Fecal immunochemical tests useful for colorectal-cancer screening

Last Updated: 2019-02-26

By Will Boggs MD

NEW YORK (Reuters Health) - One-time fecal immunochemical tests (FIT) can be useful for identifying colorectal cancer (CRC), but they lack sensitivity for advanced adenomatous polyps, according to a systematic review and meta-analysis.

"FIT annually is a good alternative to colonoscopy, but/and it requires ways to ensure annual testing of those who choose it and, most importantly, timely evaluation with colonoscopy for persons who are FIT positive," Dr. Thomas F. Imperiale from Indiana University School of Medicine, in Indianapolis, told Reuters Health by email.

While colonoscopy is the most frequently used CRC screening test in the U.S., several other countries use annual or biennial stool blood tests or a combination of stool testing and lower endoscopy. FIT is associated with higher rates of participation and acceptance, but studies evaluating its performance characteristics have shown inconsistent findings for CRC and advanced adenomas.

Dr. Imperiale and colleagues provide an updated summary of FIT performance for CRC and advanced adenomas (with colonoscopy as reference standard) based on 31 articles testing 18 FITs. The team also evaluated whether variation in performance characteristics is a function of the threshold used to define a positive test result.

Positivity thresholds ranged from 2 to 67 ug/g, with 11 studies using a positivity threshold below 10 ug/g, 17 studies using a threshold of 10 ug/g, nine using a threshold of 11 to 19 ug/g, and 26 using a threshold of 20 ug/g or greater.

The prevalence of advanced adenomas ranged from 1.26% to 12.2%, and the prevalence of CRC ranged from 0.15% to 3.48%, the researchers report in Annals of Internal Medicine, online February 26.

Sensitivity for CRC ranged from 0.91 for a threshold of 10 ug/g to 0.71 for a threshold of greater than 20 ug/g, and specificity ranged from 0.90 to 0.95, respectively, for those thresholds.

The positive likelihood ratio was 15.49 at a threshold greater than 20 ug/g, and the negative likelihood ratio was 0.10 at a threshold of 10 ug/g.

For advanced adenomas, FIT sensitivity ranged from 0.40 for a threshold of 10 ug/g to 0.25 for a threshold of 20 ug/g, with specificities ranging from 0.90 for a threshold of 10 ug/g to 0.95 for a threshold greater than 20 ug/g.

Positive likelihood ratios ranged from 3.39 at a threshold of less than 10 ug/g to 5.86 at a threshold greater than 20 ug/g, and negative likelihood ratios ranged from 0.67 to 0.79.

Overall FIT accuracy (as measured by area under the curve) was 0.94 for CRC at all thresholds and was 0.73, 0.62 and 0.69 for advanced adenomas at thresholds of 10 ug/g, greater than 10 to less than 20 ug/g, and greater than 20 ug/g, respectively.

Based on the performance of different FIT brands, the researchers suggest that OC Sensor (a quantitative FIT) might be preferred for large-scale use and OC Light (a qualitative FIT) might be preferred for small-scale use.

"Hopefully, use of FIT will reduce the proportion of unscreened persons from the current 35% to < 20%," Dr. Imperiale said. "The findings should facilitate what the USPSTF supports - use of any of the recommended screening tests without preference - and remind both patients and especially providers that there are good alternatives to colonoscopy - annual FIT being one of them - and that these options should be recommended equally well (for average-risk persons) by providers."

He added, "The data in our systematic review may help health care systems/countries that use FIT for population-based screening decide on what threshold to use when applied to their population."

Dr. James Allison of the University of California, San Francisco, who wrote a linked editorial, told Reuters Health by email, "We must increase our national screening for CRC numbers, especially in the vulnerable population - uninsured, underinsured, poor - and calling one screening test (colonoscopy) the 'best/gold standard' is not helpful or true."

"Although there were lots of studies examined, the FITs compared were actually few in number because so few FITs have had validation of their performance characteristics confirmed by use in large average risk populations," he said. "Over 120 FITs have been 'cleared' by the FDA (Food and Drug Administration) for use in the U.S. but not 'approved' by the FDA. This means many FITs can be bought at a low price, fulfilling a healthcare institution's HEDIS requirement but not guaranteeing that their patients have been screened by a well-studied FIT."

"The average-risk people screened by FIT must be sure that the FIT supplied by their MD or healthcare system has been carefully studied to confirm its advertised performance characteristics," he said. "Ask your MD!"

"The Imperiale article defined their findings more accurately than most reports by saying the one-time sensitivity for advanced adenomas was low," Dr. Allison said. "Sadly, most readers will not understand the statement. FITs are used every year or every other year in all population screening programs and so there are multiple times for it to uncover an advanced adenoma."

"Furthermore, few of all the polyps one finds in this average-risk population ever progress to fatal cancer," he said. "Polyps are given a 'bad name,' i.e., 'precancerous polyps,' and used as a scare tactic, in my opinion."

"We must change existing laws that charge copays for a colonoscopy done for a positive FIT," Dr. Allison added. "We need better and more-consistent payment policies that insure coverage of colonoscopy after an abnormal FIT test."

SOURCE: https://bit.ly/2NsQbc7 and https://bit.ly/2H2UMRo

Ann Intern Med 2019.