Colon Ca screening: risk-stratified approach

 A recent Chinese study suggested that a targeted, “risk-adapted” approach to colorectal cancer (CRC) screening might be the optimal one (see colon ca screening colo vs fit vs risk-stratified ClinGastrHep2022 in dropbox, or doi.org/10.1016/j.cgh.2022.08.003), a pre-print, pre-peer reviewed article

 

Details: 

--19,373 people from 5 provinces in China were randomized to one of 3 screening arms:

    -- one-time colonoscopy (n=3883)

    -- annual fecal immunochemical test (FIT test, n=7993)

        -- FIT test done was self-administered point-of-care qualitative test, with “a positivity threshold of 8mg Hb/g feces”, “comparable to the laboratory-based quantitative FIT”

    -- “risk-adapted screening”, whereby those judged to be at high risk were referred directly for colonoscopy, and those low risk to annual FIT testing

        -- “high risk” was defined as at least 4 on the following scale:

            -- age: 0 if 50-54yo; 1 if 55-64, 2 if 65-74

            -- sex: 0 if female, 1 if male

            -- history of CRC among 1st degree relatives: 0 if absent, 1 if present

            -- cigarette smoking: 0 if nonsmoker, 1 if current or past smoker

            -- BMI: 0 if <23, 1 if higher

-- age pretty evenly divided into 50-54, 55-59, 60-64 and 65-69 age groups (mean age 61); 58% female; 73% less than high school education; 63% BMI >23; 95% no NSAID use; 81% nonsmoker; 73% never alcohol drinkers

    -- the only difference between the screening groups was family history of CRC: 160 (4.4%) in colonoscopy group, 335 (4.3%) in FIT group, but 473 (6.2%) in the risk-adapted group

-- main outcome: detection of advanced colorectal neoplasia (CRC or advanced precancerous lesions)

    -- advanced adenoma was one with high-grade dysplasia, villous or tubulovillous histology, or size >10mm, including those with serrated lesions

-- total study was 3 years, with outcomes at the 3-year mark

 

Results:

-- participation rates (these all tended to decrease with age):

    -- colonoscopy: 42.4%

    -- FIT: 99.3%

    -- risk-adapted screening: 89.2%

        -- colonscopies were performed in 712 (49.0%) of those 1453 found to be high risk

-- overall positive FIT tests, over 3 years:

    -- FIT group: 22% (males 22%, females 20%)

    -- risk-adapted screening group: 39% (males 66%, females 21%)

-- followup colonoscopy rates in those with positive FIT tests:

    -- FIT group (low risk): 75.3%

    -- risk-adapted group: 74.1%

-- detection rate for advanced neoplasm (intention-to-treat, at the end of the 3-year study):

    -- colonoscopy: 2.76%

    -- FIT: 2.17%

    -- risk-adapted screening: 2.35%

        -- comparisons between groups (after adjusting for age, sex, and study center):

            -- colonoscopy vs FIT: OR 1.27 (0.99-1.63), p=0.053 (ie pretty close to being statistically significant)

            -- colonoscopy vs risk-adapted screening: OR 1.17 (0.91-1.49), p=0.218, not statistically significant

-- numbers of colonoscopies to detect one advanced neoplasm:

    -- colonoscopy group: 15.4

    -- FIT group: 7.8

    -- risk-adapted screening group: 10.2

-- costs for detecting 1 advanced neoplasm, about 6500 chinese yuan or $900 US, not much difference between the groups

 

Commentary: 

-- colon cancer is the 3^rd most common cancer diagnosis (1.9 million new diagnoses estimated in 2020) with the 2^nd most common cancer deaths (935,000 estimated deaths), per GLOBOCAN (Global Cancer Statistics)

    -- it should be noted that the overall prevalence of CRC is lower in China than in Europe and North America

-- international recommendations for CRC screening in average risk individuals vary a bit, but are typically colonoscopy every 10 yrs or FIT testing every 1-2 years

    -- BUT, actual CRC screening done is not so great in the US: overall 50% done in those 50-54yo, increasing to about 68% in those 55-64yo.  Screening was lowest in those with no health insurance, esp in those 50-64yo; those with income <$50K (60% screened), no health insurance (33% screened) and no primary care provider (33% screened)  (see https://www.cdc.gov/mmwr/volumes/69/wr/mm6910a1.htm). Not a great advertisement for the richest country in the world with the highest per capita health care spending….

