorough. It, apart from, like, invitation to colonoscopy with a pre-specified date, it included also an invitation folder, which was like a colorful folder explaining all the benefits and risks of colonoscopy. And it was the same in both countries that we are discussing about. It was also followed by the reminder six weeks after. And we also tried to, I mean, if people were uncertain at certain states, we tried to contact them more and more. So I think that we did lots of efforts to communicate with people who were invited to screening. But apparently in European countries, they are not so responsive. On the other hand, you should also bear in mind that when you invite people in U.S. through, for example, GPs, these are people who are already present a little bit of healthy screening bias because they are showing up. They are making visits to their general practitioners. Whereas we were trying to reach everybody in the trial region. Therefore, also some people who are really not in a good shape or not taking lots of care about their health. So this probably reflects what may happen if you're taking into account, let's say, entire population, not only like a selected part of the population, which is probably more responsive to them. So this is the response with regard to the invitation process. And you had also a comment on sedation, I think, practices. And this is also quite interesting as well, because if you have a look on the sedation practices, the lowest sedation rates were in Norway, where the highest participation rates were observed. Whereas the lowest participation rates were observed in the countries with the highest sedation rates. So it's probably not the question of sedation itself. It's a question of what people are prepared for and how you perform your colonoscopy. I agree. And I noticed that it's almost a paradox that Norway has the lowest sedation rates and had the highest participation rates. So, I mean, and that's a fascinating thing, because I've always wondered about Norway, what the reputation of colonoscopy is. We've found it to be a huge problem in the United States and because people don't like to be in pain. And they talk very badly about procedures in which they experience pain. And when they talk badly to their neighbors, this is how people get their impression. So we've reached a place in the United States where basically nobody is afraid of the procedure. A lot of people don't like the idea of having the bowel prep or perhaps taking some time off work. But we've definitely moved completely away from fear of the procedure. And the procedure has a reputation as as as being not a not a big deal. So I don't think that that's the major issue. I only brought that up to just to point out that there there are these cultural differences that have to do with sedation and there may be cultural differences. It's conceivable that we are at different points in time with regard to the quality curve, too. I'm not really fully ready to accept that that there isn't an issue with quality when you have almost 30 percent of the endoscopist, you know, with ADRs that are that are fairly below the standard. And I think the standards are easy to hit. They're not hard to hit, especially in that age range where where the patients are. And I realize that things have improved in Europe and they've improved in the United States. But I I don't totally accept that that's not something of of an issue. And I think that so anyway, I'm mostly bringing up the sedation thing just to point out that colonoscopy varies, varies a lot. And it may be a little bit hard to translate what happens in the Nordic study to the U.S. or to Germany. Even we just don't we don't know fully. So, Michael and Michael, there is a question online to follow up on the comparison between these results and those in the sigmoidoscopy trials, both individually and in your recent publication. And can that difference be attributed to your unexpectedly low prevalence of cancer? Or do you have other explanations for those differences? It's a very good question. And I don't think we have I don't think we have a good answer to that. I mean, the the observation is just as as you had on your slide, Doug. I mean, it appears that appears the effect appears to be rather similar with the data we have from the Nordic trial last week. And then that paper in annals with the four four randomized trials. Now, in that paper in annals, there were no protocol analysis. So it would be nice to do that as well. And I think that's planned for the future. It's a little bit more difficult because you have the four trials together and then you need all those confounding variables to adjust for. It's protocol analysis is difficult to do. But having said that. It's interesting. And somebody asked me a couple of days ago, well, would you recommend that we go back to FLEX-SIG? And I said, no, I mean, I did I did more than 3000 sigmoidoscopies in the NORCAP trial, which is one of the FLEX-SIG trials. I didn't like it. I think it's it's it's dirty. It's messy. It's it's it's it's less comfortable or more uncomfortable for me for the endoscopist to do that. So I don't think we should go back to that. Why is the effect of colonoscopy not larger? How should we cope with this with regard to guidelines? I am not sure at this point in time. I don't know. It's just the observation is out there. I don't know how we should handle that. I think at least we shouldn't wait for some more