Improvement of Asia-Pacific colorectal screening score combined with fecal immunochemical testing at adjusted thresholds in colorectal cancer screening
The Asia-Pacific colorectal screening (APCS) score was developed in identifying the potential high-risk population of colorectal cancer (CRC). However, the clinical utility of such a risk score has not been fully elucidated. We aimed to evaluate the diagnostic accuracy and cost to detect one advanced neoplasia (AN) of the APCS score combined with fecal immunochemical testing (FIT) at various scenarios versus FIT alone in detecting AN.
Based on an ongoing randomized controlled trial comparing colorectal cancer screening strategies (June 2018 - May 2019), 3407 participants aged 50-74 years who underwent colonoscopies were included in this study. All the participants had available colonoscopy and/or pathology reports. In addition, we prospectively collected fecal samples before the colonoscopy examination. The fecal samples were used for FIT (OC-Sensor, Eiken Chemical, Japan) following a standardized operation process. We collected detailed epidemiological questionnaire data which was further used for the calculation of the ACPS score (based on age, sex, BMI, family history of CRC in first-degree relatives and smoking, yielding risk scores of 0 to 6). For the strategy of the APCS score combined with FIT, the participants were recommended to undergo colonoscopy if they were assessed as high risk of CRC or had FIT-positive results. Diagnostic accuracy for AN was estimated based on FIT or the strategy of the APCS score combined with FIT at multiple adjusted thresholds, respectively. The cost to detect one AN per 100,000 invitees was evaluated. Indicators such as participation and compliance were modeled by summarizing up-to-date evidence from published studies.
Among the 3407 included participants, 1753 (51.5%) were men and the mean age (SD) was 60.5 (6.3) years. The participants included 28 (0.8%) CRC, 255 (7.5%) advanced adenomas, 677 (19.9%) nonadvanced adenomas, and 2447 (71.9%) benign or negative findings. The sensitivity of the combination of the APCS score and FIT for AN ranged from 27.6% (95%CI, 22.4%-33.2%) to 76.3% (95%CI, 70.9%-81.2%) at the positive threshold of 3 or higher for APCS score and of 10 μg Hb/g or higher for FIT, which was higher overall than FIT alone (ranging from 13.8%, 95% CI, 10.2%-18.3% to 17.3%, 95% CI, 13.1%-22.2%). Compared with FIT at the conventional threshold of 20 μg Hb/g alone, the strategy of APCS at the positive threshold of 4 combined with FIT at 20 μg Hb/g improved the detection rate of AN per 100,000 invitees from 1231 to 2562, with numbers of colonoscopies needed to be scoped per 100,000 invitees increasing from 2736 to 17274. Further reducing the positive threshold of APCS from 4 to 3 would detect 24.6% more AN and require 60% more colonoscopies. Costs per AN detected by the strategy of APCS at the positive threshold of 4 combined with FIT at 20 μg Hb/g and FIT (20 μg Hb/g) alone were CNY¥ 4095 to CNY¥ 3550, respectively.
The combination of APCS score and FIT showed high sensitivity to detect AN. Tailored thresholds of the APCS combined with FIT may provide better screening yield at comparable costs compared with FIT-based screening.
Legal entity responsible for the study
The study was supported by CAMS Innovation Fund for Medical Science (2017-I2M-1-006; 2019-I2M-2-002) and National Natural Science Foundation (81703309).
The presenting author has declared no conflicts of interest.