H. Lee Moffitt Cancer Center & Research Institute

Search Moffitt.org


Pierre Auguste Renoir (French, 1841-1919), The Seine (detail), 1879

Screening for Colorectal Cancer

Bernard Levin, MD

Colorectal cancer is a major cause of morbidity and mortality in the United States. Early detection of the disease at an asymptomatic stage by screening holds promise for lowering the incidence of colorectal cancer deaths, yet compliance with screening guidelines is poor. Evidence in support of the use of screening techniques for colorectal cancer is accumulating, however, and screening for this disease with fecal occult blood tests and flexible sigmoidoscopy can increase the likelihood of early detection, can decrease mortality and morbidity, and can be cost effective.

Introduction

In the United States, colorectal cancer is the second most common cancer in women (after breast cancer) and the third most common cancer in men (after prostate and lung cancers). In 1995, approximately 150,000 new cases will occur and approximately 58,000 persons will die of colorectal cancer.

Between 1987 and 1991, overall incidence rates for colorectal cancer decreased at an annual rate of 1.4%, although black men experienced a statistically nonsignificant increasing incidence rate of 0.5% per year. Mortality rates have followed a similar pattern. Over the same time period, overall mortality decreased at an annual rate of 1.8%, although mortality for black men increased at an annual rate of 0.3%.[1]

Risk Assessment

The assessment of factors associated with colorectal cancer may be expressed in terms of relative or absolute risk. Relative risk (or the risk ratio) for colorectal cancer is the ratio of the incidence rate among individuals with specific risk factors to those without specific risk factors. Absolute risk represents the probability that an individual with given risk factors will develop colorectal cancer over a defined period of time. The lifetime probability (magnitude of absolute risk) for the development of colorectal cancer for US men and women is approximately 6%. Risk factors for the development of colorectal cancer are listed in Table 1.

Rationale for Screening

The sine qua non of screening is that management of disease detected at an asymptomatic stage is more effective than treatment at the time of overt clinical presentation. In the absence of effective management, cancer screening may appear to be effective because it preferentially detects slower-growing neoplasms (length-time bias), thus advancing (in time) the diagnosis of incurable cancers (lead-time bias) and resulting in more years of life with cancer without lengthening life span. This is not the case with colorectal cancer, since effective therapy is available if cancers are detected early.

Screening can be carried out in the general population (mass screening) or under the supervision of a medical or nurse practitioner (case finding). The American Cancer Society,[2] the American Gastroenterological Association,[3] and the World Health Organization for the Prevention of Colorectal Cancer[4] have encouraged periodic fecal occult blood testing (FOBT) in conjunction with flexible sigmoidoscopy for those at average risk. Following a patient's entrance into the health case system (ie, "case finding"), the specific recommendations involve the following steps:

  • Symptomatic individuals undergo diagnostic procedures (eg, colonoscopy or double-contrast barium enema).
  • Asymptomatic individuals are evaluated for risk factors. Those in higher-risk categories may require not only genetic testing and counseling to determine inherited susceptibility, but also endoscopic evaluation. Family members may also require such measures.
  • Those without symptoms undergo FOBT annually in conjunction with flexible sigmoidoscopy every three to five years beginning at 50 years of age.

Although not recommended previously by the US Prevention Services Task Force, its recent report recommends screening for colorectal cancer with modest enthusiasm.[5]

Evaluation of Screening Procedures

Implementation of a screening program in a distinct population cannot be assumed to be beneficial to that population; the potential efficacy, benefits, and disadvantages of any cancer screening program must be based on quantitative approaches of evaluation - its validity, efficacy, compliance, and cost. The National Cancer Institute has established levels of evidence to evaluate data in support of screening procedures (Table 2). The following discussion applies to asymptomatic individuals at standard risk.

Fecal Occult Blood Testing

The fact that colorectal tumors tend to bleed is the rationale for FOBT.[6,7] The goal of testing is to select individuals for colonoscopy (or flexible sigmoidoscopy plus double-contrast barium enema, if colonoscopy is unavailable) as a result of identifying early cancers and large adenomas (more than 1 cm). Two main types of FOBTs are available: guaiac and immunochemical. Guaiac tests are based on the pseudoperoxidase activity of heme, and immunochemical tests utilize antibodies to human hemoglobin.

