Introduction
The modality for colorectal cancer (CRC) screening has transferred from guaiac fecal occult blood testing (gFOBT) to fecal immunochemical testing (FIT), but its sensitivity remains lower. While total colonoscopy (TCS) has been anticipated as a new modality for CRC screening, conclusive results based on randomized controlled trials (RCTs) are still pending. Test performance RCTs have been utilized to predict the effectiveness of new modalities for CRC screening. They were adopted for the evaluation of cancer screening using a stepwise approach in the World Endoscopic Evaluation (Young et al. Cancer 2016).
Aims & Methods
A systematic review of test performance RCTs for CRC screening was conducted. Candidate articles were searched from inception to October 14, 2023, using Ovid MEDLINE, Cochrane Central Registry, Embase, and Ichushi-web. We compared the test performance of TCS screening, referring to the following modalities: gFOBT, FIT, flexible sigmoidoscopy (FS), and CT colonography (CTC). Single and multiple rounds were searched. Test performance indicators included participation rates in CRC screening, advanced neoplasia (AN) detection rates, and CRC detection rates. The primary results were evaluated using an intention-to-screen (ITS) based network meta-analysis, and a per-protocol analysis was also performed.
Results
Seventeen RCTs (13 from Europe, two from the US, and one each from Israel and China), involving a total of 341,123 healthy screening invitees, were included. The median participation rates for the programs based on FIT, FS, and TCS were 50.4 % (28.6-94.0%), 32.3% (7.9-52.1%), and 21.5% (10.0-42.5%), respectively. In the network meta-analysis, participation rates for TCS screening were significantly lower than the programs based on FS (RR=0.58; CI: 0.37-0.89), FIT (RR=0.45; CI: 0.31-0.65), and gFOBT (RR=0.45; CI: 0.30-0.69). Detection rates for AN were consistently higher for TCS than for FIT (RR= 2.18, 95%CI:1.46-3.26) but not statistically significant for FS (RR=1.06, 95CI: 0.65-1.73). Similarly, TCS had higher CRC detection rates compared to FIT (RR= 1.80, 95%CI:1.04-3.12) and FS (RR= 1.52, 95%CI:0.87-2.64). While AN and CRC detection rates for TCS were higher than those of CTC, the differences were not statistically significant (AN: RR= 1.02, 95%CI:0.65-1.73, CRC: RR= 0.73 95%CI: 0.32-1.64). The follow-up period of single-round RCTs was limited until diagnosis, but the screening frequency varied depending on the follow-up duration, which could affect effectiveness. Screening intervals were typically two years for FIT, five years for FS, and ten years for CS. One quasi-experimental study combined the following arms from different RCTs with extended follow-up: one-time CS, one-time FS, and four rounds of FIT (Grobbee et al. Clin Gastroenterol Hepatol 2020). The participation rate was higher for FIT (73%) compared to FS (31%) and CS (24%). Cumulative AN and CRC detection rates were higher for FIT (AN = 4.5%, 95%CI:4.2-4.9, CRC =0.5% 95%CI: 0.4–0.6), but nearly equivalent for CS (AN = 2.2%, 95%CI:1.8-2.6, CRC =0.2% 95%CI: 0.1–0.3) and FS (AN = 2.3%, 95%CI:2.0-2.7, CRC =0.2% 95%CI: 0.1–0.3).
Conclusion
The single-round test performance of colonoscopy screening might be superior to that of other modalities. To confirm the priority of TCS, it is insufficient to draw conclusions based on the results of a single-round performance, given variations in screening intervals and frequencies among other modalities. Cumulative effects among different modalities should be compared within the same follow-up period.
References
Young GP, Senore C, Mandel JS, et al. Recommendations for a step-wise comparative approach to the evaluation of new screening tests for colorectal cancer. Cancer. 2016;122(6):826-39.
Grobbee EJ, van der Vlugt M, van Vuuren AJ, et al. Diagnostic yield of one-time colonoscopy vs one-time flexible sigmoidoscopy vs multiple rounds of mailed fecal immunohistochemical tests in colorectal cancer screening. Clin Gastroenterol Hepatol. 2020;18(3):667-75.