First surveillance colonoscopy at five years was noninferior to three years at interim analysis

Design

Randomised noninferiority trial (EPoS II), interim analysis at 5.5 years

Population

10,799 patients with high-risk adenomas, 5,398 in the 5-year group and 5,401 in the 3-year group

Primary outcome

Cumulative colorectal cancer incidence (5-year interim; 10-year primary)

Effect

0.77% vs 0.82%; difference -0.05 percentage points; upper boundary of 99.12% CI 0.68 (margin 0.7)

EPoS II is an ongoing noninferiority trial in eight European countries. It randomised 10,799 patients with high-risk adenomas (an adenoma of 10 mm or more, high-grade dysplasia or villous growth, or 3 to 10 adenomas of any kind) to a first surveillance colonoscopy at 5 years or at 3 years, which guidelines currently recommend. The 3-year group was also scoped at 5 years. The primary endpoint is 10-year cumulative colorectal cancer incidence, with a prespecified noninferiority margin of 0.7 percentage points.

This interim analysis at 5.5 years found a five-year cumulative colorectal cancer incidence of 0.77% with less frequent surveillance and 0.82% with more frequent surveillance, a difference of -0.05 percentage points. The upper boundary of the 99.12% confidence interval was 0.68, meeting the interim criterion. Stage at diagnosis did not appear to differ. Five patients died of colorectal cancer: three in the 5-year group and two in the 3-year group.

The result is interim, based on a five-year endpoint, with cancers and deaths too few to show small differences, and the final 10-year analysis is awaited. Guidelines have not changed. It is an argument for discussing intervals, not for changing them now, particularly in a country where colonoscopy capacity is limited and quality varies.

• Keep to your current guideline interval until the 10-year analysis and updated guidance
• Where capacity is short, discuss the trial as context when prioritising surveillance among high-risk patients
• Make sure the index colonoscopy was complete and the polyp removal was adequate, since surveillance depends on it
• Do not apply the result to patients with a family history, serrated polyposis or other higher-risk syndromes, who were not the focus

Why it matters

Surveillance colonoscopy absorbs a large share of endoscopy capacity, and a longer interval could free it if confirmed.

Don't overread it

An interim analysis at five years does not replace the planned 10-year endpoint, and guidelines have not changed.

The statistics, in plain English

The interim analysis used a stricter 99.12% confidence interval than the usual 95% because it looks at the data before the trial ends, which protects against false conclusions. An upper boundary of 0.68 is just inside the 0.7 margin, so the margin for error is small. Colorectal cancers were rare in both groups, so the comparison depends on very few events.

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.