Screening for Colorectal Cancer

Darley Petersen

Background  

Colorectal cancer (CRC) remains the second most common cause of cancer death in Western Europe and North America.

Screening for CRC has not worldwide popularity in spite of proven survival benefit in average risk persons. Incidence and mortality, both are worrying in Europe as well as in the US, Australia, and Japan. Until 1995, publications of randomized studies with Hemoccult-II have been limited to Minnesota, Nottingham, and Funen. Similar results were obtained in the US, UK, and Denmark. In the United Kingdom, a pilot-screening program has been set up in Coventry & Warwickshire and in Scotland, including a population of nearly 2 million people. In Denmark, the national health authorities have published a medical-technology-valuation (MTV) suggesting similar pilot studies in one or two counties in Denmark.

A European Group for Colorectal Cancer Screening has supported such a policy. Guidelines based on the scientific evidence are available.

Methods, Material, and Results

Screening for CRC in asymptomatic persons 45-75 years has been evaluated in Denmark from 1985 to 2002. 30,970 persons were allocated to screening and 30,968 as controls. The aim was to compare mortality rates after fecal occult blood tests (Hemoccult-II) every 2 years during a 10-year period with those of unscreened similar controls.

After 10 years, the 5 screening rounds demonstrated a reduction in mortality from CRC of 18 percent. Results were published in The Lancet in 1996 (1). Results after 13 years were published in GUT 2002 (2). Results after 18 years will be published in Scand. J Gastroent 2004 (3).

The trial was continued to demonstrate a possible reduction in incidence of CRC. The ninth screening round ended in August 2002.

In the period 1985-2002, nine screening rounds were completed:

The fecal occult blood test selected for the study, Hemoccult-II test, has been used in combination with dietary restrictions (no red meat or fresh fruit, no iron preparations, vitamin C, aspirin, or other non-steroid antirheumatics during three days before sampling). Two fecal samples were taken by participants themselves from each of three consecutive stools, which were sent to the department of clinical chemistry at Odense University Hospital and evaluated by three technicians. Those with positive tests were invited for medical history, physical examination, and full colonoscopy. A double contrast barium enema was offered, when full colonoscopy could not be obtained.

All participants received invitations by letter and two reminders during the initial screening and one reminder during the following screening rounds. Each screening period took one year to complete. Only persons accepting the first screening round were invited to the next screening round.

Persons in whom CRC or adenomas were detected after randomisation, but before they could be invited, were automatically excluded from invitation. During the screening period all persons in the screening group as well as in the control group were followed through the regional health information database, and registrations were made if cancers or adenomas or death occurred.

The study was approved by the regional ethic committee, and the registers were approved by the National Registry Board.

Organization

The organization of the study includes, besides the main investigator who is a professor in surgery, a professor specialized in pathology in the digestive tract, and who has been a co-worker throughout the study.

A professor of social medicine also has taken part since the planning phase, being responsible for calculation of sample size and planning analysis of survival and death, and nested case control studies.

A professor of health economy and his assistant evaluated the socio-economic consequences. The analysis identified efficient screening programs being cost-effective health care inter-ventions.

An impartial death committee of five persons established causes of death in patients with colorectal neoplasia, unless autopsy or recent detailed examination revealed advanced spread from CRC. The death form was filled out without knowledge of which group the person belonged.

Deaths were registered within one week, and death certificates were obtained within six months from the National Board of Health.

The screening office had two health record professionals and a part time junior doctor (PhD student).

The screening office took care of scientific correspondence of all kinds.

The present study got financial support from different sources. Applications to many foundations were necessary. The majority of the scientific staff worked for free integrating these valuable and often large quantities of paperwork. All financial resources were used for materials and technology and salary for the daily management of the screening office, which means health records, technician, and EDP-programmer.

The organization was proven suitable for other studies with different fecal occult blood tests. The effect of adding flexible sigmoidoscopy to Hemoccult-II has been evaluated in collaboration with other European countries.

Conclusion

The trial was closed in 2003, and results will be published in 2004. The scientific results represent the background for attempts to introduce screening in the general average risk population. Screening may reduce mortality from colorectal cancer, but this benefit will only be obtained on national levels with an effective organization, including well-trained health record professionals (4).

References

1. Kronborg O, Fenger C, Olsen J, Jørgensen OD, Søndergaard O. Randomised study of screening for colorectal cancer with faecal-occult-blood test. Lancet 1996; 348:1467-71.
2. Jørgensen OD, Kronborg O, Fenger C. A randomised study for colorectal cancer with faecal occult blood test. Results after 13 years and 7 biennial screening rounds. GUT 2002; 50:29-32.
3. Kronborg O, Jørgensen O, Fenger C, Rasmussen M. Randomised study of biennial screening with fecal occult blood test. Results after nine screening rounds. Scand J Gastroenterol 2004 (in print).
4. Petersen D. Screening for colorectal cancer. Abstract. The 10 th European Conference on Health Records in Dublin August 2002: Managing Health Information in the 21 st Century. Irish Journal of Medical Science 2002; 171:48.

References in connection to earlier IFHRO Congresses

• Petersen D. A prospective randomised population study at Funen in Denmark. Administrative means for the health record. The 8th European Health Records Conference: Future of Patient Records, 1995, May 21-24 in Maastricht, The Netherlands. Proceedings 1995: 247-52
• Petersen D. A prospective randomised population study at Funen in Denmark. To provide knowledge of practical implementations of clinical trial. The 12th International Health Records Congress: The Future of Health Information Management 1996, April 15-19 in Munich, Germany. Proceedings 1996: 281-6
• Petersen D. Feasibility of population screening for cancer of the colon and rectum in the next millennium. 13th International Health Records Congress: Into the next millennium. A new world record, 2000, Oct. 2-6 in Melbourne, Australia. Proceedings: Abstract A-001
• Petersen D. Screening for colorectal cancer. Irish Journal of Medical Science 2002;171:48. Abstracts from the 10th European Conference on Health Records.