Summary of Recommendations and Evidence

The U.S. Preventive Services Task Force (USPSTF) recommends screening for colorectal cancer in adults using fecal occult blood test (FOBT), sigmoidoscopy, or colonoscopy, beginning at 50 years of age and continuing until 75 years of age. The risks and benefits of these screening methods vary (Table 1). A recommendation.

The USPSTF recommends against routine screening for colorectal cancer in adults 76 to 85 years of age. There may be considerations that support colorectal cancer screening in an individual patient. C recommendation.

The USPSTF recommends against screening for colorectal cancer in adults older than 85 years. D recommendation.

The USPSTF concludes that the evidence is insufficient to assess the benefits and harms of computed tomographic (CT) colonography and fecal DNA testing as screening modalities for colorectal cancer. I statement.

Rationale

Importance. Colorectal cancer is the third most common cancer and the second leading cause of cancer death in the United States. Current levels of screening in this country lag behind those of other effective cancer screening tests; it has been estimated that attainment of goals for population colorectal cancer screening could save 18,800 lives per year.¹ Colorectal cancer incidence and mortality show health disparities, with a disproportionate burden occurring in certain minority populations, including blacks and Alaska natives.^2,3

Detection. The evidence is convincing that screening for colorectal cancer with FOBT, sigmoidoscopy, or colonoscopy detects early-stage cancer and adenomatous polyps.

Although colonoscopy is considered to be the reference standard against which the sensitivity of other colorectal cancer screening tests are compared, it is not perfect. Two types of studies to assess the sensitivity of colonoscopy—tandem colonoscopy studies, in which the same patient is studied twice, and studies comparing colonoscopy and CT colonography—show that colonoscopy may miss polyps larger than 10 mm and colorectal cancer. In addition, most of the evidence about the sensitivity of colonoscopy comes from experienced examiners in research settings. The evidence is inadequate to estimate the sensitivity in community practice; however, it is likely to be lower than in research settings.

Although single test performance is an important issue in the detection of colorectal neoplasia, the sensitivity of the test over time is more important in an ongoing screening program. Data that permit assessment and comparison of screening methods to detect colorectal neoplasia in a testing program over time from a population perspective are limited to data from analytic modeling.

Benefits of detection and early intervention. There is convincing evidence that screening with any of the three recommended tests reduces colorectal cancer mortality in adults 50 to 75 years of age. Follow-up of positive screening test results requires colonoscopy regardless of the screening test used. Because of the harms of colonoscopy, the chief benefit of less invasive screening tests is that they may reduce the number of colonoscopies required and their attendant risks.

There is adequate evidence that the benefits of detection and early intervention decline after 75 years of age. There is a substantial lead time between the detection and treatment of colorectal neoplasia and a mortality benefit, and competing causes of mortality make it progressively less likely that this benefit will be realized with advancing age.

Harms of detection and early intervention. The primary established harms of colorectal cancer screening are from the use of invasive procedures initially or in the evaluation sequence. Harms may arise from the preparation the patient undergoes before the procedure, the sedation used during the procedure, and the procedure itself.

Colonoscopy. Evidence is adequate to estimate the harms of colonoscopy. In the United States, perforation of the colon occurs in an estimated 3.8 per 10,000 procedures.⁴ Serious complications—defined as deaths attributable to colonoscopy or adverse events requiring hospital admission, including perforation, major bleeding, diverticulitis, severe abdominal pain, and cardiovascular events—are significantly more common, occurring in an estimated 25 per 10,000 procedures.⁵

Flexible sigmoidoscopy. Evidence is adequate that serious complications occur in approximately 3.4 per 10,000 procedures.⁵

Fecal tests. Evidence about the harms of fecal tests is lacking (inadequate), but the USPSTF assesses them to be no greater than small.

CT colonography. CT colonography images more than the colon. In up to 16 percent of persons having their first CT colonography, extracolonic abnormalities that require further testing are found.^5,6 Evidence is inadequate to assess the clinical consequences of identifying these abnormalities, but there is potential for benefit and harm. Potential harms arise from additional diagnostic testing and procedures for lesions found incidentally, which may have no clinical significance. This additional testing also has the potential to burden the patient and adversely impact the health system.

The risks of perforation associated with screening CT colonography in research settings are estimated to be 0 to 6 per 10,000 CT colonography studies.⁴ However, these estimates may be higher than what can be expected in screened populations because the studies included symptomatic populations.

