COLONOSCOPY LOWERS COLON CANCER MORTALITY, but only modestly. That is the researchers’ conclusion after ten years of follow-up of a large randomized trial from Europe.

The headlines have been jarring:

“Screening Procedure Fails to Prevent Colon Cancer Deaths in Large Study,” offered Bloomberg.

“In a Gold-Standard Trial, an Invitation to Colonoscopy Reduced Cancer Incidence but Not Death,” observed Statnews.

“Colonoscopy Fails to Reduce Rate of Cancer Deaths in Trial” volunteers realclearscience.com.

Have you heard about the 10-year follow-up of the large, multicenter, randomized Northern-European Initiative on Colorectal Cancer (NordICC) trial?

Researchers recruited healthy individuals in Poland, Norway, Sweden, and the Netherlands between 2009 and 2014. Most came from Poland (54,258), followed by Norway (26,411) and Sweden (3,646). The study authors did not include data from the Netherlands because of data protection law.

Should we ditch colon cancer screening?

Let’s get right to the recent report published in the New England Journal of Medicine:

The real-world risk of colorectal cancer and associated death rates appeared lower among people who had a single screening colonoscopy (compared with those not having the procedure), although only modestly so.

The researchers determined that the number needed to invite to undergo screening to prevent one case of colorectal cancer is 455. This number gives us an estimate of the effect of screening colonoscopy in the general population. Or does it?

The researchers determined outcomes based on intention-to-screen. In other words, they compared all subjects invited to have a colonoscopy screening with those who received no invitation or screening.

Here are some other study outcomes:

• The 10-year colorectal cancer risk was 0.98 percent for the screened individuals, compared with 1.2 percent for the unscreened subjects. While this 18 percent risk reduction sounds great, the absolute difference is small.
• The risk of death appeared similar in the invited group (0.28 percent) versus 0.31 percent in the usual care drop. Again, a nice relative drop (10 percent) but not particularly different in absolute numbers.
• The risk of death from any cause appeared similar in the invited and usual-care group members, at 11.03 percent and 11.04 percent, respectively.

My take — Colon cancer screening

The colonoscopy benefits would have been greater had more individuals had colorectal screening. Only 42 percent of those invited had a colonoscopy.

When the scientists adjusted for this low participation level, they discovered this:

Had all those invited to undergo screening had it, the 10-year risk of colorectal cancer would have dropped from 1.22 percent to 0.84 percent, and the risk of colorectal cancer–related death would have fallen from 0.30 percent to 0.15.

The NordICC is the first randomized trial to quantify the possible benefits of colonoscopy screening on the risk of colorectal cancer and related death.

Longer follow-up is needed to understand better whether colonoscopy is effective in this real-world analysis. Drs. Jason Dominitz and Douglas Robertson comment on the possible reasons for the low reduction in incident cancer and deaths observed in the NordICC study.

In an editorial that accompanied the study publication, they note that other studies suggest an approximate halving (40 to 69 percent) in the incidence of colorectal cancer and an up to seven-eighths (29 to 88 percent) decrease in the risk of death with colonoscopy.

We must take such observations in context: Historical non-randomized studies likely overestimate the real-world effectiveness of colonoscopy, given a lack of adjustment for incomplete adherence to testing. In addition, there is a tendency for healthier individuals to seek preventative care.

In the United States, colonoscopy is the main screening tool for colorectal cancer. In other places, participation rates may be quite different. In this sense, the real-world results in the USA may be closer to the results seen in the per-protocol analysis of this study.

I look forward to seeing if greater improvements in outcome emerge when the NordICC state has a 15-year follow-up. Finally, if the modest effectiveness of screening colonoscopy is real, can we afford to use this expensive screening tool? Or should we turn to less-invasive tactics such as FIT (fecal immunochemical testing)?

But Dr. Maas said that “around half of the patients in the study did not undergo colonoscopy, which may have negatively impacted the results. I will end with this observation about this study: Colonoscopy cannot possibly work if an individual does not have it. We need more evidence to guide us about the use of colonoscopy.

Thank you for joining me in the exploration of the effectiveness of colonoscopy.