years of follow up for the Nordic trial to compare really apples with apples, because the FLEX-SIG is 15 years. But if if it then holds, I'm not sure. I don't know. Go ahead. If I may add on on top of that, I think that we should look at two things. So first, I mean, is that's a larger difference for mortality reduction. And this is something which may be partially explained by lower than expected colorectal cancer mortality nowadays. So please remember that SIGMA trials were done, let's say, 10 to 20 years before the Nordic trial. So it means that the risk of dying from colorectal cancer in the usual care group was much higher as compared to current standards. So it may mean that the difference that we are seeing between both treatment or both screening modalities for colorectal cancer mortality reduction just simply reflects the fact that the benefit is less because we are dying less of colorectal cancer. So this is this is one thing. And the other thing for colorectal cancer incidence, it may simply reflect the time of follow up as already mentioned by Michael. In FLEX-SIG, we have 15 years follow up. In Nordic, we have 10 years follow up. And with higher detection rates at the baseline and the higher prevalence at the baseline, it's difficult to see the decrease in incidence so so quick in time. Please remember 0.5 percent prevalent cancer and 1.2 percent in the usual care group after 10 years follow up. So we would never reach 80 or 90 percent reduction with that. Thank you, Jason, there's a question online, presumably from Europe, about who can have screening colonoscopy in the US. Does it depend on insurance? Is it free, et cetera? Can you update our audience there? Sure, it's a great question. Well, so colorectal cancer screening is covered by insurance under what's called the Affordable Care Act. So anybody who has insurance can get any of the recommended screening tests done for free. Now, not everybody in the US has insurance, unfortunately. And there are some programs that are offered in various areas to provide colorectal cancer screening to people who don't have insurance. Often it's done with other tests like FIT, but it varies by region. I don't know if Doug has anything to add to that. No, I just think, you know, generally the great majority of the population now age at age 45 and older that wants to have a colonoscopy can can get one. There are other screening options that that are available, but it's you know, it's it's it's been very widely available, continues to be. Certainly. Well, the other options are also challenged by follow through to the colonoscopy, not to the same degree as enrollment first time around. But that's a challenge that we do deal with for both FIT and the more advanced testing. I wanted to ask a couple of quick questions of the Nordic folks. You know, we talk about the, as Doug said, about 29 percent of the Nordic colonoscopies had an ADR below 30 percent or below 25 percent, excuse me. But that was among the endoscopies who did at least 30 procedures. So I don't know what proportion of the procedures were covered by that. Do you have any sense of the quality of class in the other group? Like, did you lump them together and see what the overall ADR was in that other group of endoscopies? And second quick point, I think you've got a great opportunity to look at the association between sedation and colonoscopy quality metrics. I know you did some of that in your 2016 paper, Michael, but you could look at adenoma detection in the right colon. I don't know if you can get an adenomas per colonoscopy, but I'd be particularly interested in the adenoma detection in the right colon. There's a couple, as Doug said, the literature is mixed. There are a couple of recent papers that say that the ADR does go up significantly when sedation is used. Obviously, the use of sedation is not randomized in these studies, nor in your study. So there are potentially some biases. But I think it'd be really interesting to look at sedation. So if you could comment on those two issues. Yes, thank you. Number one, sedation, we have looked at that and there is no association. So there is no, the ADR is not higher, not lower with or without sedation. There's no association there, very clearly. And not for the other performance indicators either. Number two is with regards to the variation in ADR, there is variation. And we have reported that duly and transparently, I think, in that JAMA internal medicine paper in 2016. Now, going back to 2010, Michael Kaminski here was really the guy in the world who made, who had that first paper in Nijm telling us that ADR is important. Which was the start of what we are all now concerned with and talk about all the time. Look for polyps. We have training courses. We have, and we all agree on that, that that is important. So I can assure you that in the trial, there was a lot of focus about high quality. And most of the people who are on that, on the curves in that 2016 paper, most of the people who are on the higher end were the people who did most colonoscopies. And that's not apparent in the paper. Most colonoscopies in trial. The folks on the lower end, and we still worked on these people, but it's hard. As you know, it's hard to get everybody above that threshold, whatever you like it's going to be. It's hard to get everyone above there. The people there on the lower end, we worked with them. They improved, but they're still, it's not optimal. We agree with