medwireNews: Fecal immunohistochemical tests (FITs) should not be considered interchangeable, warn researchers who have found significant differences in the performance of five commonly used in the USA to detect advanced colorectal neoplasia (ACN).
“Our findings have implications for the benefits and cost-effectiveness of CRC [colorectal cancer] screening using FIT,” say Barcey Levy (University of Iowa in Illinois, USA) and colleagues.
“Tests with lower sensitivity will miss more patients with CRC and advanced polyps, and tests with higher sensitivity and lower PPV [positive predictive value] will require more colonoscopies to detect patients with actionable findings.”
The investigators compared the accuracy of five FDA-approved FITs; one quantitative test – OC-Auto-FIT (Polymedco Inc, New York, USA) –- and four qualitative assessments – Hemocult ICT (Beckman Coulter, California, USA), Hemosure iFOB (Hemosure Inc, California, USA), OC-Light S FIT (Polymedco Cancer Diagnostic, New York, UDA) , and QuickVue iFOB (Quidel, California, USA) – in 3761 individuals scheduled for screening or surveillance colonoscopy, which was used as the reference standard.
The participants were 50–85 years old (mean 62.1 years) and 63.2% were women. The majority were White, at 86.4%.
In all, 8.5% of individuals tested positive for ACN (advanced polyps or CRC), including nine (0.2%) who tested positive for CRC.
Test positivity varied fourfold across the FITs, ranging from 3.9% with Hemoccult ICT, which has a hemoglobin (Hb) cutoff of 300 µg/g, to 16.4% with OC-Light S FIT, which has a Hb cutoff of 10 µg/g.
The rate of unevaluable tests was highest with QuickVue iFOB, at 2.5%, and Hemoccult ICT, at 2.3%. This finding suggests that these FITs “are more difficult for some patients to complete correctly at home and/or need clearer instructions,” the researchers write in the Annals of Internal Medicine.
There was a significant difference in the specificity of all the tests, ranging from 85.5% with OC-Light S FIT to 96.6% with Hemoccult ICT.
Similarly, there was a significant difference in test sensitivity for Hemoccult ICT (10.1%), OC-AUTO FIT (19.1%), and OC-Light S FIT (36.7%), but comparable sensitivity for the Hemosure iFOB and QuickVue iFOB tests, at 27.4% and 30.0%, respectively, both of which have a Hb cutoff of 50 µg/g.
“Brand of FIT was the most important factor affecting sensitivity,” Levy et al note. But it was also higher for “participants with higher BMI and those with lower income,” the authors report.
The FITs were also less sensitive in women than men, “perhaps due to higher prevalence of proximal lesions,” they add.
The PPV of the tests varied from 18.2% to 28.9% and the negative predictive value varied from 92.2% to 93.3%.
The differences in the FIT tests “have practical importance for FIT-based screening programs,” say Levy et al, “as these differences affect the need for repeated FIT, the yield of ACN detection, and the number of diagnostic colonoscopies that would be required to follow-up on abnormal findings.”
They suggest that healthcare systems with the capacity for an automated FIT might consider the OC-Auto FIT, which has a Hb cutoff of 20 µg/g, because the positivity rate was mid-range (5.6%), the PPV was “acceptable,” at 28.9%, and “very few tests were unevaluable,” at 0.2%.
In settings where simple point-of-care tests are required, they propose OC-Light S FIT or Hemosure iFOB because of their low Hb threshold and high positivity rates.
“Such tests may be most useful where low adherence to testing is anticipated and where access to colonoscopy is good,” the researchers conclude.
medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2024 Springer Healthcare Ltd, part of the Springer Nature Group