Chapter 22. Colorectal Cancer Screening

Screening for colorectal cancer (CRC) can reduce disease-related morbidity and mortality. The existing evidence has led to the recommendation of CRC screening as a standard component of preventive health care.

General Considerations

CRC is the second leading cause of cancer death in the United States and Western Europe. The lifetime risk of CRC in the United States is approximately 6%. In 2010, it was estimated that 72,090 men and 70,480 women would be diagnosed and 51,370 individuals would die of CRC. However, overall disease survival has improved from 51.4% in the 1970s to 64.9% in early 2000. Increased knowledge about the pathogenesis, advances in medical and surgical care, and the increasing emphasis on CRC screening programs have contributed to these substantial gains.

CRC can be prevented through readily available screening. Prevention efforts rely on the long time interval required for a benign adenomatous polyp to progress into an invasive cancer. It is estimated that the adenoma to carcinoma sequence unfolds over a 7- to 10-year period. In addition, CRC-related deaths are preventable if the disease is detected early. When diagnosed early, the 5-year survival rate for CRC that is still confined to the primary site (localized stage) is approximately 90%. Conversely, the corresponding 5-year survival for patients with known distant metastases is only 10%. Unfortunately, only 39% of cancers are diagnosed at an early stage and up to 19% of patients have distant metastases.

Nevertheless, CRC screening is presently underutilized in the United States. Approximately half of the population is currently compliant with standard screening recommendations, despite the array of available choices. CRC screening lags far behind screening for other common malignancies, such as breast, cervical, and prostate cancer. According to recent results by the National Center for Health Statistics, only 54% of Americans aged 50 years or older had undergone any type of CRC screening and only 42% had undergone screening within the recommended time interval. At present, issues of sensitivity, specificity, and patient acceptance limit existing CRC screening methods. Several factors contribute to the lack of compliance with CRC screening, including inappropriate perception of risk (particularly if patients are asymptomatic and without a family history of CRC), dietary restrictions or burdensome cathartic preparations, the invasiveness of procedures, and perceived discomfort, pain, and embarrassment related to certain screening techniques.

Pathogenesis

In 1990, a stepwise, chronologic model for colorectal tumorigenesis was proposed. It outlined sequential alterations in key growth regulatory genes, such as APC, K-ras, and tp53, culminating in the development of a malignant neoplasm. This pathway contributes to the development of 85% of colorectal tumors that arise from preexisting adenomatous polyps. When adenomas accumulate the necessary combination of genetic mutations in the suggested stepwise manner, the end result is cancer (Figure 22–1).

The genes implicated in CRC can be divided into three categories: (1) tumor-suppressor genes, (2) proto-oncogenes, and (3) DNA repair genes.

The function of tumor-suppressor genes is to downregulate normal growth stimulatory pathways. In CRC, the genes most frequently inactivated are APC, tp53, and p16. Consistent with the Knudson "two-hit hypothesis," acquired (or somatic) mutations in both alleles of the tumor-suppressor gene are required to fully inactivate gene function and cause cancer. In autosomal-dominant CRC syndromes, a preexisting germline mutation in one allele ("first hit") is inherited and a "second hit" occurs when an acquired mutation inactivates the other allele.

Proto-oncogenes, such as K-ras, are components of signaling pathways that promote normal cellular growth and proliferation. A mutation in a proto-oncogene leads to an active gene product with a resulting tumorigenic effect.

DNA repair genes function to maintain the integrity of the genome. Individual nucleotides can be modified by biochemical reactions such as oxidation, alkylation, spontaneous deamination, and ultraviolet cross-linking. When errors occur, they are corrected through a sophisticated process known as "base excision repair." Another manner in which errors are introduced into the genome is through mispairing of nucleotides, which can occur during normal DNA replication. These errors are corrected by a DNA "mismatch repair" system. When either of these DNA repair processes is dysfunctional, deleterious mutations can accumulate in genes that directly control cellular growth and proliferation.

The approach to CRC screening in asymptomatic individuals depends on risk stratification and the likelihood of developing the disease. Screening for CRC takes place in persons who have no personal or family history of adenomatous polyps or cancer. Surveillance takes place in persons with a personal history of either adenomatous polyps or CRC, or one of the genetic syndromes that requires more intensive monitoring. Individuals who have signs or symptoms suggestive of CRC fall outside the domain of screening and should be offered appropriate diagnostic evaluation.

