Interval cancer after two rounds of a Swedish population-based screening program using gender-specific cut-off levels in fecal immunochemical test

Abstract

Objective To evaluate interval cancer (IC) after two screening rounds of the Swedish population-based screening program of Stockholm–Gotland applying gender-specific cut-off levels in the fecal immunochemical test (FIT). Methods All 60- to 69-year-olds invited to screening 2015–2019 were included. The cut-off level for a positive test was 40 µg/g in women and 80 µg/g in men. Screening-detected colorectal cancers (SD CRCs) and ICs were verified in the Swedish Colorectal Cancer Register, and the follow-up time was two years from invitation. The test sensitivity, the IC rate (ICs per 10,000 screening negatives) and the IC incidence (ICs per 100,000 person-years) relative to the background CRC incidence were assessed by gender and age. The FIT levels were compared in men and women for CRCs diagnosed within one year of the sample. Results In the second screening round, 229,187 were invited, and SD CRCs and ICs were diagnosed in 193 and 144, respectively. The IC rate was 8.9 (7.4–10.3) and test sensitivity 0.61 (0.55–0.66), and was similar in men and women. For two screening rounds, the IC rate was significantly higher in men than in women, but the IC incidence/ background CRC incidence was similar in both genders. The FIT levels in female participants with CRC were significantly lower overall, and in early-staged CRCs as compared to men, and proximal localization was more common in women. In multivariable analysis, FIT levels were significantly lower in proximal CRCs. Conclusion Over two rounds, the IC incidence relative to the background CRC incidence was similar in men and women supporting a gender-specific screening strategy. The results could be explained by lower FIT levels in women due to proximal CRC localization.