Colorectal Cancer Screening in Familiar-Risk Population: Immunochemical Fecal Occult Blood Testing Versus Colonoscopy

Colorectal Cancer

FOBT, colonoscopy, familiar risk, colorectal cancer, advanced adenoma

This study is aimed: 1) to compare the accuracy of colonoscopy vs immunochemical faecal occult blood test (iFOBT) and colonoscopy when positive for colorectal cancer (CRC) screening in familiar-risk population and; 2) to determine the complications associated with both strategies.

This is an observational, controlled, randomized phase III study to evaluate the effectiveness of the iFOBT for detecting advanced neoplasia (polyps \> 1cm in size, high grade dysplasia or with villous component, or CRC) in first degree relatives of patients with CRC. Index cases will be interviewed to obtain the family tree and their first-degree relatives will be contacted to invite them to participate in the study. Index-cases, will be randomized into one of the following two groups in order that their relatives receive the same screening strategy: A) colonoscopy; or B) annual iFOBT test (OC-Sensor®, cut off ≥50 ng/ml) and colonoscopy if positive. To determine the sensitivity and specificity of the iFOBT strategy, individuals randomized to group B will be invited to undergo a complete colonoscopy following two years follow-up. In addition, epidemiological data, personal history of disease, family history of neoplasm, characteristics of lesions at colonoscopy and histological diagnosis will be recorded. To test the hypothesis of equivalence between the iFOBT test and colonoscopy for detecting advanced colorectal neoplasm, it was considered a probability of participation, detection capability and prevalence of advanced adenomas for iFOBT of 0.750, 0.565 and 0.077, respectively, being the product of them 0.033. In the case of colonoscopy, the likelihood of participation, detection capability and prevalence of advanced adenomas in this population at risk are 0,500, 0.965 and 0.077, respectively, and their product 0.037. Accordingly, for a Type I error (alpha) of 5%, a power of 80% and a maximum deviation between the probabilities of the two tests of 0.03 the number of subjects to be included per arm is 744

Spain