CRC screening programs begin by classifying an individual's level of risk based on age, as well as personal and family medical history (Figure 22–2). This vital information helps determine when screening should be initiated, with what appropriate tests, and the frequency of subsequent examinations. CRC risk is commonly stratified into three broad categories: average risk, moderate risk, and high risk.

Average Risk

An average-risk individual is defined as an asymptomatic person without a personal or family history of colonic polyps or cancer. The majority of the general population is considered to be at average risk for CRC. Given that age is the strongest risk factor for the development of CRC and adenomatous polyps, men and women at average risk should be offered screening beginning at age 50 years.

Available screening options for average-risk individuals fall into two broad categories: (1) tests that primarily detect CRC, and (2) those that detect precancerous adenomatous polyps and CRC (Table 22–1). Tests that primarily detect cancer are fecal tests, including guaiac-based and immunochemical occult blood tests, and fecal DNA tests. These modalities provide limited opportunity for prevention because detection of premalignant adenomatous polyps is most often incidental. Tests that detect both adenomatous polyps and cancer include flexible sigmoidoscopy (FS), colonoscopy, double-contrast barium enema (DCBE), and computed tomography colonography (CTC). Because prevention is the primary goal of CRC screening, recent guidelines from the American Cancer Society, the U.S. Multisociety Task Force on Colorectal Cancer, and the American College of Radiology strongly encourage clinicians to offer those tests that are designed to detect both early cancer and adenomatous polyps if resources are available and if patients are willing to undergo an invasive examination. Noninvasive tests must be repeated at regular intervals to be effective, are less likely to prevent CRC than invasive examinations, and, if abnormal, require colonoscopy.

Nevertheless, health care providers are encouraged to focus on increasing screening rates through periodic use of any of the recommended modalities. Providers should present their patients with information about the advantages and disadvantages associated with the multiple available screening tests. In turn, patients have the opportunity to select how they wish to be screened based on their own preferences. The rationale for such an approach is that it may increase the likelihood that screening will occur. Even though the currently available screening techniques are not equal in effectiveness, cost, or associated risks, they all have been demonstrated to be cost-effective compared with no screening at all.

Moderate Risk

Individuals with a personal or family history of adenomatous polyps or CRC are considered to be at intermediate risk for the development of CRC. For these patients, colonoscopy is the recommended screening method. Other conventional screening modalities are not typically recommended in this setting.

Personal History of Adenomatous Polyps

Certain characteristics of colorectal adenomas at baseline colonoscopy are the basis for decisions about surveillance intervals. Existing data from the National Polyp Study (NPS), a pooled analysis of chemoprevention studies, as well as observational cohort studies support the presence of the following predictors for the development of future advanced adenomas or cancers: (1) three or more adenomas, (2) adenoma size greater than 1 cm, and (3) adenoma with villous features or high-grade dysplasia. A proximal location of adenomas (ascending colon or cecum) may also predict metachronous advanced adenomas, but this has not been well studied to date. In addition, there is a general consensus among many of the studies that individuals with adenomas with lesser findings (eg, one or two subcentimeter adenomas without high-grade dysplasia or villous features) are at lower risk for subsequent advanced adenomas.

By stratifying patients into "lower risk" and "higher risk" groups for future advanced adenoma development based on findings from initial colonoscopy, the NPS has provided evidence-based guidelines for surveillance of postpolypectomy patients. In a randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps, there was no better detection of advanced lesions with surveillance examination 1 year after the initial colonoscopy than with follow-up examination in 3 years. The inference made is that the rate of developing metachronous adenomas with advanced pathology is slow; therefore, the current recommendation for patients with a personal history of three or more adenomas, adenomas of 1 cm or larger, or adenomas with villous or high-grade dysplasia, or any combination of these findings, is surveillance colonoscopy in 3 years. In addition, patients who have more than 10 adenomas removed during one endoscopic examination should be examined in a shorter time interval (<3 years) based on clinical judgment and should be considered for the possibility of an underlying familial syndrome. The standard of care for a sessile, malignant polyp removed by piecemeal fashion is 3–6 months after the initial endoscopic resection.

