Fecal Immunochemical Testing for Colorectal Cancer Screening
FIT detected most cancers, but only a minority of advanced adenomas.
Fecal immunochemical testing (FIT) might be more accurate than guaiac-based fecal occult blood testing (gFOBT) in screening for colorectal cancer. In this Dutch study, 1256 average-risk patients submitted single specimens for FIT (OC-Sensor) just before undergoing screening colonoscopy.
Colonoscopy identified 8 patients (0.6%) with colorectal cancer and 113 (9%) with advanced adenomas. At a cutoff of 50 ng/mL, FIT was positive in 10% of patients. Sensitivity and specificity of FIT for detecting advanced adenomas were 38% and 93%, respectively. For colorectal carcinoma, sensitivity was 88% (i.e., FIT was positive in 7 of 8 patients with cancer), and specificity was 91%. Five of the seven FIT-positive cancers were localized (Dukes stage A). FIT detected proximal and distal advanced neoplasia with equal sensitivity.
The Journal Watch General Medicine Perspective
According to these findings, if patients were screened initially with a single FIT, most localized cancers and about one third of advanced adenomas would be detected, and 90% of patients (those who were FIT-negative) would avoid colonoscopy. Failure to detect most advanced adenomas is not necessarily a fatal flaw, if additional research shows that repeated FIT screening (i.e., at 1- or 2-year intervals) detects many of these lesions before they progress to unresectable cancers. One of our Journal Watch Gastroenterology editors, an expert in colorectal cancer screening, comments below on FIT.
— Allan S. Brett, MD
The Journal Watch Gastroenterology Perspective
Current colorectal cancer screening guidelines recommend that clinicians who use fecal blood testing switch from gFOBT to FIT (Am J Gastroenterol 2009; 104:739). Consistent results from several types of studies, including randomized, controlled trials, indicate that patient adherence (i.e., completion of the test) and test sensitivity strongly favor FIT over gFOBT. Several national screening programs outside the U.S. now are based on FIT, and cost-effectiveness analyses suggest that annual FIT is at least as cost-effective as is colonoscopy every 10 years. Several randomized, controlled trials, including one in U.S. Veterans Administration hospitals) have been organized to compare FIT and colonoscopy.
A practical problem that clinicians encounter when they try to switch to FIT is the lack of comparative performance data on the several commercial FIT assays available in the U.S. Several years ago, in a study of six commercial FITs available in Germany, researchers found that several had awful performance characteristics, including very poor specificity. However, the laboratory-based assay used in the current study (OC-Sensor) has been evaluated in many studies and is believed to perform best.
This Dutch study suggests that FIT is equally effective in both the proximal and distal colon, whereas some previous evidence had suggested better performance in the distal colon. This result is encouraging, but the endpoint for the study was advanced conventional (i.e., tubular, tubulovillous, or villous) adenomas. The study ignores (as do all FIT studies) the 30% of colorectal cancers that arise through a genetic pathway characterized by hypermethylation; the precursors of these cancers are not conventional adenomas but rather serrated lesions (sessile serrated polyps, also known as sessile serrated adenomas). Further, these serrated lesions are located primarily in the proximal colon. Endoscopically, these premalignant serrated lesions have no vessels on their surface, and some evidence shows that they don't bleed at all. The future of sensitive fecal testing that can identify both conventional and serrated precancerous lesions in the proximal colon is more likely to lie in fecal DNA testing than in FIT.
— Douglas K. Rex, MD
de Wijkerslooth TR et al. Immunochemical fecal occult blood testing is equally sensitive for proximal and distal advanced neoplasia. Am J Gastroenterol 2012 Oct; 107:1570. (http://dx.doi.org/10.1038/ajg.2012.249)
- Medline abstract (Free)
Published in Journal Watch General Medicine November 20, 2012
To ensure that your Reader Remark is not formatted as one long paragraph, precede new paragraphs with either a blank line or an indentation.