Incidence and Mortality
In 2004, an estimated 146,940 new cases of and 56,730 deaths from colon and rectal cancers were expected.20 Colorectal cancer (CRC) is the third leading cause of new cancer cases (11% of all new cases) and cancer deaths (10% of all cancer deaths) in both men and women.20 In 2000, the age-adjusted incidence rate in nine Surveillance, Epidemiology, and End Results Program (SEER) registries was 55 per 100,000; the age-adjusted mortality rate was 21 per 100,000.21 The lifetime risk from birth of being diagnosed with CRC is about 6%; the lifetime risk of dying from CRC is about 2%. Thus, about 1 in 3 people who develop CRC die of this disease. Between 1992 and 2001, mortality from CRC declined by 1.8% per year93 and incidence declined by 0.8% annually in the United States.94 The early detection and removal of precancerous colorectal polyps may have contributed to the decline in CRC incidence and mortality.95
The major screening tests currently available for CRC screening are the fecal occult blood test, sigmoidoscopy, colonoscopy, and double-contrast barium enema. These tests are used to identify precancerous or cancerous lesions in the colon and rectum. No one screening strategy has been shown to be superior to the others, although they differ in regard to accuracy, effectiveness, and potential harms.
Fecal occult blood testing (FOBT) has been examined in three RCTs involving more than 250,000 people followed for up to 18 years.96,97 All three trials found a reduction in CRC mortality from 15% to 33%, with an absolute risk reduction for CRC deaths ranging from 0.8 per 1000 with biennial screening in the United Kingdom over 8 years of follow-up98 to 4.6 per 1000 with annual screening in Minnesota during 18 years of follow-up.99 The Minnesota study also noted a 17% to 20% decrease in incidence of CRC.99 The sensitivity of a single test is approximately 30% to 50%, with a specificity of 90% to 98%, depending on how the test is done. Fecal occult blood tests find about 25% to 50% of patients with colorectal cancer, but only 2% of patients with a positive test had cancer in the Minnesota trial.
The effectiveness of sigmoidoscopy to reduce CRC deaths has been examined in three well-designed case-control studies.100-102 These studies showed a mortality reduction of 60% to 80%.97 In a small, randomized trial of sigmoidoscopy, in which persons with polyps were followed up with colonoscopy, the incidence of colorectal cancer was decreased by 80% but no decrease in mortality was found.103 Using full examination of the colon as the "gold standard," sigmoi-doscopy has been found to identify 70% to 80% of patients with advanced adenomas or cancer.104,105 The sensitivity and specificity of sigmoidoscopy are difficult to determine, because all visible polyps are typically removed, many of which may have little to no malignant potential.
The ability of screening colonoscopy to reduce colorectal cancer morbidity or mortality has not been directly studied to date. Data from studies of other modalities have been extrapolated to support the effectiveness of colonoscopy. Because it is often used as the gold standard, determining its sensitivity and specificity has been difficult. A recent study by Pickhardt et al., comparing optical colonoscopy with CT virtual colonoscopy,106 in which 1,233 patients underwent both procedures, found the sensitivity of optical colonoscopy for adenomatous polyps to be 88% to 92%, depending on the size of the polyps. As with sigmoidoscopy, the natural history of many polyps found on colonoscopic examination is not known; thus, the potential for identifying false positives must be considered.
No screening studies of double-contrast barium enema with a mortality outcome have been published; thus, the accuracy and effectiveness of this procedure are unknown.96 Its sensitivity is likely lower than that of endoscopic procedures, but if it misses primarily polyps that are small and not likely to progress to invasive cancer, its effectiveness for screening may be adequate.
