colonoscopy does decrease colon cancer incidence

 There has never been a randomized clinical trial confirming the benefit of colonoscopy, though there are RCT trials suggesting that sigmoidoscopy is beneficial. A new study (the NordICC trial) tried to fill that gap, finding that at 10 years follow-up,​ there was a decrease in colon cancer associated with colonoscopy (see colon ca colonoscopy dec ca at 10yrs nejm2022 in dropbox or DOI: 10.1056/NEJMoa2208375) 

 

Details: 

-- a pragmatic randomized trial of likely healthy men and women age 55 to 64 selected from population risk registries in Poland, Norway, Sweden, and the Netherlands between 2009 in 2014 

-- those chosen through the registry were offered one-time screening colonoscopy (none had had colonoscopy before) versus the group with no invitation (usual care group). 

    -- there were no organized colorectal cancer screening of any kind in these areas at the beginning of this trial

-- median age 59 years, 41% women 

-- there was a pretty large difference in screening population by country: 61% participated in Norway, 33% in Poland, and 40% in Sweden (no comment on the Netherlands)

-- colonoscopy characteristics: 91% had at least a good bowel preparation, 97% had cecum intubation, and 31% had adenoma detection  

-- primary endpoints: risk of colorectal cancer and related death

-- secondary endpoint: death from any cause 

-- follow-up data was available for 85,585 participants: 28,220 in the invited group and 11,843 of them underwent screening (42%); and 56,365 in the usual care group 

-- median follow-up 10 years 

 

Results: 

-- colorectal cancer was found in 62 people (0.5% of those screened)

    --early stage cancer (stage A or B) found in 0.38% of the invited colonoscopy group and 0.44% in the usual care group 

    --late stage cancer (stage C or D) found in 0.40% of the invited colonoscopy group and 0.50% in the usual care group 

--adenomas were detected and removed in 3634 people in the screened group (30.7%)

 

-- per-protocol analysis (ie those who agreed to the colonoscopy): 

    -- risk of colorectal cancer at 10 years: 0.84% (0.68-1.00) in the invited group and 1.22% (1.13-1.32) in the usual care group, 31% risk reduction, RR 0.69 (0.55-0.83), highly statistically significant

    -- risk of death from colorectal cancer at 10 years: 0.15 (0.09-0.23) in the invited group, 0.30 (0.26-0.36) in usual care, 50% risk reduction, RR 0.50 (0.27-0.77), now statistically significant 

 

-- number needed to screen to prevent one case of colorectal cancer: 455 (270-1429) 

-- risk of death or any cause was 11.03% in the invited group and 11.04% in the usual care group, not statistically significant 

  

-- Adverse events: 15 participants had major bleeding after polyp removal (all could be stopped endoscopically), but only 1 perforation in the Netherlands and no screening-related deaths within 30 days after colonoscopy 

 

Commentary: 

-- colon cancer is quite prevalent: third most common type of cancer and second leading cause of death worldwide; it is responsible for almost 10% of annual global cancer incidence 

-- of note, a 2018 data review found that colorectal cancer mortality was decreased 15% with a guaiac fecal sampling (see https://www.nejm.org/doi/full/10.1056/NEJMsr1714643, or DOI: 10.1056/NEJMsr1714643) 

    -- some of the guaiac tests done had low specificity for cancer (esp those with stool rehydration), and the decreased mortality may partially be related to the relatively high use of colonoscopy (sort of a relatively gross screen by guaiacs and perhaps the decreased mortality was largely related to the large number of colonoscopies done after the positive screen??)

-- there are impressive data showing that sigmoidoscopy is beneficial, and it does seem reasonable to assume that a colonoscopy (which includes a sigmoidoscopy for the distal part of the colon) would be superior since the right side of the colon is visualized only by colonoscopy. 

    --However, the benefits of picking up and treating right-sided colonic lesions (beyond the reach of the sigmoidoscope) seems to be significantly less than left-sided ones, as found in several studies see  (see http://gmodestmedblogs.blogspot.com/2017/04/one-time-flex-sig.html , finding that the one-time sigmoidoscopy was effective for about 15 years after screening; and see colonoscopy right sided lesions not helped AIM2009 in dropbox, or DOI: 10.7326/0003-4819-150-1-200901060-00306). It would be useful in this context to have a study comparing sigmoidoscopy (much less invasive, costly, difficult for patients, safer) versus colonoscopy. In fact 4 large RCTs on sigmoidoscopy have found a 22% decrease in colorectal cancer and 26% decrease in related deaths (see the detailed review for USPSTF in JAMA; 2021; 325: 1978-1998) 

    -- it was notable that the NordICC trial above did find that for the 42% who had colonoscopy, there was  a 31% decrease in colorectal cancer and 50% decrease in colorectal cancer related deaths [but, there was a significantly higher 58% to 87% patient participation in the sigmoidoscopy trials, and these sigmoidoscopy results included all-comers and not just the ones who had a sigmoidoscopy. So hard to compare these results).