In screening programs, an individual with a positive FOBT has a 30% to 45% chance of having an adenoma and a 3% to 5% chance of having colorectal cancer. Traditional card-based guaiac tests (Hemoccult) will detect 40% to 80% of asymptomatic colorectal cancers.[8,9] Thus, this type of test will miss at least 20% of colorectal cancer under ideal testing conditions and possibly up to 50% if the appropriate testing procedures are not followed. When dietary and medication restrictions are followed, a guaiac-based test such as Hemoccult is highly specific, ie, 98% of healthy individuals will be negative.[10] The implication of a low positive-predictive value for cancer (eg, 5.6% for unhydrated and 2.2% hydrated in the Minnesota trial) is that the majority of positive results are falsely positive but still require further investigation. However, some additional benefit will accrue from the identification and removal of adenomas.

Four large randomized trials (Nottingham, UK; G&umlo;teborg, Sweden; Funen, Denmark; and Minnesota) and one nonrandomized trial (New York) have been or are being conducted (Table 3). All involve the use of a hydrated or unhydrated guaiac-based test kit (Hemoccult). These studies vary considerably in design as well as in method of follow-up of those with positive results.

The New York study, which assigned individuals to screening and control groups, has reported a 43% reduction in colorectal cancer mortality with annual screening in the "first-timers" subgroup after nine years of follow-up.[7] The University of Minnesota trial is the first prospectively randomized trial of screening to have unequivocally demonstrated a significant reduction in colorectal cancer mortality.[11] After 13 years of follow-up, the investigators observed a 33% reduction with annual screening and a nonsignificant reduction of approximately 5% with biennial screening. Data are still being collected in the latter group. Because approximately 80% of the samples were hydrated, the positivity rate was high (9.8%). This resulted in a rate of colonoscopy at approximately 35%. In other studies, rehydration has been found to produce a high positivity rate, especially in a community setting.[12] The other randomized studies have yet to report definitive results, although after 38 months of follow-up in the Danish study, a 19% decrease in colorectal mortality in the screened group has been reported (P=0.24).[13]

Some controversy remains concerning the possible mechanisms responsible for the mortality reduction in the Minnesota trial and other trials of FOBT. Estimates of sensitivity for early, surgically curable cancers from diverse studies (that are often very different in design) range from 30% to 92% (the latter in the Minnesota trial). Critics have suggested that the high colonoscopy rate in the Minnesota trial may be responsible for the substantial reduction in mortality.[14] In these trials, adenomas may be detected because of bleeding (but fewer than 30% of cancers and large polyps bleed sufficiently to be detected by FOBT). Subsequent analysis has suggested that colonoscopy was not random and that chance colonoscopy cannot be assumed to be the cause of the reduction in mortality. For example, colonoscopic polypectomy in the National Polyp Study resulted in an over 80% reduction in subsequent colon cancer.[15] These speculations do not detract from the importance of the positive result in the Minnesota trial. However, it is important to understand the explanations for the results so that future use of FOBT may be even more efficient and cost effective.

New Tests

Immunochemical tests for human hemoglobin have been studied in the United States, Japan, and Australia. Data from recent nonrandomized trials show that an immunochemical test (HemeSelect) had a 97% sensitivity for colorectal cancer and a 76% sensitivity for adenomas of more than 1 cm in diameter. Estimated specificity was 97.8%.[16] Future large-scale studies may incorporate such immunochemical tests, of which a major advantage is the absence of a need for dietary modification. Tests under development and evaluation are aimed at detecting mutant genes in the stool of patients with adenomas or cancer based on the molecular pathogenesis of colorectal cancer.[17]In preliminary studies, identical ras mutations could be identified both in the stool and in cancer or adenoma.[18,19] Circulating DNA also can be detected in the blood of patients with neoplasms of the alimentary tract.[20] It remains to be seen whether these molecular diagnostic tests will be useful.[19]

Flexible Sigmoidoscopy

Two recent case-control studies of sigmoidoscopic screening have provided the first evidence from controlled studies that sigmoidoscopy can reduce colorectal cancer mortality.[21,22] In both studies, a reduction in mortality of approximately 70% to 80% was reported from cancers within reach of the sigmoidoscope. The beneficial effect extended for approximately 10 years. Because 50% of all colorectal cancers are detectable by the 60-cm sigmoidoscope, these studies suggest that periodic sigmoidoscopic screening could reduce overall colorectal cancer mortality by one third. Although current guidelines established by the American Cancer Society recommend flexible sigmoidoscopy screening every three to five years, a recent study by Rex et al[23] supports lengthening this interval to five years. A randomized, controlled study by the National Cancer Institute to evaluate effectiveness of screening for colorectal cancer with periodic flexible sigmoidoscopy is in its early phases in the United States,[24] as are similar trials in the United Kingdom.[25] These are ambitious studies that will take approximately 15 years to complete and analyze.

