Colorectal Cancer Screening and Surveillance

Who Should Be Screened? Determining Risk

The age to begin screening, the modality chosen for screening, and the interval between examinations depends on the presence or absence of known risk factors for CRC. Persons with no risk factors are considered to be at average risk of CRC with a cumulative lifetime risk of approximately 5% in both men and women. Patients at higher than average risk of CRC include patients with a personal or family history of CRC or polyps and patients with Crohn disease or ulcerative colitis. Patients with a genetic predisposition to CRC are at the highest lifetime risk of CRC. Hereditary CRC syndromes account for 5% of all CRC cases and warrant the most aggressive screening and surveillance.

The USMSTF, USPSTF, and ACG recommend that screening for CRC begin at age 50 years in average-risk, asymptomatic men and women. The ACG further suggests that African Americans start screening at age 45 years because of the increased incidence of CRC before age 50 years in this population.

Screening Average-Risk Persons

CRC screening tests can be divided into cancer prevention tests with the potential to identify both cancer and polyps and cancer detection tests that are primarily effective in diagnosing CRC ( Table 55-1 ).

Colonoscopy every 10 years is the preferred CRC prevention test, and annual fecal occult blood testing with high-sensitivity guaiac-based testing or the fecal immunochemical test (FIT) is the preferred cancer detection test. Table 55-2 summarizes the current screening guidelines for average-risk individuals issued by the ACG, USPSTF, and USMSTF.

Colonoscopy

Colonoscopy is frequently cited as the gold standard CRC screening test because it has both diagnostic and therapeutic capabilities. Additionally, abnormalities detected on other CRC screening tests lead to a colonoscopy.

No randomized clinical trials of the effect of screening colonoscopy on CRC mortality have been performed, but epidemiologic and cohort studies have shown that exposure to colonoscopy is associated with up to a 75% decrease in CRC incidence, which includes a durable protective effect up to and greater than 10 years. This protective effect extends to patients who have no polyps detected on colonoscopy, in addition to those who have had a polyp removed. A recent follow-up on the National Polyp Study evaluating the long-term benefit of colonoscopy with polypectomy found a 53% reduction in CRC mortality when compared with the expected deaths from CRC in the general population. An inverse relationship between colonoscopy screening and death from CRC was found in a recent large Canadian population–based study. These investigators found that for every 1% increase in colonoscopy screening, a 3% decrease in CRC occurs.

The reduction in colorectal cancer mortality after colonoscopy varies by the site of the cancer, with less protection in the proximal than in the distal colon. In a Canadian case-control study that matched more than 10,000 patients who died with CRC to 5 control subjects each, colonoscopy was associated with lower mortality from left-sided colon cancer (adjusted odds ratio [OR], 0.33; 95% confidence interval [CI], 0.28 to 0.39) but not right-sided CRC (OR, 0.99; 95% CI, 0.86 to 1.14). It is suggested that the biologic differences between proximal and distal CRC (a higher proportion of CpG island methylation phenotype, DNA microsatellite instability, and BRAF mutations in proximal tumors and sessile serrated adenomas/polyps), the higher proportion of flat polyps in the proximal colon, and the technical skill of the endoscopist affect the miss rate of premalignant lesions and the development of interval CRC.

Quality indicators have been established to standardize colonoscopy and aid in determining areas for continuous quality improvement. The minimum quality standards include but are not limited to the cecal intubation rate (>90% for diagnostic cases and >95% for screening examinations), high quality of bowel preparation, colonoscopy withdrawal time (>6 minutes in screening examination), and adenoma detection rate (ADR; >25% in a mixed gender population aged ≥50 years or >30% for men and >20% for female patients older than 50 years). The ADR is defined as the proportion of screened subjects in whom at least one adenoma is detected. Endoscopists who meet these standards have decreased interval cancer rates. Compared with physicians who have an ADR greater than 20%, the hazard ratios for interval CRC based on ADRs of less than 11%, 11% to 14.9%, and 15% to 19.9%, were 10.94, 10.75, and 12.50, respectively ( P = .02 for all comparisons). In a recent study published, Corley and colleagues found a continued protection against CRC with an ADR above 32%. Each 1% increase in the ADR was associated with a 3 % decrease in the risk of CRC.

One major factor affecting the quality of mucosal inspection and subsequently ADR is the quality of bowel preparation. The American Society of Gastrointestinal Endoscopy/ACG Task Force recommends that the examination be considered adequate if it allows detection of polyps greater than 5 mm. Inadequate (poor or fair) bowel preparation is associated with an adenoma miss rate as high as 42% when a repeat colonoscopy is performed within 3 years of the initial suboptimal examination. The quality of bowel preparation worsens with a longer time interval between completion of the preparation and the start of colonoscopy. A meta-analysis of nine studies concluded that receiving at least half of the bowel preparation on the day of the examination, also known as “split-dose” bowel preparation, resulted in more excellent and good results (OR, 3.46; 95% CI, 2.45 to 4.89) compared with night-before dosing. Therefore, to maximize bowel preparation quality, it is recommended that the colonoscopy be started within 3 to 5 hours of the last dose of preparation.

Flexible Sigmoidoscopy

Three of four randomized controlled trials have shown that flexible sigmoidoscopy decreases CRC incidence up to 25%. Two of the three trials found up to a 30% reduction in mortality from CRC.

Flexible sigmoidoscopy is often performed without sedation and does not usually require an oral bowel preparation, which is one barrier to colonoscopy. However, the examination is limited to the left colon, and about 37% of lesions are beyond the reach of the flexible sigmoidoscope. Patients for whom precancerous lesions are detected with sigmoidoscopy should be referred for a colonoscopy.