However, it is important to acknowledge that the timing of surveillance colonoscopy in patients with a history of adenomatous polyps is still evolving. The existing studies assessing subsequent risk for neoplasia after colonoscopic polypectomy have only followed patients for 5–6 years. This is a significant limitation and has led to the current projection that patients without advanced adenomas can wait at least 5 years (and perhaps up to 10 years) for repeat colonoscopy. As a result, further studies are needed to evaluate this "low risk" group in order to confirm the strategy of these intervals.

Personal History of CRC

Individuals with a history of CRC are at risk for recurrent cancer and metachronous neoplasms and require endoscopic surveillance after surgical resection. Anywhere from 2% to 7% of patients with CRC have one or more synchronous cancers in the colon or rectum at the time of initial diagnosis; it is therefore important to perform a complete colonoscopy in the preoperative period. In cases where an obstructing colonic or rectal lesion is detected, CTC or DBCE should be considered perioperatively, and a complete colonoscopy should be performed 3–6 months after surgery. Once the colon is cleared of any synchronous lesions, a postoperative surveillance colonoscopy is recommended at 1 year to evaluate for any metachronous lesions. If the examination at 1 year is normal, the subsequent examination should be at 3 years. If this examination is also found to be normal, surveillance colonoscopy can thereafter be extended to every 5 years. Neither individual randomized controlled clinical trials of intense surveillance with annual colonoscopy nor a meta-analysis of these trials for the purpose of detecting recurrent disease have shown a survival benefit for patients with CRC.

Colonoscopy at 1 year after surgical resection is recommended based on reports of a high incidence of metachronous second cancers noted within the first 2 years after resection. Among an aggregate of studies reporting results from postcancer surveillance colonoscopy, there was an incidence rate of 0.7% metachronous cancers in the first 2 years after resection of the initial primary cancer. This estimate is consistent with data from a tumor registry review in Nebraska, which calculated an annual incidence of 0.35% per year for metachronous cancers. This is considered sufficient information to warrant a colonoscopy at 1 year following surgical resection. However, this should not diminish the importance of a high-quality colonoscopic examination in the perioperative period to exclude synchronous neoplasms.

In addition, it is important to distinguish between rectal and colon cancer because of the differing rates of local recurrence. The recurrence of colon cancer at the anastomotic site occurs in only 2–4% of patients. In contrast, local recurrence rates of rectal cancer when patients have undergone a low anterior resection can be 10 or more times higher. High recurrence rates of rectal cancer are partly a function of surgical technique. Local recurrence rates of cancer can be reduced by using a surgical technique called mesorectal excision, as well as administering radiation and chemotherapy in the neoadjuvant, preoperative setting to patients with locally advanced disease. However, because reported local recurrence rates for rectal cancer across the United States are generally higher than those achieved in case series using total mesorectal excision, there is a rationale for performing periodic examinations of the rectum. Although effectiveness has not been proven, performing proctoscopy at 6-month intervals for the first 2 years after surgical resection can be considered for the detection of a surgically curable recurrence of the original rectal cancer.

When colon or rectal cancer is endoscopically resected and surgery is not needed, follow-up endoscopic examination to inspect the biopsy site within 1 year is a reasonable strategy. As previously noted, colonoscopy is considered the test of choice for the detection of metachronous neoplasms in patients with a history of CRC. CTC has not been evaluated adequately in this setting, and guaiac-based fecal occult blood testing (FOBT) has been considered to have a very low positive predictive value after colonoscopic evaluation.

Family History of CRC or Adenomatous Polyps

An individual's risk of CRC is increased if there is a family history of adenomatous polyps or CRC. The screening recommendations based on familial risk are derived largely from the observed colon cancer risk in relatives of patients with CRC and adenomas diagnosed before age 60 years (Table 22–2). Data from a meta-analysis of 27 studies assessing familial risk of CRC and adenomatous polyps report a relative risk of CRC when a first-degree relative was affected with CRC to be 2.4. The relative risk for CRC if the first-degree relative had an adenomatous polyp was 1.9, with age effects similar to those observed for cancer. In addition, if more than one relative was affected and CRC was diagnosed before age 45 years, the relative risk increased to 4.2 and diminished to 2.2 and 1.8 for ages 45–59 years and older than 59 years, respectively. Therefore, familial risk needs to be readily identified and should prompt the early initiation of screening with colonoscopy.

At present, the U.S. Multisociety Task Force on Colorectal Cancer recommends that for patients who report one first-degree relative under 60 years or two first-degree relatives of any age with adenomatous polyps or CRC, screening colonoscopy should start at age 40 years or 10 years younger than the earliest diagnosis of an affected relative (whichever one comes first) and continue every 5 years (Figure 22–2, Table 22–3). Screening should start at the same time for those patients with a single first-degree relative with CRC or adenomas diagnosed after age 60 years, but surveillance should be performed as for average-risk individuals. The rationale for starting screening at age 40 years is that the incidence of CRC in these patients resembles the risk in persons with no family history but precedes it by approximately 10 years. In addition, a patient reporting a second- or third-degree relative with adenomatous polyps or CRC does not confer additional risk, and therefore, average-risk screening recommendations are sufficient.

However, it is important to note that no studies to date have reported a reduction in mortality in persons with a family history of CRC or adenomatous polyps who undergo screening. Therefore, the present screening recommendations are considered provisional, and further evidence is needed to better delineate the CRC risk in relatives with adenomatous polyps or CRC as this information continues to evolve.

High Risk

For individuals with a hereditary CRC syndrome, such as familial adenomatous polyposis (FAP) or Lynch syndrome (also known as hereditary nonpolyposis colorectal cancer, or HNPCC), genetic counseling and special screening protocols are recommended (Table 22–4). Readily identifying these individuals is of great importance given the known benefit of intensive endoscopic surveillance and prophylactic surgery on morbidity and mortality. Recognizing these syndromes also has an impact on referral for predictive genetic testing, wherein identifying gene carriers improves the efficiency of cancer surveillance and helps identify family members who require intense management versus those who can receive the standard of care. Additionally, for patients with longstanding inflammatory bowel disease such as Crohn disease or ulcerative colitis, surveillance colonoscopy with systematic biopsies should be performed because the risk of CRC is increased in both conditions.

Familial Adenomatous Polyposis

FAP is an autosomal-dominant syndrome that is associated with mutations in the adenomatous polyposis coli (APC) gene. Affected individuals classically develop hundreds to thousands of colorectal adenomas at a young age and have a risk of CRC approaching 100% in the absence of prophylactic colectomy. Annual sigmoidoscopy beginning at age 10–12 years is recommended for persons who have a genetic diagnosis or are at risk of having FAP in order to determine if they are expressing the genetic abnormality. Germline genetic testing should be recommended in persons with an FAP phenotype and should be performed after patients (or parents of children) undergo genetic counseling. If a mutation is identified, other family members can be tested to discern the presence or absence of the same mutation with nearly 100% accuracy. Sigmoidoscopy is performed until affected individuals develop polyps, at which point total colectomy is recommended. Family members who test negative for the gene mutation are considered to be at average risk for CRC.

Although classic FAP can be readily identified, a subset of patients has a less obvious phenotype. Patients with 10 or more cumulative colorectal adenomas but less than 100, are at risk for a variant of FAP called attenuated FAP (AFAP). In these patients, the age of onset of adenomas is approximately 10 years later than with classic FAP. AFAP also has an autosomal-dominant mode of inheritance and, as with classic FAP, is due to mutations in the APC gene. The true incidence and frequency of AFAP is unknown; however, it may account for up to 10% of adenomatous polyposis families and should be considered in individuals with multiple adenomas. Colonoscopy should be used as the initial screening modality in patients suspected of having AFAP, and the age to initiate screening is based on the age of polyp expression or CRC diagnosis in the family. It is reasonable to start colonoscopic screening 10 years earlier than the earliest age of known polyp or CRC diagnosis in the family. As with FAP, germline genetic testing should be considered in persons presenting with the AFAP phenotype.

Lynch Syndrome

Lynch syndrome exhibits an autosomal-dominant pattern of inheritance, and affected individuals have a near 80% lifetime risk of developing CRC without intervention. It is predominantly caused by a mutation in one of four of the DNA mismatch repair genes, MSH2, MLH1, MSH6, and PMS2. There are several classification systems for the clinical diagnosis of Lynch syndrome, ranging from the most exclusive Amsterdam Classification to the most inclusive revised Bethesda Guidelines, all relying on personal or family medical histories, or both. Commercial genetic testing is currently available for individuals suspected of having Lynch syndrome, as well as for family members who are at risk for the condition.

Once Lynch syndrome has been diagnosed based on clinical criteria or through genetic testing, an intensive screening program should be instituted. Screening colonoscopy should be recommended every 1–2 years beginning at age 20 and 25 years, or 10 years prior to the youngest CRC case diagnosed within the family (whichever occurs earlier). Annual colonoscopy is recommended after the age of 40 years.

Inflammatory Bowel Disease

The CRC risk is similar in both Crohn colitis and ulcerative colitis, with an estimated lifetime risk of 15–40%. Although direct supporting evidence is lacking, an expert consensus panel has recommended colonoscopic surveillance every 1–2 years for patients after 8 years of pancolitis or after 15 years in those with limited left-sided colitis. If the extent of disease cannot be accurately assessed, patients should undergo surveillance colonoscopy beginning within 8–10 years of disease.

To consider surveillance effective, an extensive biopsy protocol during colonoscopy should be followed. Experts have recommended that biopsy specimens should be taken every 10 cm in all four quadrants and that additional biopsy specimens should be obtained from anastomotic sites (in cases of previous colonic resections), strictures, and mass lesions (other than pseudopolyps). Polyps should be removed by polypectomy, with biopsies of the adjacent flat mucosa to discern if dysplasia is present. In cases where high-grade dysplasia or multifocal dysplasia is found in flat mucosa, patients should be advised to undergo colectomy. However, it is imperative that pathologic confirmation be made by an experienced pathologist in the field of dysplasia in inflammatory bowel disease prior to colectomy. Controversy exists in cases of low-grade dysplasia in flat mucosa and the need for colectomy. If low-grade dysplasia is unifocal or found in areas of inflammation, attempts should first be made to treat the active colitis.

In cases where a mass lesion is detected, it is important to differentiate whether it is a dysplasia-associated lesion or mass (DALM) or if it has arisen from a sporadic adenoma. A DALM is a dysplastic lesion believed to have arisen because of the cancer potential of the colitis and is an indication for colectomy. In contrast, a lesion arising from a sporadic adenoma should be removed in its entirety during colonoscopy and does not require colectomy unless the lesion is not amenable to complete endoscopic resection. Differentiating between DALMs and adenoma-related masses can be difficult and requires extensive review of the endoscopic appearance, histologic findings, presence of dysplasia in the surrounding flat mucosa, patient's age, and duration of disease.

The decision for colectomy should be individualized in cases of long-standing disease based on a number of concomitant factors known to increase CRC risk. These factors include extensive active colitis or ongoing colitis-related symptoms, personal history of primary sclerosing cholangitis, a family history of CRC, age, and life expectancy.

Special Considerations: Hyperplastic & Serrated Polyps

At this time, there is no evidence to suggest that patients with a few small, distally located hyperplastic polyps are at increased risk for CRC. Therefore, such patients should undergo surveillance for CRC similar to that of average-risk individuals.

However, there has been recent literature to suggest that all hyperplastic polyps are not histologically similar and that some variants of hyperplastic polyps may evolve into a unique type of adenoma, called a serrated adenoma. Serrated adenomas resemble hyperplastic polyps with dysplasia and have been linked to the development of sporadic adenocarcinoma with microsatellite instability. Polyps of this type are often large, sessile, and located in the proximal colon. It is important for these neoplastic lesions to be completely removed, and surveillance should follow those recommendations as for typical adenomas.

In addition, there is a recently described syndrome of "hyperplastic polyposis." The following features have been used to define this entity: (1) at least five histologically diagnosed hyperplastic polyps proximal to the sigmoid colon, of which two are greater than 1 cm in diameter, or (2) any number of hyperplastic polyps occurring proximal to the sigmoid colon in an individual who has a first-degree relative with hyperplastic polyposis, or (3) more than 30 hyperplastic polyps of any size distributed throughout the colon. A limited number of studies have suggested an increased risk of CRC in such individuals, although the magnitude of risk has not been determined. A case series of 15 patients fulfilling the preceding criteria did not find any CRC cases within 3 years of follow-up colonoscopic evaluation. Presently, the optimal management for these patients is unclear and requires further study.

Fecal Occult Blood Testing

Fecal occult blood testing (FOBT) is the most widely available screening modality for CRC. It is a noninvasive and inexpensive test used to detect the presence of occult blood in the stool of asymptomatic persons. FOBT is recommended as an annual examination and is the only screening test for which randomized controlled trials have shown a reduction in CRC-related incidence by 33%, and mortality benefits ranging from 15% to 20%. This is due primarily to the detection of CRC at an earlier stage or the detection of adenomatous polyps with the ability to remove these lesions during colonoscopy. It is important to emphasize that positive FOBT results need to be appropriately followed by complete visualization of the colon through colonoscopy.

Two types of FOBT exist: the standard guaiac test and the newer immunochemical tests (see following section). Guaiac-based FOBT is widely used for CRC screening worldwide. It is performed by the patient at home and requires the collection of six stool samples from three consecutive stools. The guaiac test detects blood by reacting with peroxidase activity of the heme portion of hemoglobin. As a result, false-positive tests may result from the ingestion of red meat or peroxidase-containing foods, and patients must be made aware of the necessary dietary restrictions. In addition, rehydration results in an increase in the false-positive rate, and should not be performed.

When FOBT is used alone as a screening strategy, it is only 33–50% sensitive in one-time testing, with some improvement if testing is repeated every 1–2 years. The specificity of FOBT as a screening modality is high (estimated to be >95%). False-positive results can stem from other causes of gastrointestinal blood loss or other substances in the stool that might cause a positive guaiac reaction. Patients are instructed to avoid ingestion of vitamin C, iron, aspirin, and nonsteroidal anti-inflammatory medications for 3 days prior to performing FOBT as these substances may interfere with the accuracy of the test. False-negative results can also occur because colorectal lesions can bleed intermittently and blood may not always be present throughout the entire stool. Therefore, a single test of a stool sample during digital rectal examination is not an adequate substitute for FOBT. There is an estimated fivefold decrease in sensitivity with a single test performed during digital rectal examination, and a higher yield has been noted when sampling three consecutive stools for occult blood.

The disadvantages related to FOBT are that current tests fail to detect many polyps and some cancers. Also, individuals who have a false-positive result will subsequently need to undergo the discomfort, cost, and risk of colonoscopy without benefit. In addition, there is reported evidence of low referral rates for colonoscopy following an abnormal FOBT. In the United States, one survey found that up to 30% of patients with positive FOBT did not receive a recommendation for colonoscopy.

In evaluating an FOBT screening program, it is also important to measure programmatic adherence and performance over time. FOBT is ineffective if patients have one test but do not return for repeat testing or do not have colonoscopy for positive tests. When FOBT is recommended, patients must be informed that the test needs to be repeated every 1–2 years if negative and that colonoscopy should be performed if the test is positive.

Fecal Immunochemical Testing

Newer fecal immunochemical testing (FIT) reacts with antibodies that are specific for the globin portion of the human hemoglobin molecule, and therefore, special dietary restrictions to avoid false-positive tests are not needed. Multiple studies have reported that FIT has a better sensitivity and specificity than guaiac-based FOBT for the detection of CRC, in which the sensitivity ranged from 47.1% to 69% and the specificity ranged from 88.2% to 97.1%; in contrast, the sensitivity of the guaiac test for detecting CRC has been reported to be as low as 37.1%, with a specificity of 86.7%. In addition, the effectiveness of FIT for CRC screening has been shown using "gold standard" endoscopy results for FIT-negative patients. However, given the wide variability in detecting advanced adenomas, which ranges from 25% to 72% based on the type of immunochemical test used, FIT should be best regarded as an early cancer detection test.

There are a number of unresolved issues regarding FIT that require further study, including the number of samples needed for optimal sensitivity and specificity, the cutoff point for hemoglobin detection that provides the best sensitivity and specificity, how to ensure hemoglobin stability in collected samples, and whether there is an advantage of a quantitative FIT over a qualitative test.

