From the AGA Journals

Low-risk adenomas may not elevate risk of CRC-related death

View on the News

What’s the best timing for CRC surveillance?

Despite evidence suggesting that colorectal cancer (CRC) incidence and mortality can be decreased through the endoscopic removal of adenomatous polyps, the question remains as to whether further endoscopic surveillance is necessary after polypectomy and, if so, how often. The most recent iteration of the United States Multi-Society Task Force guidelines endorsed a lengthening of the surveillance interval following the removal of low-risk adenomas (LRAs), defined as 1-2 tubular adenomas <10 mm with low-grade dysplasia, while maintaining a shorter interval for high-risk adenomas (HRAs), defined as advanced adenomas (villous histology, high-grade dysplasia, or >10 mm) or >3 adenomas.

Dr. Reid M. Ness, Vanderbilt University Medical Center, Nashville, Tenn.

Dr. Reid M. Ness

Dr. Duvvuri and colleagues present the results of a systematic review and meta-analysis of studies examining metachronous CRC incidence and mortality following index colonoscopy. They found a small but statistically significant increase in the incidence of CRC but no significant difference in CRC mortality when comparing patients with LRAs to those with no adenomas. In contrast, they found both a statistically and clinically significant difference in CRC incidence/mortality when comparing patients with HRAs to both those with no adenomas and those with LRAs. They concluded that these results support a recommendation for no difference in follow-up surveillance between patients with LRAs and no adenomas but do support more frequent surveillance for patients with HRAs at index colonoscopy.

Future studies should better examine the timing of neoplasm incidence/recurrence following adenoma removal and also examine metachronous CRC incidence/mortality in patients with sessile serrated lesions at index colonoscopy.

Reid M. Ness, MD, MPH, AGAF, is an associate professor in the division of gastroenterology, hepatology, and nutrition at Vanderbilt University Medical Center and at the VA Tennessee Valley Healthcare System, Nashville, campus. He is an investigator in the Vanderbilt-Ingram Cancer Center. Dr. Ness has no financial relationships to disclose.



Unlike high-risk adenomas (HRAs), low-risk adenomas (LRAs) have a minimal association with risk of metachronous colorectal cancer (CRC), and no relationship with odds of metachronous CRC-related mortality, according to a meta-analysis of more than 500,000 individuals.

Abhiram Duvvuri, MD, of the University of Kansas Medical Center, Kansas City

Dr. Abhiram Duvvuri

These findings should impact surveillance guidelines and make follow-up the same for individuals with LRAs or no adenomas, reported lead author Abhiram Duvvuri, MD, of the division of gastroenterology and hepatology at the University of Kansas, Kansas City, and colleagues. Currently, the United States Multi-Society Task Force on Colorectal Cancer advises colonoscopy intervals of 3 years for individuals with HRAs, 7-10 years for those with LRAs, and 10 years for those without adenomas.

“The evidence supporting these surveillance recommendations for clinically relevant endpoints such as cancer and cancer-related deaths among patients who undergo adenoma removal, particularly LRA, is minimal, because most of the evidence was based on the surrogate risk of metachronous advanced neoplasia,” the investigators wrote in Gastroenterology.

To provide more solid evidence, the investigators performed a systematic review and meta-analysis, ultimately analyzing 12 studies with data from 510,019 individuals at a mean age of 59.2 years. All studies reported rates of LRA, HRA, or no adenoma at baseline colonoscopy, plus incidence of metachronous CRC and/or CRC-related mortality. With these data, the investigators determined incidence of metachronous CRC and CRC-related mortality for each of the adenoma groups and also compared these incidences per 10,000 person-years of follow-up across groups.

After a mean follow-up of 8.5 years, patients with HRAs had a significantly higher rate of CRC compared with patients who had LRAs (13.81 vs. 4.5; odds ratio, 2.35; 95% confidence interval, 1.72-3.20) or no adenomas (13.81 vs. 3.4; OR, 2.92; 95% CI, 2.31-3.69). Similarly, but to a lesser degree, LRAs were associated with significantly greater risk of CRC than that of no adenomas (4.5 vs. 3.4; OR, 1.26; 95% CI, 1.06-1.51).

Data on CRC- related mortality further supported these minimal risk profiles because LRAs did not significantly increase the risk of CRC-related mortality compared with no adenomas (OR, 1.15; 95% CI, 0.76-1.74). In contrast, HRAs were associated with significantly greater risk of CRC-related death than that of both LRAs (OR, 2.48; 95% CI, 1.30-4.75) and no adenomas (OR, 2.69; 95% CI, 1.87-3.87).

The investigators acknowledged certain limitations of their study. For one, there were no randomized controlled trials in the meta-analysis, which can introduce bias. Loss of patients to follow-up is also possible; however, the investigators noted that there was a robust sample of patients available for study outcomes all the same. There is also risk of comparability bias in that HRA and LRA groups underwent more colonoscopies; however, the duration of follow-up and timing of last colonoscopy were similar among groups. Lastly, it’s possible the patient sample wasn’t representative because of healthy screenee bias, but the investigators compared groups against general population to minimize that bias.

The investigators also highlighted several strengths of their study that make their findings more reliable than those of past meta-analyses. For one, their study is the largest of its kind to date, and involved a significantly higher number of patients with LRA and no adenomas. Also, in contrast with previous studies, CRC and CRC-related mortality were evaluated rather than advanced adenomas, they noted.

“Furthermore, we also analyzed CRC incidence and mortality in the LRA group compared with the general population, with the [standardized incidence ratio] being lower and [standardized mortality ratio] being comparable, confirming that it is indeed a low-risk group,” they wrote.

Considering these strengths and the nature of their findings, Dr. Duvvuri and colleagues called for a more conservative approach to CRC surveillance among individuals with LRAs, and more research to investigate extending colonoscopy intervals even further.

“We recommend that the interval for follow-up colonoscopy should be the same in patients with LRAs or no adenomas but that the HRA group should have a more frequent surveillance interval for CRC surveillance compared with these groups,” they concluded. “Future studies should evaluate whether surveillance intervals could be lengthened beyond 10 years in the no-adenoma and LRA groups after an initial high-quality index colonoscopy.”

One author disclosed affiliations with Erbe, Cdx Labs, Aries, and others. Dr. Duvvuri and the remaining authors disclosed no conflicts.

Next Article: