Fecal immunochemical testing, or FIT, may miss more colorectal cancers and advanced precancerous lesions in women than in men when using the standard positivity threshold, according to a retrospective analysis presented at the American College of Gastroenterology annual meeting. The finding comes from a study of over 30,000 screening participants conducted by Thomas F. Imperiale, MD, of Indiana University School of Medicine.
Gender Differences in FIT Detection
Men showed higher fecal hemoglobin levels and greater FIT sensitivities for both colorectal cancers (CRCs) and advanced precancerous lesions (APLs) compared with women, the study found. At a threshold of 100 ng/mL, the standard cutoff for FIT tests, CRC sensitivity was 77.4% for men versus 62% for women (P=0.03). APL sensitivity followed a similar pattern: 25.8% for men and 20.2% for women (P&less;0.001).
Specificity was slightly lower for men at 94.2% compared with 95.3% for women (P&less;0.001). To achieve equivalent CRC sensitivity for women as men, the FIT threshold would need to drop to 12 ng/mL, which would reduce specificity to 75.7%. A threshold of 65 ng/mL would match APL sensitivity but lower specificity to 92.8%.
Threshold Adjustments and Implications
Adjusting thresholds to reach 95% specificity required different cutoffs by sex: 120 ng/mL for men and 94 ng/mL for women. Imperiale noted that lowering the threshold for women to improve cancer detection raises questions about downstream effects, including more false positives and follow-up procedures. The optimal threshold would depend on the specific test used, as performance varies between brands.
Read Also: Embryo Transfer More Effective Than IUI for IBD Fertility
Current population-based screening programs prioritize maximizing sensitivity, Imperiale said, but other factors also matter. Understanding test performance and long-term outcomes is essential for balancing benefits and risks. FIT has an average sensitivity of 79% and specificity of 94%, with characteristics varying by threshold and test type. Differences in sex-based lesion distribution may warrant adjusted thresholds.
The analysis used data from the DeeP-C and BLUE-C studies, validation sets for multi-target stool DNA testing. Among 30,165 participants (mean age 63.4 years; 53.4% women), 0.54% had CRC and 9.6% had APLs. No significant sex differences emerged in fecal hemoglobin levels for CRCs or APLs among those with nonadvanced neoplasia or normal colonoscopies.
While lesion size did not differ between sexes, location did. Women had a higher proportion of proximal CRCs (48.1% vs. 31% for men), with markedly lower CRC sensitivity at 100 ng/mL (50% vs. 80.8%, P=0.01). Women also had more proximal APLs (61.2% vs. 53.6%), with reduced sensitivity (12.8% vs. 17.7%, P=0.01).
Clinical Considerations and Limitations
Multivariable analysis identified male sex, current or former smoking, BMI over 30, larger lesion size, and distal/rectal location as independent predictors of FIT sensitivity for advanced neoplasia. Despite these findings, Imperiale cautioned that the study had limitations, including use of a single test brand, a single threshold, and one application. The benefits of adjusting thresholds may increase with broader programmatic use.
