Screening for Colorectal Cancer. Part 1: Screening-Tests and Project Design

Duration: December 2009 - March 2010, Update: December 2010

External review: Franz Piribauer
Suggested by: Viterio-Schweiz GmbH, Irisweg 6, CH-8700 Küsnacht, Switzerland under contract with Krebsliga Schweiz, CH-3001 Bern, www.swisscancer.ch
Publication: HTA-Project report 41a: Screening for Colorectal Cancer. Part I: Screening-Tests and Project Design - https://eprints.aihta.at/873

Background:
Colorectal cancer (CRC) is a malignant tumor arising within the walls of the large intestine. Among both men and women CRC was the third most common non-skin cancer and also the third-highest cause of cancer deaths in the US in 2009. In terms of age-standardized incidence rates, there is little difference from one European country to another. CRC has a recognizable, protracted pre-malignant stage that is relatively easy to treat. If the disease is detected early, a person’s chances of survival are considerably higher than if it is detected at a later stage. That is why screening for CRC has been introduced in various organizational modes in a number of countries.

Aims and research questions:
The Swiss cancer league (Krebsliga Schweiz) requested a review of the secondary literature (health technology assessments, systematic reviews, meta-analyses) on CRC-screening to inform about policy options in this realm in December 2009 and an update in November 2010.

Research questions:
1. What screening-tests are available for colorectal cancer? What are the respective test characteristics and what are the respective test’s wider implication for a colorectal cancer-screening program?

2. What questions and central aspects are to be considered in the context of designing an organized population-based screening-program for colorectal cancer?

Methods:
A systematic literature search limited to secondary literature (health technology assessments, systematic reviews of the literature, meta-analyses) published from 1999-2009 was performed in Dec. 2009. This was supplemented by a small unsystematic search for literature on recent developments in molecular screening-tests. An update search for secondary literature was performed in November 2010.

Part 1 – Clinical findings:

Significance of colonoscopy in screening for colorectal cancer
Colonoscopy is the final common pathway of all screening for colorectal cancer (CRC) and is used for biopsy and polyp removal. For a screening-test in the (healthy) general population colonoscopy is invasive and prone to serious complications. Screening-yield and rates of complications are strongly dependent on the individual operator and on quality assurance. As a result, training and continued education of endoscopists as well as monitoring of both detection and complication rates are key to high screening-quality.

Effectiveness of screening for CRC
No data is currently available on the impact of CRC-screening on all-cause mortality. Four randomized controlled trials on screening for faecal occult blood as a first-line test (gFOBT) showed a relative risk reduction of 15% for disease-specific CRC-mortality. Results from one large randomized controlled trial in the UK on once only flexible sigmoidoscopy as a first-line test showed a relative risk reduction of 31% for diseases-specific CRC-mortality and a reduction in CRC-incidence of 23%. Preliminary findings from a randomized controlled trial in Norway showed no impact of screening. Results from ongoing sigmoidoscopy trials in the USA and Italy are expected later, as well as results after longer follow up from the Norwegian trial. Two randomized controlled studies on screening with colonoscopy as a first-line test will yield results no sooner than ten years from now. There is only limited evidence on test characteristics (sensitivity, specificity, complication rates) in real life screening-settings.

International screening-activities
In many countries the evaluation of evidence, the planning and at times the coordination of CRC-screening are done by a national institution. A few countries – England, Scotland, Finland and Australia – run organized population-based programs. Ireland will introduce such a program in 2012. However, most screening is not population-based but opportunistic. Often participation rates are low. Some countries – Japan, Italy and Germany – have programs that have been under way for many years. In the European Union about 70% of the population has access to some mode of CRC-screening. The most common first-line screening-test is gFOBT, to a degree also iFOBT. In some countries endoscopic-screening – colonoscopy, flexible sigmoidoscopy – is used as an alternative or in combination with FOBT. Mostly due to remuneration decisions by health insurers in the US colonoscopy is the most common first-line screening-test there.

Choice of first-line test
When considering first-line screening-tests on which to base an organized program, the test’s impact on participation is more important than its individual test-sensitivity. Program-sensitivity largely depends on participation rates. Recent developments in first-line screening include quantitative iFOBTs. CT-colonoscopy, capsule endoscopy and new molecular tests are not yet viable alternatives for use in population-based mass-screening.

Improving screening-effectiveness
An upper age-limit for CRC-screening is recommended. An integrated screening-program combines screening with screening-relevant considerations in diagnosis, treatment and surveillance. Along with standardized documentation and regular evaluation, this integrated program-design provides the quality necessary to justify screening average risk-populations. Giving thorough attention to the design of the surveillance regime is important, because its thresholds determine the numbers of surveillance-colonoscopies resulting from CRC-screening. Incremental implementation of a national population-based screening-program, with pilot testing and incremental roll-out, should be considered.

Securing comprehensive program-financing
Apart from the financing the actual delivery of screening services, funding of a program infrastructure is necessary. Population-based screening-programs require significant initial investment for setting up sophisticated overhead structures and long term sustainable financing for ongoing documentation, quality assurance and evaluation. Also, ongoing funding of both program- and provider-independent information dissemination to potential screening-participants and funds for regular program evaluation through an external institution needs to be secured.