Relatively low risk and nonaggressive stage of colorectal cancer in individuals with negative baseline fecal immunochemical test results: A cohort study

摘要

Fecal immunochemical testing (FIT) is used as a screening tool because of its adequate sensitivity to detect colorectal cancer (CRC). Whether individuals with negative baseline FIT results have a low risk of CRC over a long period of follow-up remains unclear. This retrospective cohort study investigated the characteristics of CRC in individuals with normal baseline FIT results. We retrospectively evaluated individuals who participated in the Taiwanese biennial FIT-based CRC screening program at a single hospital between March 2010 and December 2017. Patients who had positive FIT results (>= 30 mu g Hb/g feces) and then underwent confirmatory colonoscopy were enrolled for further analysis. Patients were divided into two cohorts: target group (consisting of patients with negative FIT results in the baseline round, followed by positive FIT results and confirmatory colonoscopy in any one of the subsequent rounds) and reference group (consisting of patients with positive FIT results in the baseline round and confirmatory colonoscopy). The detection rate and characteristics of CRC were compared between two groups. A P value of <.05 was considered statistically significant. A total of 51 789 individuals underwent 80 195 tests in four rounds of biennial FIT based CRC screening. The positivity rate of FIT was 8.2% (n = 6591). The colonoscopy compliance rate was 70.3% (n = 4635). From 4635 individuals, 1650 and 2985 were included in the target and reference groups, respectively. The CRC detection rate in the target and reference groups was 2.3% (38/1650) and 5.0% (149/2985), respectively (odds ratio = 0.45; 95% confidence interval = 0.31-0.64). The CRC detection rate in the target group increased with screening intervals (P for trend = .001). After excluding patients with unknown stage, 35 patients in the target group and 130 patients in the reference group were included for subsequent analysis. Lesions were located mainly in the distal colon in both groups (82.9% [29/35] vs. 82.3% [107/130]; P = 1). The distribution of the CRC stage according to the seventh edition of the American Joint Committee on Cancer staging system in the target vs reference group was as follows: Stage 0: 48.6% vs 20.0%; Stage I: 11.4% vs 30.0%; Stage II: 20.0% vs 18.5%; Stage III: 17.1% vs 22.3%; and Stage IV: 2.9% vs 9.2%; (P = .008). CRC detection rate declined considerably if the baseline FIT result was negative. CRC detected in patients with negative baseline FIT results showed a high proportion of nonaggressive stage. Approximately 80% of CRCs were identified in the left-sided colon, suggesting the limited ability of repetitive FIT to detect proximal CRC. These results support the benefits and necessity of consecutive FIT screening even when the baseline FIT result is negative.