A finding of any type of polyp in the colon increases the risk for colorectal cancer (CRC), according to new findings from a large Swedish study.
At 10 years, the cumulative colorectal cancer incidence was 1.6% among patients with hyperplastic polyps, 2.5% among those with sessile serrated polyps, 2.7% for tubular adenomas, 5.1% for tubulovillous adenomas, and 8.6% for villous adenomas, as compared with 2.1% for the control group.
However, a higher risk for colorectal-related death was only observed in patients with sessile serrated polyps, tubulovillous adenomas, or villous adenomas.
The study was published online March 16 in Lancet Gastroenterology & Hepatology.
An important clinical implication from this study is that colonoscopy surveillance after polypectomy needs to be further tailored based on the baseline colonoscopy findings, the authors comment.
“Our findings suggest that more intense surveillance may be considered for individuals with villous adenomas and sessile serrated polyps, whereas the interval can be extended for individuals with tubular adenomas or hyperplastic polyps,” elaborated study author Mingyang Song, MD, ScD, assistant professor of clinical epidemiology and nutrition at the Harvard T.H. Chan School of Public Health in Boston, Massachusetts.
“Second, our findings also highlight the importance of quality improvement of colonoscopy for colorectal cancer prevention, in order to minimize the risk of missing/incomplete resection,” he added.
“Key Contribution to the Field”
This is the first study to assess both colorectal cancer incidence and mortality after polyp removal according to histopathological subtypes, and it therefore “marks a key contribution to the field,” write Reinier G.S. Meester, PhD, of Erasmus University Medical Center in Rotterdam, the Netherlands, and Uri Ladabaum, MD, MS, of Stanford University School of Medicine in California, in an accompanying editorial.
The main contribution, however, is the analysis of isolated serrated polyps, which includes hyperplastic polyps compared with sessile serrated polyps, they point out, as only a limited number of studies have assessed long-term cancer outcomes in these patients.
These findings show “for the first time that sessile serrated polyps are independently associated with colorectal cancer mortality,” Meester and Ladabaum write. “Although patients with this type of polyp were relatively rare in the study and accounted for less than one in 50 cancer cases, this number might represent underreporting.”
Overall, the results showed that none of the patient groups (who had polyps found) had risks that were lower than the general population. “This finding should prompt us to ask why polyp removal did not decrease the risk to below general population risk, and whether improved inspection, detection, and resection methods can ever achieve such an outcome,” the editorialists write.
However, they add that the “fact that associations with cancer were stronger before 2003 is consistent with the optimistic thought that colonoscopy quality improvements over time could be preventing more cancers.”
Speaking to Medscape Medical News, study author Song said there may be two major reasons why polyp removal did not decrease that individual’s risk of developing CRC to the risk seen in the general population.
“First, individuals with polyps have an inherently high risk of developing colorectal cancer,” he said, “And while removal breaks the carcinogenic process, they may still at risk of developing recurrent lesions — which again highlights the importance of surveillance.”
The second reason is that poor-quality colonoscopy may result in incomplete removal of the polyps or missing some important lesions. “This puts patients with polyps at a higher risk of developing cancer,” Song explained. “This is particularly relevant to our study because of the use of the historical data, which showed a higher risk elevation for earlier years than later years.”
While progression from adenoma to cancer accounts for most cases of colorectal cancer, about 20% to 30% of cases are associated with sessile serrated polyps. As they are primarily found in the proximal colon and have a subtle and flat endoscopic appearance, sessile serrated polyps can be easily missed or incompletely removed endoscopically, the authors write. This results in their disproportionate contribution to so-called “interval colorectal cancers” that are diagnosed in patients who are still well within the recommended surveillance periods following polypectomy.
Risk Higher in All Groups
In this study, Song and colleagues analyzed data on 178,377 patients with colorectal polyps who had undergone polypectomy from Sweden’s nationwide gastrointestinal ESPRESSO histopathology cohort (1993–2016). They were matched with 864,831 individuals from the general population who served as a reference group.
At a median of 6.6 years of follow-up, there were 4278 incident colorectal cancers and 1269 colorectal cancer related deaths in patients with polyps, and 14,350 incident colorectal cancers and 5242 colorectal cancer deaths in the reference group.
Individuals in all polyp groups except for those with hyperplastic polyps had a higher incidence of colorectal cancer (P <. 0001) as compared with the general population, but only those with tubulovillous and villous adenomas had a greater mortality.
After adjusting for potential confounders, all polyp subtypes had a positive association with colorectal cancer incidence, including hyperplastic polyps (multivariable hazard ratio [HR], 1.11; P = .02), sessile serrated polyps (HR, 1.77; P < .0001), tubular adenomas (HR, 1.41; P < .0001), tubulovillous adenomas (HR, 2.56; P < .0001), and villous adenomas (HR, 3.82; P < .0001).
For colorectal cancer mortality, there was a positive association for sessile serrated polyps (HR, 1.74; P = .02), tubulovillous adenomas (HR, 1.95; P < .0001), and villous adenomas (HR, 3.45; P < .0001), but not with hyperplastic polyps (HR, 0.90, P = .20) or tubular adenomas (HR, 0.97, P = .63).
Information regarding risk factors such as polyp size and multiplicity, quality, and indication of endoscopy were not available; nor were lifestyle behaviors such as smoking, obesity, and diet. Thus, it is unknown how that may have affected the data. “We think they may have somewhat strengthened our observed associations, because individuals with risk factors are more likely to develop polyps and [are] also at a higher risk of colorectal cancer,” said Song. “But this is certainly a limitation of our study.”
The study was funded by the US National Institutes of Health, the American Cancer Society, the American Gastroenterological Association, and the Union for International Cancer Control.
Song reports personal fees from Shire, Synergy, and Bayer Pharma, and grants from AstraZeneca, Takeda, and Gelesis outside the submitted work. Two other coauthors also report relationships with pharmaceutical companies. The complete list can be found with the original article. Ladabaum reports personal fees for advisory (UniversalDx, Lean) and consultant (Covidence, MotusGI, Quorum, Clinical Genomics) roles.