Five-year surveillance after high-risk polyps held up against three

Design

Multinational randomised noninferiority trial (interim analysis)

Population

10,799 patients with high-risk adenomas after polyp removal

Primary outcome

Cumulative colorectal cancer incidence (10-year primary; interim at 5.5 years)

Effect

5-year incidence 0.77% vs 0.82%; difference -0.05 points, upper 99.12% CI 0.68 (margin 0.7)

EPoS II, a noninferiority trial in eight European countries, randomised 10,799 patients with high-risk adenomas (a large, high-grade or villous adenoma, or 3-10 adenomas) to a first surveillance colonoscopy at five years or the currently recommended three years, with colorectal cancer incidence as the endpoint.

In this interim analysis at 5.5 years, the five-year cumulative incidence of colorectal cancer was 0.77% with five-year surveillance and 0.82% with three-year surveillance (difference -0.05 percentage points), meeting the prespecified noninferiority margin, with no clear difference in cancer stage at diagnosis.

The direction is reassuring: for many patients with high-risk adenomas, a longer first interval may be safe, easing colonoscopy demand. But this is an interim read of a trial whose primary endpoint is cancer at ten years, and current guidelines still recommend three years, so it informs the conversation rather than overturning practice today.

• Five-year first surveillance was noninferior to three years for colorectal cancer incidence at interim.
• Cancer incidence was 0.77% (5-year) versus 0.82% (3-year), difference -0.05 points.
• Cancer stage at diagnosis did not clearly differ between groups.
• This is interim; the primary endpoint is colorectal cancer at ten years.
• Guidelines still recommend three-year surveillance for now.

Why it matters

It challenges the assumed need for three-year surveillance and could ease colonoscopy capacity if confirmed.

Don't overread it

Interim analysis of a trial whose primary endpoint is cancer at ten years; it does not yet change guideline-recommended intervals.

The statistics, in plain English

Noninferiority means five-year surveillance was not meaningfully worse within a pre-set margin; because this is an interim analysis of a ten-year endpoint, it is supportive, not definitive.

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.