that. But these people have done relatively few procedures. Now, during the revision process in the New England Journal, and at least one of you know this, but I won't go into details because this is confidential information between the authors and the reviewers in the journal. We were asked and we prepared and we performed an associational analysis of ADR and cancer after screening. And there was no association. But of course, the event rate is very low. And the number of colonoscopies in this study is far too small. I mean, there would need to be a big difference to be able to show that. Because we have only 13,000 or so colonoscopies in the trial. And our colleagues' paper has 380,000 and Michael's has about 250,000 colonoscopies. So with this study, I don't think we will have any news regarding the association of ADR versus interval cancer. It's just too small to establish that. You mentioned the review process with NEJM and the confirmation or absence of that with some of the findings they expected. One of the major societies in the U.S. has strongly criticized not the study, but the treatment of the study by media. And I think that's rightly so. Many would say that it could be entirely anticipated from the structure of the abstract and the nature of American media to look at an abstract and not dig deeper and not even pursue the per-protocol analysis. I think it's a fair question or critique for the NEJM as to how they may or may not have requested revision, not to change the data or the emphasis, but to elucidate a little bit more in the upper-level initial presentation of an abstract. Any thoughts or comments from any of the panel on that point? I can comment on that. In fairness, as we talked about, the textbook recommendation is intention to treat always first for all randomized trials in medicine. And I think it's a good principle. So in fairness, I mean, that's what everybody does. And I think everybody should do. However, then for this trial and other trials where the participation is not 99% or 96%, but it was 42, of course, per-protocol is important. And so with that recognition, I think, yeah, maybe there should be a sentence about the per-protocol analysis in the abstract. There wasn't space, you know, it's 250 words. And the priority is where it is. And it should be on the intention to screen. And that's how it was. And I think then it's up to us as experts. And it's up to the media to relay the whole body of information. As we all, I think, try to do. And we have emphasized that all the interviews we had in the last week, you have to look at both. You have to look at both analysis. And you can't just take the one and then think you have the whole message. It's not fair. And it's not correct. Great, thank you. There's one last technical question I'm going to ask Doug to comment on. And that's regarding more recent technical advancements with high-resolution white light and use of virtual chromoendoscopy or narrowband imaging. If any of those were available or not available in the study, it would be of interest and comments on what that might be doing to our current detection. OK, so I think that I can comment on that because I also had some comments on ADR. So all the scopes which were used for the purpose of the trial, high-definition scopes, so no standard definition scopes, they were bought for that purpose. And all those scopes had available narrowband imaging at the time. I mean, it's a different system that we are having right now. But it was the thing that was available at the time. And I really want to emphasize that the overall quality, if you look at the ADR of 31% from European perspective, is pretty high. I mean, we have lower colorectal cancer incidences compared to US. And we are not seeing, even in the most experienced hands, as high ADRs as you are seeing in the United States. So something like 70%, which happens in US population, we are not there. I mean, there is no single endoscopist in my country which reaches 70%. And I presume that there are some very good endoscopists. 50% is doable for us, but 70%, not at all. So 31%, I think that is really high. And also, I mean, the distribution of colonoscopies performing colonoscopies within the trial is that those with really high ADR values were doing most of colonoscopies. So I don't think that we should put so much emphasis on the, let's say, suboptimal ADR. Because overall, we are meeting standards. And also, if you look at the cancers after that, I mean, we cannot try to analyze them. But what you see is that there is not so many post-colonoscopy colorectal cancers in the screening group, which means that we are probably doing really fine. Thank you very much. Our hour is drawing to a close. And we greatly appreciate the contributions and the nature of this study and its size and complexity. It looks like that's the end of our broadcast. And we look forward to 15-year data. Thanks very much for joining us today, guys. Appreciate it. Thank you, everybody. This was very nice. I appreciate it, too. And I hope we can continue our discussion the next time we see each other. It was lovely. Thank you so much. Yeah, take care. Have a great day. Hope to see you soon. Have a great day. Go to bed, Jason. Take care. Thank you all for joining us.

webinar

American Society for Gastrointestinal Endoscopy

study

colonoscopy screening

colorectal cancer

death

Nordic European Initiative on Colorectal Cancer

participants

random assignment