 

-- there were several findings of note in the above study:

    -- participation rates for colonoscopy were remarkably low at 42%, but risk-adapted screening at 89%; and for those with FIT tests that were positive, the number of colonoscopies increased pretty dramatically to 75% (ie, it seems that the low intervention of FIT testing, when positive, led most people to get colonoscopies)

        -- and the overall numbers of colonoscopies needed to detect an advanced neoplasm was 50% more in the colonoscopy group than the risk-adapted group

     -- though the advanced neoplasm detection rate by colonoscopy surpassed the other 2 groups, by the end of the 3 years (with 2 more FIT tests done), the rates of advanced neoplasm detection was comparable between the groups

     -- there were some significant differences between men and women: women were more likely to participate in the risk-adapted screening, men were much more likely to have a positive FIT test. this latter difference in likely because men are at higher CRC risk, and smoking is more of an issue in men

 

Limitations:

-- though this was a randomized controlled study, there is (as typically the case) a bias inherent in the manner that patients were recruited: they used newspaper/television/stream media advertisement. so this recruitment reflects the skewed population of those accessing these media and willing to be included (ie, this was not a random assignment of all in a community). Also, another bias limiting generalizability of results is that those participating in the study (also pretty typical for studies) did receive added support (active follow-up by study staff) than what happens in clinical practice

-- no screening was done in those <50yo in this study, so not sure these results would apply to a younger cohort

-- they used a Chinese point-of-care FIT test, which they assert is comparable to the lab-based test, though I am not sure how accurate this is especially without comparisons in other countries (since China has a lower prevalence of CRC, which limits the positive and negative predictive value of the test, it would be helpful to have confirmation of the point-of-care test done in a higher risk country)

-- the risk model used has several issues:

    -- there is a binary cutpoint for each of these 5 criteria, and several really are continuous variables (should age 54 and 55 be different by 1 point?)

    -- they do not specify an age range for family history: should the same score apply if there is a family history of CRC in a 35yo relative vs a 65yo one? we do know that multiple people in the family with early onset CRC confers a higher risk vs a single older relative, but these groups are not differentiated in their high risk model

        -- there are some old and new genetic markers of increased CRC risk. Would these add to the risk prediction model? Do they vary significantly by geographical areas (and would they be different in non-Chinese populations?).  would these genetic variants add to the risk prediction if one incorporated into the risk model CRC at young age and/or CRC in first-degree relatives??  (ie would taking these latter factors into account eliminate or minimize the higher CRC risk anticipated by genetic predisposition???)

    -- should a past smoker have the same risk score as a current one? What if they quit 30 years ago?

    -- should there be a single cutpoint for BMI risk? The last blog on BMI suggested that BMI >30 was associated with higher risk than 25-30: see http://gmodestmedblogs.blogspot.com/2022/09/colon-cancer-bmi-as-risk-factor-for.html

-- this was only a 3-year study (they intend to perform further analyses in the future). But this makes it hard to understand the really important clinical outcomes of CRC detection and CRC mortality (there were too few actual cancers detected in this 3-year study to draw real conclusions)

-- there was also some “contamination”: 110 low-risk patients had colonoscopy without prior FIT testing, 229 high-risk had FIT first vs colonoscopy

-- the overall adherence to follow-up colonoscopy in those with positive FIT tests seemed to be low in this study (?why), and this could affect the generalizability of the results to other settings.

-- hard to interpret the cost data, since there are such differences in China or the countries with largely public health care systems vs the market-based/variable but very high costs it the US (eg, in Boston there can be a huge difference in the cost of colonoscopy depending on the site of the test/ability of insurers to limit or equalize costs)

  

So,

-- especially in light of the pretty poor uptake in CRC screening in the US (especially for colonoscopy), there might be a real benefit for age or other specific criteria for when to start screening and the frequency of CRC screenings. This trial did suggest that there was a higher colonoscopy rate in those considered to be high-risk than those just assigned to the colonoscopy arm, and in those who were informed that their FIT test was positive

-- there certainly needs to be more studies and more robust data to alter our current recommendations, including identifying and quantitating the important high-risk criteria, and showing that those at high risk benefit clinically from earlier and/or more intensive screening (specifically assessing decreased CRC incidence and mortality, and all-cause mortality), and that those at low risk do not have increased CRC incidence/mortality from less intensive screening

-- one added benefit of improved understanding of the relative risks attributable to CRC is that it might help patients alter those risk factors that are modifiable, though studies would need to confirm this

    -- however, it is likely these risk factors should be modified anyway, since they are likely to be many of the same ones (eg smoking, high BMI) involved in other high-prevalence diseases such as heart disease, other cancers, etc)

geoff

 

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

  

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