A variety of different types of polyps occur in the colon and rectum. Hyperplastic polyps are the most common of those that have little potential for becoming malignant. They cannot be distinguished visually from adenomatous polyps, so biopsy is required for diagnosis. Whether the presence of distal hyperplastic polyps increases the risk of proximal neoplastic polyps is uncertain.107 A systematic review of 18 studies108 estimated a 21% to 25% risk for any proximal neoplasia in patients with a distal hyperplastic polyp, includ ing a 4% to 5% risk of an advanced neoplasm (cancer or polyp with severe dysplasia or villous histology). In 4 of the studies in which colonoscopy was performed regardless of distal findings, however, the relative risk of finding any proximal neoplasia was 1.3 (95% confidence interval, 0.9-1.8).
Two-thirds of all colonic polyps are adenomatous, which are defined as dysplastic and thus have malignant potential. Most colorectal cancers arise from adenomatous polyps. Some proportion of these grow from small (less than 5 mm) to large (greater than 1.0 cm) to cancer, generally over a period of 10 years or longer. The proportion that makes this transition is thought to be small; adenomatous polyps occur in 30% to 40% of adults over the age of 50, but the risk of developing colorectal cancer is only about 6%.107 However, removal of adenomatous polyps is associated with a reduced risk of colorectal cancer incidence and mortality.
The harms of screening for colorectal cancer include the risk of the screening tests themselves, the risks of the subsequent workup from positive screening tests, the potential for false-negative screening results, and the potential for overdiagno-sis and treatment of lesions that would not have become malignant over the person's lifetime. No direct adverse effects of FOBT exist (other than the inconvenience and some patients' distaste for performing the test). Both sigmoidoscopy and colonoscopy are associated with low risks for major complications, including bleeding and perforation of the colon during the examination. A large population-based study of Medicare beneficiaries aged 65 and older found perforation rates of nearly 1 per 1,000 for sigmoidoscopy and 2 per 1,000 for colonoscopy.109 The risk of death following colonic perforation was 52 to 65 per 1,000 perforations.
Fecal occult blood tests may miss small adenomas, as these lesions frequently do not bleed. Whether that represents a true negative or a false negative is uncertain, as these small adenomas may not be likely to develop into neoplastic lesions. Even though some consider colonoscopy to be the optimal examination of the colon and rectum for detection of precancerous and cancerous lesions, studies have shown that significant lesions (i.e., those larger than 1 cm) may be missed. The Pickhardt study comparing virtual with optical colonoscopy found that virtual colonoscopy missed 5 of 59 advanced neoplasms (defined as adenomatous polyps 10 mm or more in diameter or demonstrating high-grade dysplasia, villous changes, or cancer) and optical colonoscopy missed 7 of the 59 lesions.106
The risk of overdiagnosis and treatment of lesions that do not have long-term malignant potential (false-positive lesions) is more difficult to quantify. Most adenomas (60%-75%) are smaller than 1cm on endoscopic examina-tion.107 The risk for high-grade dysplasia increases from 1% in small adenomas (less than 5 mm) to 6% for medium-sized adenomas (5-10 mm) to 21% for large adenomas (greater than 1cm).107
In a recent study of a screening colonoscopy program at a work site,110 the authors created a clinical index to stratify risk for advanced proximal neoplasia (defined as an adenoma 1 cm or larger or one with villous histology, severe dysplasia, or cancer) and to identify a subgroup at low risk for whom screening sigmoidoscopy alone might be sufficient. Scores were based on age, sex, and distal findings. In the validation arm of the study, the 47% of the cohort determined to be in the low-risk subgroup had a risk for advanced proximal neo-plasia of 0.4%. Use of the index in this population identified 92% of persons with advanced proximal neoplasia. The number needed to screen (NNS) to detect advanced proximal neoplasia among patients with any distal polyp was 16 and, among everyone, the NNS was 36. The NNS to extend one life from colorectal cancer mortality was not calculated and would be higher.
Colorectal cancer screening reduces death from colorectal cancer and decreases the incidence of invasive cancer by finding and removing adenomatous polyps. These benefits of screening, however, are tempered somewhat by the effort involved, the harms of the screening procedures themselves, and the possibility of overdiagnosis and overtreatment of small lesions with low malignant potential. As tests with greater sensitivity are developed, the risk of overdiagnosis increases.
Was this article helpful?
Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.