Accuracy of a faecal immunochemical test in patients under colonoscopy surveillance of colorectal adenoma and cancer.

Abstract

Surveillance of colorectal neoplasia place great strain on colonoscopy resources, and faecal immunochemical tests (FIT) are under-investigated for this purpose. The aim of this study was to report the outcome of FIT among patients scheduled for post-polypectomy and post-resection colorectal cancer (CRC) surveillance. Patients scheduled for colonoscopy surveillance at five endoscopy units in mid-Sweden in 2016-2020 were eligible. They provided a faecal sample from 2 separate days, which were analysed by iFOBT QuikRead go® (Aidian Oy). Both the colonoscopies, and the FIT analyses were conducted by staff blinded to the other. Out of 216 included patients, 157 (73%) underwent both a complete colonoscopy and had at least one FIT analysed prior to the examination. The indication for surveillance was previous adenoma in 69 (44%) and post-resection CRC in 88 (56%) patients. Two (1%) in the CRC surveillance group were diagnosed with a metachronous CRC, whereas 49 (56%) patients in the CRC surveillance, and 17 (25%) in the adenoma group had no pathology identified at colonscopy (P < 0.001). The proportion of patients diagnosed with adenomas requiring surveillance according to European Society of Gastrointestinal Society (ESGE) guidelines 2020 was 6 (7%) in the post-CRC resection versus 7 (10%) in the adenoma surveillance group (P = 0.4). Based on one FIT and at cut-off 10 µg Hb/g, sensitivity for CRC was 100%, specificity 83% (95% confidence interval [CI]: 77-89), Positive Predictive Value (PPV) 7% (-2 to 16) and Negative Predictive Value (NPV) 100%. All patients with an adenoma requiring surveillance had a FIT below this cut-off. Adding a second FIT decreased the specificity. Larger studies to evaluate the accuracy and consequences of using FIT for surveillance of colorectal neoplasia are needed. FIT may be more interesting for post-resection CRC surveillance than follow-up of adenoma.