First surveillance colonoscopy at five years was noninferior to three years after high-risk adenoma removal

Design

Randomised noninferiority trial, interim analysis at 5.5 years, eight European countries

Population

10,799 patients with high-risk adenomas after polypectomy

Primary outcome

Cumulative incidence of colorectal cancer (interim: at 5 years)

Effect

0.77% vs 0.82%; difference -0.05 percentage points; upper bound 0.68 (margin 0.7)

EPoS II is an ongoing noninferiority trial in eight European countries. It randomised 10,799 people with high-risk adenomas (at least one adenoma of 10 mm or more, high-grade dysplasia or villous growth, or three to ten adenomas) to a first surveillance colonoscopy at five years or at three years. The three-year group was also scoped at five years. Guidelines currently recommend three years.

This is an interim analysis after 5.5 years of follow-up. Cumulative colorectal cancer incidence at five years was 0.77% with the less frequent schedule and 0.82% with the more frequent one, a difference of -0.05 percentage points. The upper bound of the one-sided 99.12% confidence interval was 0.68 against a noninferiority margin of 0.7, so the criterion was met, narrowly. Stage at diagnosis did not appear to differ much. Five people died of colorectal cancer: 3 in the five-year group and 2 in the three-year group.

The prespecified primary end point is incidence at ten years, so the final answer is not yet in, and the upper bound sits close to the margin. Current guidelines have not changed, and one interim analysis is not a reason to depart from them. For services stretched by long colonoscopy waiting lists, this adds evidence that a five-year interval for this group may be reasonable, once guidelines catch up.

• Keep to your current guideline interval until it is revised; this is an interim analysis.
• Five-year cancer incidence was 0.77% vs 0.82% with five-year and three-year first surveillance.
• The upper confidence bound (0.68) was close to the 0.7 margin, so the margin of safety is small.
• Use this evidence when prioritising surveillance lists if capacity is limited, and follow local guidance.

Why it matters

It questions whether three-year surveillance is needed for everyone with high-risk adenomas, a recommendation with large workload implications.

Don't overread it

This is an interim five-year analysis of a ten-year trial; the guideline interval of three years has not changed.

The statistics, in plain English

Noninferiority means showing that the new schedule is not worse than the old by more than a set margin, here 0.7 percentage points. The upper bound of 0.68 sits just inside that margin, so the data support noninferiority but with little room to spare.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.