Colonoscopy

It has been suggested that colonoscopy be tested as a screening tool for individuals at 60 years of age.[25] This idea is supported by a recent report that flexible sigmoidoscopy has only a 44% sensitivity for colonic adenomas. It is proposed that colono scopy every 10 years be investigated as a primary screening modality at and after 60 years of age. A national study to evaluate colonoscopy as a screening technique is being developed.[26]

Double-Contrast Barium Enema Examinations

Radiologists maintain that barium enema examinations are safer, are less time consuming for the patient, have the potential of examining the entire colon, and are less expensive than endoscopic studies.[27] While a barium enema can detect lesions equal to or greater than 1 cm in diameter in 85% of individuals, it is considerably less sensitive for smaller lesions. Practical problems include a lack of adequate numbers of radiologists skilled in the performance of high-quality double-contrast barium enema examinations. Virtual colonoscopy,[28] a new technique involving a helical computed tomography scan and a three-dimensional reconstruction, is presently in the experimental/developmental phase.

Compliance

Despite the available screening procedures for early detection of colorectal cancer, compliance is poor. Estimates (outside of controlled clinical trials) are that 18% to 50% of individuals avail themselves to screening by flexible sigmoidoscopy and occult blood testing.[29] Public and professional education is needed in this respect.

Cost Effectiveness

Screening can reduce mortality associated with colorectal cancer by detecting the disease in its early stage and thus preventing its development. Ideally, the benefits of such a screening program should exceed the costs incurred. Wagner et al[30] assessed the cost effectiveness of a periodic program of colorectal cancer screening for the elderly (Table 4). They examined four different schedules, each of which consisted of annual FOBT beginning at 65 years of age and continuing until either death or the age of 85 years plus (1) flexible sigmoidoscopy every three years, (2) flexible sigmoidoscopy every five years, (3) one flexible sigmoidoscopy at 65 years of age, or (4) no sigmoidoscopy. It was assumed that members of the study group would undergo repeated screening tests, follow-up diagnostic testing in the event of positive screening results, polypectomy in the event of polyp detection during screening or follow-up, and periodic surveillance after polypectomy to search for new polyps. Regardless of the screening schedule used, colorectal cancer screening is potentially expensive. If the entire population of 65-year-old persons complied with recommendations for periodic screening, the costs would be as high as $2.6 billion. However, the costs of treating advanced cancer, as well as the enhanced quantity and quality of life produced by each successful early detection, also need to be considered. The model is sensitive to the cost of follow-up surveillance by colonoscopy; further decreases of the cost of colonoscopy will reduce the overall costs substantially. A recent cost effectiveness model emphasized the significance of compliance as a determinant of effectiveness of screening.[31]

The cost effectiveness of colorectal cancer screening is estimated to be approximately $40,000 per year of life gained. These costs compare favorably with the costs of other preventive services for the elderly. For example, annual breast screening with m ammography would cost approximately $34,500 per year of life gained. Hemodialysis for end-stage renal disease, a service specifically legislated as a Medicare benefit, has been estimated to cost $36,000 per year of life saved. The absence of Medicare coverage for colorectal cancer screening is a serious deterrent to its widespread implementation.

Conclusions

It is encouraging that colorectal mortality rates continue to fall for the majority of the population. Although not perfect, colorectal cancer screening can be reasonably cost effective. With the advent of improved initial tests (immunochemical FOBTs or molecular genetic-based tests) and a reduction in colonoscopy expenses, the costs are likely to diminish even further.

Future advances in endoscopy and imaging may enhance feasibility and compliance of colorectal screening. Health care providers should not only be aware of the techniques currently available to further decrease the mortality associated with colorectal cancer, but also encourage their patients to avail themselves of these methods. It is to be hoped that third-party reimbursement (including Medicare) will support the widespread use of colorectal cancer screening as an effective, efficient preventive measure.

