Waiting five years for the next colonoscopy was as safe as three after risky polyps

Among people who have had high-risk polyps removed, waiting five years before the next surveillance colonoscopy was no worse than waiting three, judged by how many went on to develop colorectal cancer [s1]. That is the interim finding of EPoS II, a European randomised trial published in the New England Journal of Medicine on September 16, and it bears directly on guidelines that currently recommend the shorter, three-year interval [s1].

The stakes are practical. After an adenoma — a benign but potentially precancerous polyp — is found and removed, patients are placed on a surveillance schedule of repeat colonoscopies. Those procedures carry small risks, consume scarce endoscopy capacity, and are unpleasant enough that many people skip them. If a longer gap is genuinely as safe, it frees resources and spares patients without costing lives; if it is not, cancers are missed. The only way to know is to test it head to head.

The trial

EPoS II, conducted in eight European countries, enrolled patients with high-risk adenomas, defined as at least one adenoma 10 mm or larger, or with high-grade dysplasia or villous growth, or 3 to 10 adenomas of any kind [s1]. Participants were randomly assigned to undergo their first surveillance colonoscopy either 5 years after the polyps were removed or 3 years after; the three-year group also underwent a colonoscopy at 5 years [s1]. Three-year surveillance is what current guidelines advise [s1].

The primary end point is the cumulative incidence of colorectal cancer at 10 years, tested for noninferiority with a prespecified margin of 0.7 percentage points on the upper boundary of the confidence interval for the difference between groups [s1]. The results reported now come from a planned interim analysis after 5.5 years of follow-up, with inverse-probability weighting used to account for people who did not attend their 5-year colonoscopy [s1].

What it found

A total of 10,799 patients underwent randomisation, 5,398 to the five-year group and 5,401 to the three-year group [s1]. The five-year cumulative incidence of colorectal cancer was 0.77% with the less-frequent schedule and 0.82% with the more-frequent one — a difference of −0.05 percentage points [s1]. The upper boundary of the confidence interval for that difference was 0.68, inside the 0.7-point margin, which met the criterion for noninferiority [s1].

The number of cancer deaths was small in both groups and did not diverge: 5 patients died of colorectal cancer, 3 (0.06%) in the five-year group and 2 (0.04%) in the three-year group [s1]. The distribution of cancer stage at diagnosis did not appear to differ substantially between the groups — reassuring, because the fear with a longer interval is not only more cancers but later-stage, harder-to-treat ones [s1].

The limits

This is an interim analysis at 5 years of a trial whose primary end point is measured at 10 [s1]. The question that matters most — whether the two schedules remain equivalent once the full follow-up accrues — is not yet answered, and the statistical plan reflects that: the interim result is reported with a one-sided 99.12% confidence interval, with a wider allowance reserved for the final analysis to keep the overall error rate in check [s1]. The finding applies to patients with high-risk adenomas specifically, not to lower-risk polyps or to average-risk screening intervals [s1]. And missing data from people who skipped the 5-year colonoscopy had to be handled statistically rather than observed directly [s1].

An accompanying editorial framed the message as "when more is not better," a reminder that in surveillance, additional procedures are a cost as well as a safeguard, and that intervals should be set by evidence rather than caution alone [s2].

What to watch

The 10-year analysis is the one that will settle whether guidelines should lengthen the interval. In the meantime, EPoS II adds to a run of findings reshaping colorectal surveillance — from the lowering of the screening start age to 45 to trials of drugs that cut recurrence after polyp removal and AI systems that change how many polyps endoscopists find in the first place. The interval question is the one that touches the most patients, because nearly everyone who has had an adenoma removed is on a surveillance clock.

This article describes trial results and screening intervals for informational purposes only. It is not medical advice; decisions about surveillance timing rest with a clinician who knows the case.

Sources

• [s1] Jover R, et al. Colonoscopy Intervals and Colorectal Cancer Incidence after Adenoma Removal. New England Journal of Medicine, published online 2026-09-16.
• [s2] Shaukat A, et al. When More Is Not Better — Colonoscopy Surveillance after High-Risk Adenomas. New England Journal of Medicine, published online 2026-09-16.