BACKGROUND: Colonoscopy and the faecal immunochemical test are accepted strategies for colorectal cancer screening in the average-risk population (ie, people aged =50 years without personal or family history of colorectal cancer). In this trial, we aimed to compare whether invitation to screening with faecal immunochemical test was non-inferior to colonoscopy in a screening programme.
METHODS: COLONPREV was a pragmatic, randomised, controlled, non-inferiority trial done at 15 tertiary hospitals across eight regions of Spain. Eligible participants were presumptively healthy and aged between 50 years and 69 years without a personal history of colorectal cancer, adenoma or inflammatory bowel disease, family history of hereditary or familial colorectal cancer (ie, two or more first-degree relatives with colorectal cancer or one diagnosed before age 60 years), severe comorbidities, or previous colectomy. Participants were randomly assigned (1:1) to one-time colonoscopy or biennial faecal immunochemical test before invitation to screening. The primary endpoint was colorectal cancer mortality at 10 years, assessed in the intention-to-screen population. An absolute difference of less than 0·16 percentage points was required to show non-inferiority. This trial was registered with ClinicalTrials.gov, NCT00906997.
FINDINGS: Between June 1, 2009, and Dec 31, 2021, 57 404 individuals were randomly assigned to receive an invitation for colonoscopy (n=28 708) or the faecal immunochemical test (n=28 696). The intention-to-screen population consisted of 26 332 individuals in the colonoscopy group and 26 719 in the faecal immunochemical test group. In the intention-to-screen population, participation in any form of screening was 31·8% in the colonoscopy group and 39·9% in the faecal immunochemical test group (risk ratio [RR] 0·79 [95% CI 0·77 to 0·82]). Faecal immunochemical testing was non-inferior to colonoscopy with regard to the risk of colorectal cancer mortality at 10 years: the risk was 0·22% (55 deaths) in the colonoscopy group and 0·24% (60 deaths) in the faecal immunochemical test group (risk difference -0·02 [95% CI -0·10 to 0·06; RR 0·92 [95% CI 0·64 to 1·32]; pnon-inferiority=0·0005).
INTERPRETATION: Participation in screening was higher among individuals invited to faecal immunochemical test screening than colonoscopy screening. On the basis of participation observed in this study, a faecal immunochemical test-based programme was non-inferior to a colonoscopy-based programme for colorectal cancer-related mortality.
FUNDING: Fundación Científica de la Asociación Española contra el Cáncer and Instituto de Salud Carlos III.
| Discipline Area | Score |
|---|---|
| Public Health |  |
| Family Medicine (FM)/General Practice (GP) | |
| General Internal Medicine-Primary Care(US) | |
| Gastroenterology |  |
This shows the benefit of FIT and will be useful in future practice.
Useful information. An invitation to FIT was non-inferior to an invitation to colonoscopy. The real question, however, is whether offering more than one modality and allowing patients to choose would improve overall participation and outcomes. Participation in either test (FIT or colonoscopy) was still very low in this study cohort.
Bottom line: it's harder to detect outcome differences in an RCT where most people fail to participate in colon cancer screening, even when just FIT testing as opposed to colonoscopy.
I think a lot of primary care providers see FIT testing as "inferior" to c-scope, so it is educational to read this analysis showing that, likely related to more uptake, recommending FIT is NOT an inferior strategy.
28,708 were randomised to colonoscopy, of whom 5293 were screened by colonoscopy and 3074 by FIT. 28,646 were randomised to FIT, of whom 10,525 were screened by FIT and 126 by colonocopy. How generalisable are these results? Could the results be due to the way patients were recruited and advised?