ABSTRACT
BACKGROUND & AIMS: Age, sex, smoking, and family history are risk factors for colorectal cancer in Asia. The Asia-Pacific Colorectal Screening (APCS) scoring system was developed to identify subjects with a high risk for advanced neoplasm (AN). We tested an algorithm that combined APCS scores with fecal immunochemical test (FIT) in colorectal cancer screening. METHODS: We performed a multicenter prospective study, enrolling asymptomatic individuals older than 40 years old in 12 Asia-Pacific regions from December 2011 to December 2013. APCS scores were calculated for each individual (0-1 = low risk [LR], 2-3 = medium risk [MR], and 4-7 = high risk [HR] for AN). LR and MR subjects were offered FIT and referred for early colonoscopies if FIT results were positive. HR subjects were offered colonoscopies. The proportions of subjects with ANs were determined for each group based on colonoscopy findings; odd ratios for LR and MR subjects were calculated compared to LR individuals. We calculated the sensitivity of the APCS-FIT algorithm in identifying subjects with AN. RESULTS: A total of 5657 subjects were recruited: 646 subjects (11.4%) were considered LR, 3243 subjects (57.3%) were considered MR, and 1768 subjects (31.3%) were considered HR for AN. The proportions of individuals with an AN in these groups were 1.5%, 5.1%, and 10.9%, respectively. Compared with LR group, MR and HR subjects had a 3.4-fold increase and a 7.8-fold increase in risk for AN, respectively. A total of 70.6% subjects with AN (95% confidence interval: 65.6%-75.1%) and 95.1% subjects with invasive cancers (95% confidence interval: 82.2%-99.2%) were correctly instructed to undergo early colonoscopy examination. CONCLUSIONS: The APCS scoring system, which is based on age, sex, family history, and smoking, is a useful tool for determining risk for colorectal cancer and advanced adenoma in asymptomatic subjects. Use of the APCS score-based algorithm in triaging subjects for FIT or colonoscopy can substantially reduce colonoscopy workload.
Subject(s)
Biomarkers, Tumor/analysis , Colorectal Neoplasms/metabolism , Decision Support Techniques , Early Detection of Cancer/methods , Feces/chemistry , Immunohistochemistry , Adult , Age Factors , Aged , Algorithms , Asia/epidemiology , Biomarkers, Tumor/genetics , Colonoscopy , Colorectal Neoplasms/epidemiology , Colorectal Neoplasms/genetics , Colorectal Neoplasms/pathology , Female , Genetic Predisposition to Disease , Humans , Male , Middle Aged , Pedigree , Predictive Value of Tests , Prognosis , Prospective Studies , Risk Assessment , Risk Factors , Sex Factors , Smoking/adverse effects , Smoking/epidemiology , Time FactorsABSTRACT
OBJECTIVES: We tested the hypothesis that the risk of colorectal cancer (CRC), advanced colorectal neoplasia (ACN), and colorectal adenoma among screening participants with different first-degree relatives (FDRs) affected by CRC was similar. METHODS: A multi-center, prospective colonoscopy study involving 16 Asia-Pacific regions was performed from 2008 to 2015. Consecutive self-referred CRC screening participants aged 40-70 years were recruited, and each subject received one direct optical colonoscopy. The prevalence of CRC, ACN, and colorectal adenoma was compared among subjects with different FDRs affected using Pearson's χ2 tests. Binary logistic regression analyses were performed to evaluate the risk of these lesions, controlling for recognized risk factors including age, gender, smoking habits, alcohol drinking, body mass index, and the presence of diabetes mellitus. RESULTS: Among 11,797 asymptomatic subjects, the prevalence of CRC was 0.6% (none: 0.6%; siblings: 1.1%; mother: 0.5%; father: 1.2%; ≥2 members: 3.1%, P<0.001), that of ACN was 6.5% (none: 6.1%; siblings: 8.3%; mother: 7.7%; father: 8.7%; ≥2 members: 9.3%, P<0.001), and that of colorectal adenoma was 29.3% (none: 28.6%; siblings: 33.5%; mother: 31.8%; father: 31.1%; ≥2 members: 38.1%, P<0.001). In multivariate regression analyses, subjects with at least one FDR affected were significantly more likely to have CRC (adjusted odds ratio (AOR)=2.02-7.89), ACN (AOR=1.55-2.06), and colorectal adenoma (AOR=1.31-1.92) than those without a family history. The risk of CRC (AOR=0.90, 95% confidence interval (CI) 0.34-2.35, P=0.830), ACN (AOR=1.07, 95% CI 0.75-1.52, P=0.714), and colorectal adenoma (AOR=0.96, 95% CI 0.78-1.19, P=0.718) in subjects with either parent affected was similar to that of subjects with their siblings affected. CONCLUSIONS: The risk of colorectal neoplasia was similar among subjects with different FDRs affected. These findings do not support the need to discriminate proband identity in screening participants with affected FDRs when their risks of colorectal neoplasia were estimated.