Positivity rates and performances of immunochemical faecal occult blood tests at different cut-off levels within a colorectal cancer screening programme

Abstract

BackgroundImmunochemical faecal occult blood tests have greater sensitivity for colorectal cancer screening than guaiac-based tests; however the number of positive tests required is still under discussion. MethodsA direct comparison of Hemoccult II with two immunochemical quantitative tests (OC-Sensor and FOB-Gold) using a 2-sample strategy was performed in over 30,000 patients undergoing colorectal cancer screening in France. ResultsPositivity ratio between immunochemical tests and Hemoccult II varied between 2.2 (OC-Sensor) and 2.4 (FOB-Gold) for the lowest cut-off value and 1.5–1.4 for the highest cut-off value. The positive predictive value for colorectal cancer was similar for immunochemical tests and Hemoccult II, and significantly higher for immunochemical tests for advanced adenomas. The detection rate of both colorectal cancer and advanced adenomas was higher with immunochemical tests than with Hemoccult II. With the 2-sample strategy and the lowest cut-off value the detection rate of colorectal cancer almost doubled and for advanced adenomas quadrupled. ConclusionFor colorectal cancer screening with immunochemical faecal occult blood tests, an acceptable strategy would be 2-day sampling with at least one positive test at a cut-off between 150 and 200ng/mL (OC-Sensor) and 176 and 234ng/mL (FOB-Gold). Data on the ease of test interpretation and cost-effectiveness now necessary to make definitive choices.