Once-only colonoscopy or two rounds of faecal immunochemical testing 2 years apart for colorectal cancer screening (SCREESCO): preliminary report of a randomised controlled trial

Anna Forsberg (Lead / Corresponding author), Marcus Westerberg, Chris Metcalfe, Robert Steele, Johannes Blom, Lars Engstrand, Kaisa Fritzell, Mikael Hellström, Lars-Åke Levin, Christian Löwbeer, Andreas Pischel, Ulf Strömberg, Sven Törnberg, Yvonne Wengström, Anders Ekbom, Lars Holmberg, Rolf Hultcrantz

Research output: Contribution to journalArticlepeer-review

33 Citations (Scopus)
158 Downloads (Pure)

Abstract

Background: Screening for colorectal cancer is done with lower gastrointestinal endoscopy or stool-based tests. There is little evidence from randomised trials to show primary colonoscopy reduces mortality in colorectal cancer. We aimed to investigate the effect of screening with once-only colonoscopy or two rounds of faecal immunochemical test screening on colorectal cancer mortality and incidence.

Methods: We did a randomised controlled trial in Sweden (SCREESCO). Residents in 18 of 21 regions who were age 60 years in the year of randomisation were identified from a population register maintained by the Swedish Tax Agency. A statistician with no further involvement in the trial used a randomised block method to assign individuals to once-only colonoscopy, two rounds of faecal immunochemical testing (OC-Sensor; 2 years apart), or a control group (no intervention; standard diagnostic pathways), in a ratio of 1:6 for colonoscopy versus control and 1:2 for faecal immunochemical testing versus control. Masking was not possible due to the nature of the trial. The primary endpoints of the trial are colorectal cancer mortality and colorectal cancer incidence. Here, we report preliminary participation rates, baseline findings, and adverse events from March, 2014, to December, 2020, in the two intervention groups after completion of recruitment and screening, up to the completion of the second faecal immunochemical testing round. Analyses were done in the intention-to-screen population, defined as all individuals who were randomly assigned to the respective study group. This study is registered with ClinicalTrials.gov, NCT02078804.

Findings: Between March 1, 2014, and Dec 31, 2020, 278 280 people were included in the study; 31 140 were assigned to the colonoscopy group, 60 300 to the faecal immunochemical test group, and 186 840 to the control group. 10 679 (35·1%) of 30 400 people who received an invitation for colonoscopy participated. 33 383 (55·5%) of 60 137 people who received a postal faecal immunochemical test participated. In the intention-to-screen analysis, colorectal cancer was detected in 49 (0·16%) of 31 140 people in the colonoscopy group versus 121 (0·20%) of 60 300 in the faecal immunochemical test group (relative risk [RR] 0·78, 95% CI 0·56–1·09). Advanced adenomas were detected in 637 (2·05%) people in the colonoscopy group and 968 (1·61%) in the faecal immunochemical test group (RR 1·27, 95% CI 1·15–1·41). Colonoscopy detected more right-sided advanced adenomas than faecal immunochemical testing. There were two perforations and 15 major bleeds in 16 555 colonoscopies. No intervention-related deaths occurred.

Interpretation: The diagnostic yield and the low number of adverse events indicate that the design from this trial, both for once-only colonoscopy and faecal immunochemical test screening, could be transferred to a population-based screening service if a benefit in disease-specific mortality is subsequently shown.

Original languageEnglish
Pages (from-to)513-521
Number of pages9
JournalLancet Gastroenterology and Hepatology
Volume7
Issue number6
Early online date14 Mar 2022
DOIs
Publication statusPublished - Jun 2022

ASJC Scopus subject areas

  • Gastroenterology
  • Hepatology

Fingerprint

Dive into the research topics of 'Once-only colonoscopy or two rounds of faecal immunochemical testing 2 years apart for colorectal cancer screening (SCREESCO): preliminary report of a randomised controlled trial'. Together they form a unique fingerprint.

Cite this