How accurate are the new fecal DNA and fecal immunochemical tests as screening tests for colorectal cancer?
Fecal DNA testing is more sensitive but less specific than fecal immunochemical testing (FIT), and as a result, has a higher false-positive rate. It is also more expensive than other noninvasive alternatives such as FIT. We do not know which test will be better at reducing mortality. (Level of Evidence = 2b)
Fecal DNA testing looks for abnormalities characteristic of the DNA in colorectal cancer, whereas FIT is an improved version of the older tests that detect fecal occult blood but requires only a single stool specimen. Adults 50 to 85 years of age at average risk of colorectal cancer who were undergoing screening colonoscopy were invited to participate in the study. A total of 11,016 agreed and underwent the required tests. Of that group, 689 were excluded because of an insufficient specimen for fecal DNA or a specimen that leaked in shipping; 304 others were excluded because of inadequate colonoscopy; and 34 had an insufficient sample for FIT. Of the final group of 9,989 participants, 65 received a diagnosis of cancer.
Fecal DNA testing was 92% sensitive and 87% specific, whereas FIT was 74% sensitive and 95% specific. Although this looks like a clear advantage for fecal DNA (60 of 65 cancers detected compared with 48 of 65 for FIT), it is important to look a bit further. The lower specificity for fecal DNA meant that there were nearly three times as many false-positive results that would have required a follow-up colonoscopy if fecal DNA was the sole screening test (1,231 vs. 472 for FIT). Using fecal DNA, there would have been 22 colonoscopies per cancer detected compared with 11 using FIT. Also, fecal DNA testing requires the entire stool specimen, collected using a small bucket that hangs in the toilet, and costs approximately $400 to $800 (FIT costs approximately $3 to $40). Because colorectal cancer takes several years to progress from adenoma to cancer, FIT could be performed annually at a much lower cost and would likely detect many of the initially missed cancers in subsequent years.
Study design: Diagnostic test evaluation
Funding source: Industry
Setting: Outpatient (any)
Reference: ImperialeTFRansohoffDFItzkowitzSHet alMultitarget stool DNA testing for colorectal-cancer screening. N Engl J Med.2014; 370( 14): 1287– 1297.