Colorectal cancer screening – effect on mortality and incidence rate of colorectal cancer. Overview of documentation and international recommendations
Systematic review
|Updated
Key message
Background
Colorectal cancer (CRC) is the second most common cancer in Norway and the incidence is increasing for both men and women. The majority of CRCs develops from benign adenomatous polyps through the adenoma-carcinoma stage. CRC may have no detectable symptoms while it is developing and spreading to nearby lymph
nodes. The survival is strongly connected with the tumour stadium at the time of diagnosis since early detection and treatment can prevent CRC. This type of cancer is potentially well suited for screening and several screening methods are available.
The Norwegian Directorate of Health requested a review (hasteoppdrag) of CRC
screening in September 2008. The goal was to summarize the evidence on the effect
of population-based CRC screening on mortality and incidence of CRC and to evaluate the accuracy of each screening method. An overview of screening methods used internationally was also requested and is also provided in the report.
Method
The following electronic databases were searched for relevant systematic reviews,
meta-analyses, HTA reports, and other relevant evidence-based literature on CRC
screening: Cochrane Database of Systematic Reviews (CDSR Issue 3 2008), Database of Abstracts of Reviews of Effects (DARE Cochrane Library Issue 4, 2008), HTA (Cochrane Library Issue 4, 2008), and Medline Ovid (1950 to September Week3 2008). The reference lists of retrieved studies were scanned for relevant articles or international guidelines. No systematic search for primary studies was performed. Cost-effectiveness issues were not addressed.
Results
The search resulted in 154 hits. Numerous international guidelines and screening
recommendations and other potentially relevant articles and reports were identified
through the manual search. A total of 53 articles in full text were reviewed against
specified inclusion-exclusion criteria. The most commonly used CRC screening tests are the Fecal Occult Blood Test (FOBT), flexible sigmoidoscopy, and colonoscopy. A positive test result for FOBT will usually be followed by colonoscopy. The sensitivity and specificity of the FOBT vary for different brand types (gFOBT and FIT). A systematic review reports the pooled estimate of CRC mortality reduction using the gFOBT as a 16% reduction in the relative risk of CRC mortality as a consequence of annual or biannual screening. There was a 15% relative risk reduction in CRC mortality when only results from biennial
screening programs were included. Attendance rates are diminishing over
time and the FOBT screening does not seem to affect the incidence of CRC.
The literature identified in this review supports the suitability of flexible sigmoidoscopy and colonoscopy as a screening tool to detect colorectal abnormalities. Endoscopic screening can theoretically prevent development of CRC, but we are still awaiting the evidence from well performed controlled studies. Results from completed randomised trials with sigmoidoscopy will be soon published. A randomized controlled trial with colonoscopy screening is planned(NORDCCAP- II/NordICC). Other screening methods, such as virtual colonoscopy or DNA stool testing for tumor markers are also under study and the evidence of these methods has not been assessed yet.
Conclusions
Several randomized screening trials have demonstrated a reduction in CRC mortality by repeated FOBT testing annually or biannually followed by colonoscopy for participants with positive test results.
There is so far no solid evidence on CRC mortality reduction for the other primary screening tests that are available, but this will change with the reporting of findings from ongoing trials. It seems that colorectal screening has no effect on total mortality reduction. Most of the countries that have implemented CRC screening program use stool tests (gFOBT or FIT) as the primary screening tests followed by colonoscopy for all participants with a positive FOBT.