Radiation exposure from CT colonography is reported to be 10 mSv per examination. The harms of radiation at this dose are not certain, but the linear no-threshold model predicts that one additional person per 1,000 would develop cancer in his or her lifetime at this level of exposure.⁷ The lifetime cumulative radiation risk from the use of CT colonography to screen for colorectal cancer should be considered in the context of the growing cumulative radiation exposure from the use of other diagnostic and screening tests that involve radiation exposure. On the other hand, improvements in CT colonography technology and practice are lowering this radiation dose.

USPSTF assessment. The USPSTF concludes that, for FOBT, flexible sigmoidoscopy, and colonoscopy to screen for colorectal cancer, there is high certainty that the net benefit is substantial for adults 50 to 75 years of age. The USPSTF concludes that, for adults 76 to 85 years of age, there is moderate certainty that the net benefits of screening are small.

The USPSTF concludes that, for adults older than 85 years, there is moderate certainty that the benefits of screening do not outweigh the harms.

The USPSTF concludes that there is insufficient evidence to assess the sensitivity and specificity of fecal DNA testing for colorectal neoplasia, and that therefore the balance of benefits and harms cannot be determined for this test.

The USPSTF concludes that, for CT colonography, evidence to assess the harms related to extracolonic findings is insufficient, and the balance of benefits and harms cannot be determined.

Clinical Considerations

PATIENT POPULATION

These recommendations apply to adults 50 years and older, excluding those with specific inherited syndromes (Lynch syndrome or familial adenomatous polyposis) and those with inflammatory bowel disease. The recommendations do apply to persons with first-degree relatives who have had colorectal adenomas or cancer, although for those with first-degree relatives who developed cancer at a younger age or with multiple affected first-degree relatives, an earlier start to screening may be reasonable. Data suggest that colorectal cancer has a higher mortality rate in blacks. The reasons for this differential are not well known, and the recommendations are intended to apply to all ethnic and racial groups.

When the screening test leads to the diagnosis of clinically significant colorectal adenomas or cancer, the patient will be followed by a surveillance regimen and recommendations for screening are no longer applicable. The USPSTF did not address evidence for the effectiveness of any particular surveillance regimen after diagnosis or removal of adenomatous polyps.

SCREENING TESTS

The relative sensitivity and specificity of the different colorectal screening tests with adequate data to assess cancer detection—colonoscopy, flexible sigmoidoscopy, and fecal tests—can be depicted as follows:

Sensitivity: Hemoccult II < fecal immunochemical tests ≤ Hemoccult SENSA ≈ flexible sigmoidoscopy < colonoscopy

Specificity: Hemoccult SENSA < fecal immunochemical tests ≈ Hemoccult II < flexible sigmoidoscopy = colonoscopy

For the operator-dependent tests—flexible sigmoidoscopy, CT colonography, and colonoscopy—better operator training and more experience have a high likelihood of improving sensitivity. Approaches related to certification, such as quality standards and possibly minimum volume requirements, could be used to achieve the goal of improving operator performance and, therefore, test sensitivity. Assurance of performance of high-quality endoscopy should be part of all screening programs.

Because several screening strategies have similar efficacy, efforts to reduce colon cancer deaths should focus on implementation of strategies that maximize the number of persons who get screening of some type. The different options for colorectal cancer screening tests are variably acceptable to patients; eliciting patient preferences is one step in improving adherence. Ideally, shared decision making between physicians and patients would incorporate information on local test availability and quality, as well as patient preference.

SCREENING INTERVALS AND STARTING AND STOPPING AGES

Screening for colorectal cancer reduces mortality through detection and treatment of early-stage cancer, and detection and removal of adenomatous polyps. The degree to which each of these mechanisms contributes to a reduction in mortality is unknown, although it is likely that the largest reduction in colorectal cancer mortality during the 10 years after initial screening comes from the detection and removal of early-stage cancer. Colonoscopy is a necessary step in any screening program that reduces mortality from colorectal cancer. This reduction in mortality comes at the expense of significant morbidity associated with colonoscopy. Evidence does not currently allow a differential estimate of colonoscopy-related morbidity for different age groups or for examinations done with or without biopsy.

In this context, the best measure for the morbidity that results from any screening program for colorectal cancer is the number of colonoscopies required to achieve a reduction in mortality. Although improvements in mortality will generally be associated with increasing morbidity that results from the screening and surveillance program, the goal of a screening program should be to maximize the number of life-years gained while minimizing the harms.