    -- in part, as a result of the RCTs showing sigmoidoscopy benefit, many parts of the world have not adopted the colonoscopy screening method

  

 Limitations: 

-- this is not a real RCT, none of the patients who were selected from the registry knew that they would be in a study, and those in the control group were never even aware that this study was being done. This may well be manifested by the fact that only 42% offered colonoscopy screening actually accepted it. In an RCT there would have been prior discussion of this being an acceptable procedure. For example, the RCTs for sigmoidoscopy had between 58 and 87% actually screened.

    -- also, those who elected to have colonoscopy in this study might be fundamentally different than those not participating, creating a selection bias. perhaps they thought there was a  problem with their colon, elected to get tested, but had a higher pretest likelihood of having a cancer than the general population, distorting the comparison 

    -- and perhaps most useful to us in this study is the people who actually selected colonoscopy, where there was in fact a 31% risk reduction for colorectal cancer incidence and 50% for cancer deaths (though, again, the role of selection bias?)

    -- the advantage of a pragmatic study such as this one is that it more truly reflects what happens in actual clinical practice (not a very high uptake of colonoscopies, as opposed to a real RCT where people are at least open to it)

        -- though there seem to be more colonoscopies done in some countries where colonoscopy screening has been promoted more (eg US), more so than in these 4 countries in the study where no screening was done before the study 

-- as a study in these 4 countries, there were likely some differences in what demographic and comorbidity variables documented within the countries; and the population understanding of colonoscopy (benefits vs risks) were different than in other countries where colonoscopies are done routinely. Again limiting generalizability

-- no data on distal vs proximal cancers, sex or age at screening, or even what type of adenomas they found (tubular adenomas vs more aggressive types??) 

-- this was a single colonoscopy. And, typically one finds improved results after a subsequent screening test, since the first one is picking up prevalent colon cancers (perhaps there a long time and more advanced); one of the markers of a subsequent colonoscopy would be not just the numbers of people with incident colon cancer (ie developed during the screening interval), but also if they were stage shift to an earlier stage and treatable colon cancer over time. The colon cancer stages in those diagnosed in the single colonoscopy NordICC trial were quite similar. 

-- the 10-year horizon for the study may be insufficient to find mortality differences. they do intend to continue the analysis of events into the future, for the 15-year outcomes

-- The age group of the study, 55 to 64 years old, is quite different from current recommendations and their results may not be applicable to those more than 45-55 or >65yo (the more standard ages of screening now)

-- we have no information about baseline patient factors that are associated with colon cancer: genetic ones (eg adenomatous polyposis syndromes, Lynch syndrome, perhaps those with BRCA1 and 2); inflammatory bowel disease; and (though some mixed data) obesity, insulin resistance/diabetes, eating red and processed meats, tobacco, alcohol, androgen deprivation therapy for prostate cancer, cholecystectomy, coronary heart disease, H Pylori infection, hemolytic e. coli infection, and even the microbiome (eg, see microbiome colon cancer JAMAonc2015 in dropbox, or doi:10.1001/jamaoncol.2015.1377, one of several studies implicating Fusobacterium nucleatum) 

 

so, an interesting study that further supports the benefits of colonoscopy screening. And, given that no colonoscopy screening had been done in these countries, it is not surprising that this study was designed this way. But the results reflect that this study is not so useful in our current colonoscopy culture, which is why the per-protocol subgroup (those who had the colonoscopy) may be a more reliable outcome measure given the low participation rate in NordICC. And in this group there was significant difference for both colon cancer prevalence and colon cancer-related deaths by that analysis.

-- to me,  this study raises again the question of the role of sigmoidoscopy, as noted in prior blogs. It would be really useful if a study were done and found that sigmoidoscopy outcomes mirrored those of colonoscopy and we could refer for this less invasive, cheaper, and lower risk procedure that has a much easier prep. We do have FIT testing as an acceptable screening. Would be great to throw in that as well into the RCT (FIT testing is also lacking rigorous RCTs, but is equally acceptable to colonoscopy as a screening method)…

 

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

For access to the dropbox, go to link: https://www.dropbox.com/sh/0bmvtita8mzms11/XDTwHySFFg

Then go to "clinic", then to "clinical stuff" for articles, or go to https://www.dropbox.com/sh/nyle22q1fn6lkpk/AAB9B2hBj5Kw4gtrJAkI-UF8a?dl=0 for the powerpoint presentations

 

to get access to all of the blogs (2 options):

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list

 

-- of course we do have other less-invasive options, such at FIT testing annually (see http://gmodestmedblogs.blogspot.com/2021/05/colorectal-cancer-screening-new-recs.html for the 2021 USPSTF guidelines)