Equivalency of fecal immunochemical tests and colonoscopy in familial colorectal cancer screening

Gastroenterology. 2014 Nov;147(5):1021-30.e1; quiz e16-7. doi: 10.1053/j.gastro.2014.08.004. Epub 2014 Aug 13.


Background & aims: Colonoscopy is the recommended screening procedure for first-degree relatives of patients with colorectal cancer (CRC), but few studies have compared its efficacy for CRC detection with that of other screening strategies. We conducted a controlled randomized trial to compare the efficacy of repeated fecal immunochemical tests (FITs) and colonoscopy in detecting advanced neoplasia (advanced adenoma or CRC) in family members of patients with CRC.

Methods: In a prospective study, 1918 first-degree relatives of patients with CRC were randomly assigned (1:1 ratio) to receive a single colonoscopy examination or 3 FITs (1/year for 3 years; OC-Sensor; cutoff ≥10 μg hemoglobin/g feces, corresponding to 50 ng hemoglobin/mL buffer). The strategies were considered to be equivalent if the 95% confidence interval of the difference for the detection of advanced neoplasia was ±3%. Follow-up analyses were performed to identify false-negative FIT results and interval CRCs.

Results: Of all eligible asymptomatic first-degree relatives, 782 were included in the colonoscopy group and 784 in the FIT group. In the intention-to-screen analysis, advanced neoplasia was detected in 33 (4.2%) and 44 (5.6%) first-degree relatives in the FIT and colonoscopy groups, respectively (odds ratio = 1.41; 95% confidence interval: 0.88-2.26; P = .14). In the per-protocol analysis, 28 first-degree relatives (3.9%) in the FIT group and 43 (5.8%) in the colonoscopy group had advanced neoplasia (odds ratio = 1.56; 95% confidence interval: 0.95-2.56; P = .08). FIT missed 16 of 41 advanced adenomas but no CRCs. The FIT strategy required endoscopic evaluation of 4-fold fewer individuals to detect 1 advanced neoplasia than the colonoscopy strategy.

Conclusions: Repeated FIT screening (1/year for 3 years) detected all CRCs and proved equivalent to colonoscopy in detecting advanced neoplasia in first-degree relatives of patients with CRC. This strategy should be considered for populations where compliance with FITs is higher than with colonoscopy. ClinicalTrials.gov number: NCT01075633 (COLONFAM Study).

Keywords: Colon Cancer; Colonoscopy Resources; Diagnostic Yield; Familial Colorectal Cancer.

Publication types

  • Comparative Study
  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't
  • Video-Audio Media
  • Webcast

MeSH terms

  • Adenoma / chemistry
  • Adenoma / diagnosis*
  • Adenoma / genetics
  • Adenoma / pathology
  • Adult
  • Biomarkers, Tumor / analysis*
  • Colon / pathology*
  • Colonoscopy*
  • Colorectal Neoplasms / chemistry
  • Colorectal Neoplasms / diagnosis*
  • Colorectal Neoplasms / genetics
  • Colorectal Neoplasms / pathology
  • Early Detection of Cancer / methods*
  • False Negative Reactions
  • Feces / chemistry*
  • Female
  • Genetic Predisposition to Disease
  • Hemoglobins / analysis*
  • Heredity
  • Humans
  • Immunochemistry*
  • Logistic Models
  • Male
  • Middle Aged
  • Odds Ratio
  • Patient Compliance
  • Pedigree
  • Predictive Value of Tests
  • Prospective Studies
  • Risk Factors
  • Spain
  • Time Factors


  • Biomarkers, Tumor
  • Hemoglobins

Associated data

  • ClinicalTrials.gov/NCT01075633