In a report prepared for the USPSTF by two groups in the Cancer Intervention and Surveillance Modeling Network, investigators conducted microsimulation analyses that applied programs of screening to standard populations of adults in the United States.⁵ These analyses permitted a comparison of expected outcomes among testing strategies involving the fecal tests, flexible sigmoidoscopy, or colonoscopy. In the models, the predicted total number of colonoscopies included those resulting from surveillance after detection of colorectal neoplasia. The models assumed lifetime monitoring by colonoscopy every three to five years, depending on the number and size of the adenomas detected. It is not the intent of the USPSTF to endorse this particular approach to surveillance, but standardizing the approach to surveillance is necessary to compare screening strategies in the models.

For all screening modalities, starting screening at 50 years of age led to a balance between life-years gained and colonoscopy risks that was more favorable than commencing screening earlier. Despite the increasing incidence of colorectal adenomas with age, for persons previously screened, the gain in life-years associated with extending screening from 75 to 85 years of age was small in comparison to the risks of screening persons in this decade. For adults who have not previously been screened, decisions about first-time screening in this age group should be made in the context of the individual's health status and competing risks, given that the benefit of screening is not seen in trials until at least seven years later. For persons older than 85 years, competing causes of mortality preclude a mortality benefit that outweighs the harms.

Screening programs incorporating FOBT, sigmoidoscopy, or colonoscopy will all be effective in reducing mortality. Modeling evidence suggests that population screening programs between 50 and 75 years of age using any of the following three regimens will be approximately equally effective in life-years gained, assuming 100 percent adherence to the same regimen for that period⁸: (1) annual high-sensitivity FOBT, (2) sigmoidoscopy every five years combined with high-sensitivity FOBT every three years, and (3) screening colonoscopy at intervals of 10 years.

The strategies differ in the total number of colonoscopies that would be required to gain similar numbers of life-years. Annual high-sensitivity FOBT (sensitivity of at least 70 percent) that has a false-positive rate less than 10 percent (specificity greater than 90 percent) is estimated to require the fewest colonoscopies while achieving a gain in life-years similar to that seen with screening colonoscopy every 10 years. Currently available tests that meet both specifications include Hemoccult SENSA guaiac testing and fecal immuno-chemical tests with characteristics similar to those of the Magstream quantitative test.

Although use of an annual FOBT with a lower sensitivity has been demonstrated to reduce colorectal cancer mortality in randomized controlled trials, modeling suggests that the number of life-years gained will be greater with the strategies using higher-sensitivity tests.

For all screening modalities, the effectiveness decreases substantially as adherence to the regimen declines. At the individual level, adherence to a screening regimen will be more important in life-years gained than will the particular regimen selected. Current data are insufficient to predict adherence to any specific screening regimen at the population level.

CONSIDERATIONS FOR PRACTICE WHEN EVIDENCE IS INSUFFICIENT

CT Colonography

Potential preventable burden. A screening program that incorporates the option of CT colonography could help reduce colorectal cancer mortality in the population if patients who would otherwise refuse screening found it an acceptable alternative.

Potential harms. The potential harms from evaluation of incidental findings found with CT colonography may be large. The lifetime cumulative radiation risk from use of CT colonography to screen for colorectal cancer should be considered, as well as the growing cumulative radiation exposure from the use of other kinds of diagnostic and screening procedures that involve radiation exposure.

Current practice. CT colonography performed by trained and experienced radiographers may not be currently available in many parts of the United States.

Costs. Patient time and burden to participate in colorectal cancer screening using test strategies that require bowel preparation are substantial. A CT colonography screening strategy that did not involve bowel preparation would decrease the burden of adherence. The cost of CT colonography is high.

Fecal DNA

Potential preventable burden. Fecal DNA has potential as a highly specific test, and it could reduce harms associated with follow-up of false-positive test results.

Current practice. Fecal DNA tests are evolving, and no test is widely used.

Costs. Fecal DNA is likely to have a high cost per test.

OTHER APPROACHES TO PREVENTION

Dietary approaches, such as avoidance of red meat and alcohol or consumption of diets very high in fiber, have been suggested to protect against the risk of colorectal adenomas, but these claims are based on associations present in observational studies that have not been substantiated in trials. Certain nonsteroidal anti-inflammatory drugs (NSAIDs) are associated with regression and decreased incidence of colonic adenomas, but the harms of daily NSAID use in asymptomatic persons led the USPSTF to recommend against this use in persons not at increased risk.

USEFUL RESOURCES

In 2007, the USPSTF recommended against the use of aspirin or NSAIDs for prevention of colorectal cancer (D recommendation, available at http://www.ahrq.gov/clinic/uspstf/uspsasco.htm).