References

  1. Brown M. National Cancer Institute, 1994 (personal communication).
  2. Levin B, Murphy GP. Revision in American Cancer Society recommendation for the early detection of colorectal cancer. CA Cancer J Clin. 1992;42:296-299.
  3. Fleischer DE, Goldberg SB, Browning TH, et al. Detection and surveillance of colorectal cancer. JAMA. 1989;261:580-585.
  4. Winawer SJ, St John J, Bond J, et al. Screening of average-risk individuals for colorectal cancer. Bull World Health Organ. 1990;68:505-513.
  5. Guide to Clinical Prevention Services: Report of the U.S. Preventive Services Task Force. 2nd ed. Baltimore, Md: Williams & Wilkins; 1995.
  6. Simon JB. Occult blood screening for colorectal carcinoma: a critical review. Gastroenterology. 1985;88:820-837.
  7. Winawer SJ, Schottenfeld D, Flehinger BJ. Colorectal cancer screening. J Natl Cancer Inst. 1991;83:243-253.
  8. Kronborg O, Fenger C, Olsen J, et al. Repeated screening for colorectal cancer with fecal occult blood test: a prospective randomized study at Funen, Denmark. Scand J Gastroenterol. 1989;24:599-606.
  9. Hardcastle JD, Thomas WM, Chamberlain J, et al. Randomised, controlled trial of faecal occult blood screening for colorectal cancer: results for first 107,349 subjects. Lancet. 1989;1:1160-1164.
  10. Allison JE, Feldman R, Tekawa IS. Hemoccult screening in detecting colorectal neoplasm: sensitivity, specificity and predictive value. Long-term follow-up in a large group practice setting. Ann Intern Med. 1990;112:328-333.
  11. Mandel JS, Bond JH, Church TR, et al. Reducing mortality from colorectal cancer by screening for fecal occult blood. N Engl J Med. 1993;328:1365-1371.
  12. Levin B, Hess K, Johnson C, et al. Screening for colorectal cancer: the impact of rehydration on the Hemoccult fecal occult blood test. Gastroenterology. 1995;108:A496.
  13. Kronborg O, Fenger C, Worm J, et al. Causes of death during the first 5 years of a randomized trial of mass screening for colorectal cancer with fecal occult blood test. Scand J Gastroenterol. 1992;27:47-52.
  14. Lang CA, Ransohoff DJ. Fecal occult blood screening for colorectal cancer: is mortality reduced by chance selection for screening colonoscopy? JAMA. 1994;271:1011-1013.
  15. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med. 1993;329:1977-1981.
  16. St John DJ, Young GP, Alexeyeff MA, et al. Evaluation of new occult blood tests for detection of colorectal neoplasia. Gastroenterology. 1993;104:1661-1668.
  17. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell. 1990;61:759-767.
  18. Sidransky D, Tokino T, Hamilton SR, et al. Identification of ras oncogenes in the stool of patients with curable colorectal tumors. Science. 1992;256:102-105.
  19. Smith-Ravin J, England J, Talbot IC, et al. Detection of c-Ki-ras mutations in faecal samples from sporadic colorectal cancer patients. Gut. 1995;36:81-86.
  20. Sorenson GD, Pribish DM, Valone FH, et al. Soluble normal and mutated DNA sequences from single-copy genes in human blood. Cancer Epidemiol Biomarkers Prev. 1994;3:67-71.
  21. Selby JV, Friedman DG, Quesenberry CP Jr, et al. A case-control study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med. 1992;326:653-657.
  22. Newcomb PA, Norfleet RG, Storer BE, et al. Screening sigmoidoscopy and colorectal cancer mortality. J Natl Cancer Inst. 1992;84:1572-1575.
  23. Rex DK, Lehman GA, Ulbright TM, et al. The yield of a second screening flexible sigmoidoscopy in average-risk persons after one negative flexible examination. Gastroenterology. 1994;106:593-595.
  24. Gohagan J. PLCO Cancer Screening Trial. 1995 (personal communication).
  25. Atkin WS, Cuzick J, Northover JM, et al. Prevention of colorectal cancer by once-only sigmoidoscopy. Lancet. 1993;341:736-740.
  26. Winawer S. 1995 (personal communication).
  27. Gelfand DW, Ott DJ. The economic implications of radiologic screening for colonic cancer. AJR Am J Roentgenol. 1991;156:939-943.
  28. Vining D. 1994 (personal communication).
  29. Ahlquist DA. Occult blood screening: obstacles to effectiveness. Cancer. 1992;70:1259-1265.
  30. Wagner JL, Herdman RC, Wadhwa S. Cost effectiveness of colorectal cancer screening in the elderly. Ann Intern Med. 1991; 115:807-817.
  31. Lieberman DA. Cost-effectiveness model for colon cancer screening. Gastroenterology. 1995;109:1781-1790.

    From the Division of Cancer Prevention at The University of Texas M.D. Anderson Cancer Center, Houston, Tex.

    Back to Cancer Control Journal Volume 3 Number 1

© Copyright 1996 - 2009 H. Lee Moffitt Cancer Center & Research Institute