The positive predictive value for colorectal cancer of second-round fecal immunochemical testing was half that of first-round testing among average-risk patients, according to a report by Dr. Maaike J. Denters and colleagues in the March issue of Gastroenterology.
Moreover, there was no significant difference in positive predictive value (PPV) between participants who had performed a guaiac fecal occult blood test in the first round and those who had performed a fecal immunochemical test (FIT) in the first round (Gastroenterology 2012 March [doi:10.1053/j.gastro.2011.11.024]).
The researchers wrote that the differences between the two tests should not be overemphasized, and "no large difficulties are to be expected should a switch from a guaiac-based program to a FIT-based program be desired in screening programs currently using guaiac tests."
Dr. Denters, of the Academic Medical Centre, University of Amsterdam, and colleagues randomized 4,990 average-risk persons aged 50-74 years to a guaiac (n = 2,119) or FIT test (n = 2,871).
Tests were sent to participants and mailed back to the researchers. No dietary instructions were included with the guaiac test. Two years later, FIT kits were sent to all participants with negative results on the first round of testing.
Participants with a positive test result after either round of testing received an invitation for a consultation at the screening center, where a colonoscopy was recommended, barring contraindications.
Overall, 293 participants tested positive on the first round – 233 in the FIT cohort and 60 in the guaiac group. Thus, the positivity rate was 8.1% for the FIT test and 2.8% for the guaiac test.
A total of 239 of the positive patients underwent colonoscopy. In the guaiac group (n = 53), 24 had advanced adenomas as their most advanced finding. These were defined as any adenoma 10 mm or greater, or with a villous component greater than 20%, or with high-grade dysplasia. That meant there was a PPV of 45% for advanced adenomas. Eight patients had colorectal cancer (CRC), for a PPV for cancer of 15%.
Among the FIT-positive group (n = 186), there were 88 advanced adenomas (PPV = 47%) and 12 cancers (PPV = 6%).
Overall, there were 20 cancers, representing a PPV for cancer of 8% for all first-round positive patients.
In the second round of testing, among the FIT-after-guaiac patients who underwent colonoscopy following a positive second-round test result (n = 122), there were 53 advanced adenomas (PPV = 43%) and 5 cancers (PPV = 4%).
Similarly, the FIT-after-FIT cohort had 50 advanced adenomas detected (PPV = 38%) and 4 cancers (PPV = 3%).
Overall, this totaled nine cancers, yielding a PPV for cancer of just 4% in the second round. "In the second round, fewer cases of advanced neoplasia were detected after a positive test result, while the chances of finding CRC were halved," the authors wrote.
However, "despite a significant decrease in the PPV for CRC in a second round of screening, a substantial number of significant lesions are detected in a second screening round," independent of the type of test used in the first round, although this finding applies more to advanced adenomas than to cancer, they wrote.
The authors pointed to one potential limitation: They chose a low hemoglobin cut-off level for FIT positivity (50 ng Hb/mL) compared with other studies that have used a value of 75 ng Hg/mL or even 100 ng Hg/mL.
"It is very well possible that in further screening rounds [at this level,] positivity rates will stay relatively high, resulting in many colonoscopy procedures, whereas PPV will decrease further," they wrote. "The choice for a cut-off level will be a fine balance between these two parameters and will be influenced by economic, behavioral, and other parameters, and differ per country."
The study was funded by the Netherlands Organization for Health Research and Development (ZonMw). The authors had no conflicts to disclose.