Colorectal cancer (CRC) is the second most common cause of cancer death among Americans. The gold-standard screening test, colonoscopy, is performed on roughly 15 million people in the US every year. In 2021, because of rising rates of CRC in younger people, the US Preventive Services Task Force lowered the recommended screening age from 50 to 45. For certain patients, including those with a family history of early-onset CRC or diseases like inflammatory bowel disease, screening may be started even earlier than age 45.
Past research shows that colonoscopy is associated with as much as a 69% decrease in new cases of colorectal cancer and an 88% decrease in the risk of death from it. So when several news reports described findings from a recent randomized trial on colonoscopy as disappointing (see examples here and here), it came as a surprise to many — including some experts like us who combed through the study carefully.
Why is colonoscopy the gold standard for colorectal cancer screening?
As many people know, colonoscopy involves insertion of a long flexible tube with a camera at the end into the rectum. A highly trained gastroenterologist or surgeon steers the tube up the colon and looks for growths called polyps or other abnormalities, including early cancers. Not only can colonoscopy identify potentially precancerous polyps known as adenomas, but it also allows their removal. A similar (though less extensive) screening test called flexible sigmoidoscopy evaluates only the lower colon, and can miss cancers further inside the colon that tend to have a worse prognosis.
Other screening tests for CRC cannot locate or remove polyps. They look for blood or abnormal DNA in stool, which may signal a polyp or cancer. This includes fecal occult blood tests and fecal immunochemical tests, which detect blood, and Cologuard, which detects blood and abnormal DNA.
These key factors all make colonoscopy the preferred choice.
What does this new study on colonoscopy tell us?
Published in The New England Journal of Medicine, the study was a randomized controlled trial of 84,585 people between the ages of 55 and 64 in Norway, Poland, and Sweden who had not previously undergone screening. Participants either received an invitation to have a screening colonoscopy or did not receive an invitation. The researchers followed participants for 10 to 15 years to compare the number of colorectal cancers and deaths from CRC in each group.
Only 42% of people invited to have colonoscopy accepted the invitation. Data from everyone invited, regardless of whether they actually underwent colonoscopy, is known as an "intention-to-screen" analysis. As many news reports correctly noted, the intention-to-screen analysis showed an 18% reduction in later colorectal cancers and no significant reduction in deaths. Importantly, though, when only people who actually had colonoscopy were analyzed (known as a "per-protocol" analysis), colonoscopy reduced the number of colorectal cancers by 31% and of CRC-associated deaths by 50%.
Why are these findings so different?
In this study, a relatively low percentage (42%) of persons who were invited to undergo colonoscopy actually had the procedure, compared with a rate of 60% of adults in the US, where colonoscopy is broadly recommended. This low rate of participation is the major reason why the intention-to-screen analysis showed lower rates of detection and death than the per-protocol analysis. After all, you cannot find something if you do not look for it.
This isn't to say that the intention-to-screen analysis is meaningless. It draws our focus to real-world barriers leading to the low participation rate. For example:
- Did the investigators' invitation adequately convey colonoscopy's benefits, given the absence of an opportunity for a detailed, personalized discussion with a physician?
- Were people able to take time off from work for the procedure?
- Did some find the prep too daunting?
We know these factors can prevent people from following through with a recommended colonoscopy.
What else is important to know?
An important limitation of the study is that the rate of detection of pre-cancerous polyps by colonoscopy was lower (31%) than typical in the US (about 40%). A lower detection rate may translate to removal of fewer polyps, thereby weakening potential for fewer future deaths from CRC.
Finally, cancer does not grow overnight. Precancerous lesions can take years to develop into cancer if unchecked. Therefore, the authors plan to repeat their analysis in 15 years to learn if the benefit of screening colonoscopy increases over time.
The bottom line: Should you undergo screening colonoscopy or not?
Yes! Unequivocally, yes.
The take-home message of this study is that when people have screening colonoscopies, deaths from colorectal cancer are reduced by half. That's a huge reduction! And while the rate of CRC deaths is often the bottom line in studies, it's important to consider hardships associated with a diagnosis of cancer — financial costs, physical costs of surgery, chemotherapy, and radiation — as well as the suffering caused by the disease itself. This study teaches us that colonoscopies work quite well when performed, and that we still have work to do to make colonoscopy more accessible so that more people benefit from screening.