Article Text

PDF

Population based endoscopic screening for colorectal cancer
  1. W S Atkin,
  2. J M A Northover
  1. Cancer Research UK Colorectal Cancer Unit, St Mark’s Hospital, Northwick Park, Middlesex, UK
  1. Correspondence to:
    W S Atkin
    wendy.atkin{at}cancer.org.uk

Statistics from Altmetric.com

We propose that there should be a national colorectal cancer screening programme in the UK, and that there is a good case to base it on a single flexible sigmoidoscopy (FS) at approximately 60 years of age, possibly supplemented by faecal occult blood testing (FOBT) after age 60 years, predominantly for the detection of early proximal cancers.

CASE FOR SCREENING

There is no longer doubt that screening is an effective method of reducing colorectal cancer (CRC) incidence and mortality rates. The US Preventive Services Task Force1 recently reviewed the evidence and gave a grade A recommendation that all men and women over the age of 50 years should be screened for CRC. In the UK, the Department of Health has demonstrated its commitment to CRC screening by funding a national demonstration pilot to assess the feasibility and acceptability of FOBT in the general population.2 Together with the MRC it is also funding a randomised trial of a single FS screen offered at around age 60 years with colonoscopy for those found at FS to have high risk adenomas. The trial recently reported3 that FS is acceptable, feasible, and safe but has yet to report on the magnitude and duration of efficacy. The FOBT pilot was completed in mid-2002 and results are due to be presented to ministers in early 2003. Colonoscopy is not considered a feasible option for mass screening in the UK because of the personal commitment required of the screenee, the risks of perforation, and manpower issues.

GOAL FOR SCREENING: PREVENTION IS BETTER THAN EARLY DETECTION

The FOBT regimen under consideration in the UK is the unrehydrated guaiac test, offered biennially between the ages of 50 and 69 years. In two European trials4,5 this regimen produced a 15–18% reduction in CRC mortality rates. Case control studies, which allow estimates of the maximum effect achievable assuming complete compliance, show that FOBT and FS reduce CRC mortality rates by a similar proportion (approximately 35%).6,7 Importantly, the effect of a single FOBT in protecting against the diagnosis of a fatal CRC appears to abate after two years whereas the effect of a single FS examination lasts at least 10 years. Therefore, FOBT must be repeated at least every two years while FS is effective when offered very infrequently. We have suggested that it needs to be done only once if offered at approximately 60 years of age.8

The most important advantage of FS is its ability to prevent CRC through detection and treatment of adenomas, the precursor of most CRC. The evidence comes from several large case control9 and cohort10,11 studies and a small Norwegian trial.12 The reduction in CRC incidence seen after 18 years in the US trial of rehydrated FOBT13 was almost certainly due to the very high rate of colonoscopy resulting from the low specificity of the test. (In this trial, 38% of the annually and 28% of the biennially screened groups had at least one colonoscopy for a positive test.) No reduction in CRC incidence has been observed in European trials which used the unrehydrated FOBT, probably because cumulative colonoscopy rates were much lower (approximately 4%).

Prevention has advantages over early cancer detection. Firstly, the costs of the screening programme can be offset against avoided treatment costs (surgery, adjuvant chemo- and radiotherapy), imaging investigations, and clinical follow up. CRC is an expensive disease to manage and is set to become more costly with the introduction of new imaging, adjuvant, and palliative regimens. It has been suggested that FS screening may even be cost saving in the long term.14

Screening for cancer induces more anxiety than screening for premalignant lesions. While there have been no direct comparisons of FOBT and FS, work on cervical screening suggests that anxiety on receiving an abnormal result is lessened when people are given a leaflet explaining that cancer is unlikely.15 Such reassurance cannot be given with FOBT because 10% of people with a positive test result will be found to have cancer.

COMPLIANCE AND ACCEPTABILITY

Compliance rates with a single FS have been reported to vary between 39% and 50% in the UK.16 When FS and FOBT are offered together, attendance rates have been low,16,17 possibly because of confusion about the role of each test in that situation. Evidence that higher uptake rates with a single FS screen are achievable in a programme comes from Norway12 and Northern California (Selby J, personal communication) where they are above 70%. Maximum effectiveness with FOBT screening requires long term regular screening. FOBT offered biennially between 50 and 69 years of age represents 10 opportunities to fail to comply. Studies show that approximately 60% will do the test once but less than 40% will complete all tests offered.4,18 If FOBT screening is started at age 50 years, when CRC incidence rates are low, it is likely that compliance rates will have fallen by age 60 years (five rounds later) when cancer incidence is increasing. It might be better to start after age 60 years to detect the cancers (mainly proximal) not prevented by a prior FS screen.

Advantages of flexible sigmoidoscopy for colorectal cancer screening

  • Detects adenomas, precursors of cancer.

  • Reduces incidence and therefore morbidity.

  • By preventing disease, costs of screening can be offset against avoided treatment costs.

  • Takes only five minutes and there is an immediate result.

  • Small polyps can be removed during screening.

  • Very low complication rate.

  • Infrequent examination.

  • High attendance rates.

  • Can be performed by nurses.

The FS screening regimen in the UK trial is quick, taking only 4–5 minutes, and an immediate result is available for most screenees; it is safe, with only one perforation in 50 000 FS3,19; it is efficient in that screening and polypectomy can be achieved on a single visit in 95% of individuals3; and it minimises the need for subsequent colonoscopy. The 5% who are offered a colonoscopy are those whose screen detected polyps have characteristics which indicate a higher risk of proximal cancer.20 With FOBT, a 1–2% positivity rate at each screening round results in a cumulative colonoscopy rate across the 50–69 year age range of 10–20% among compliers. Reducing the age range, by starting FOBT at 60 years, would decrease the cumulative positivity rate and also increase cost effectiveness.21

We can now predict that there will be manpower problems with whatever form of screening is selected. Nurses and non-specialist medical endoscopists can perform screening FS as well as gastroenterologists,22 but they are currently in short supply. Both FS and FOBT will put pressure on already stretched colonoscopy services, so it is essential to address these issues now. This will require training new endoscopists (medical and non-medical) and retraining established endoscopists to meet the rigorous standards required of a screening programme. The rewards would be great. The dramatic reductions in incidence rates of cancers of the sigmoid colon and rectum over the past 15 years in the USA have been attributed to sigmoidoscopy and polypectomy.23 During this same time period in the UK, CRC incidence rates have remained unchanged.24

Our aim in the UK must be to eradicate CRC rather than to increase, through earlier detection, the duration of the cancer experience and thereby the physical and psychological burden of suffering from the disease. Ultimately, it might be cost effective to exploit the long benign phase in CRC development using DNA markers in stool, which have the potential to be both sensitive and specific for precancerous lesions with high malignant potential. In the meantime, a policy based on a single FS at approximately 60 years of age offers the best prospect in the UK.

REFERENCES

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.