Flexible Sigmoidoscopy

FS allows for direct visualization of the distal third of the colon, ideally from the rectum to the splenic flexure. Several retrospective, case-controlled studies have suggested that FS is associated with a near 70% reduction in cancer mortality resulting from tumors found within reach of the endoscope. A recent multicenter, randomized controlled trial conducted in the United Kingdom found that in an 11-year follow-up of individuals undergoing a single FS for screening initiated between the ages of 55 and 64 years, FS substantially reduced colorectal cancer incidence and mortality by 33% and 43%, respectively. The reduction noted in this study was less compared to results from prior case-control studies and is attributed to the dominance of screen-detected prevalent cancers within the first 4 years of follow-up. The magnitude of reduction in cumulative incidence will likely increase as this cohort is followed in order to examine the important issue of the long-term effects of one screening examination using FS.

FS is recommended every 5 years for average-risk individuals. A 5-year interval is supported by a number of case-controlled trials and cohort studies reporting a mortality benefit from the use of FS up to 10 years from the last screening examination. In a recent study comparing the appropriateness of a 5-year interval with a 3-year interval, there was no difference in the detection of advanced neoplasia (0.9% vs 1.1% at 3- and 5-year intervals, respectively). In addition, multiple studies have reported a low frequency of low-risk and advanced lesions, and there is some speculation that an even longer interval than 5 years between sigmoidoscopic examinations may be reasonable.

To improve the use of FS as a screening modality, certain clinical factors as well as endoscopic findings noted on FS have been considered important predictors of proximal colonic disease. These predictors include the presence of multiple adenomas, any adenoma with villous histology or dysplasia, age older than 65 years, male gender, and a family history of CRC. Hyperplastic polyps detected on FS are not associated with increased adenoma detection elsewhere in the colon.

The advantages cited for FS are that it is inexpensive, requires no sedation, involves a simple bowel preparation, and can be performed by primary care physicians and nonphysicians, including physician assistants and nurse practitioners. In addition, the risk of perforation is lower than that associated with colonoscopy. However, the interval for surveillance is shorter with FS than with colonoscopy because FS is a less sensitive examination. This reduced sensitivity is partly due to variable operator experience as a wider variation in adenoma detection rates among different providers has been reported. In addition, the preparation used for colon cleansing prior to FS may be less than optimal, and discomfort in a nonsedated patient may interfere with the completeness of the examination.

Currently, any adenomatous polyp found on FS warrants complete colonic visualization by colonoscopy to exclude the presence of proximal lesions. Such a strategy has been associated with an increase in the adenoma detection rate to approximately 70%. However, these findings highlight a major drawback of FS in that it can miss anywhere from 30% to 65% of advanced neoplasia located proximally. This may be a particular concern for the use of FS in older patients and women, in whom higher rates of proximal neoplasia have been reported. Recent data comparing the performance of FS to colonoscopy in the detection of advanced neoplasia reported that in subjects between 55 and 59 years of age, the incremental benefit of colonoscopy over FS was very small and statistically insignificant. In contrast, for patients 60 years of age or older, the detection rate for colonoscopy was significantly higher than for FS (odds ratio [OR], 2.00; 95% confidence interval, 1.30–3.09). When considered with prior studies, these results suggest that FS could achieve high detection of advanced neoplasia in patients younger than 60 years and is less effective with those older than 60 years. Future studies to determine a combined approach for CRC screening using FS for persons in the sixth decade followed by colonoscopy once subjects are older than 60 years may provide valuable information that may, in turn, conserve colonoscopy resources.

A second shortcoming of FS is that patient compliance with this screening modality is poor. Only 15–30% of eligible persons regularly undergo FS, and this invariably limits its utility in CRC prevention. Lastly, low rates of referral for colonoscopy and colonoscopy completion have been reported, and better measures to assure appropriate follow-up are warranted.

Combined Fecal Occult Blood Test & Flexible Sigmoidoscopy

This combined approach couples annual FOBT with sigmoidoscopy every 5 years. The occult blood test is performed prior to sigmoidoscopy given that a positive result would warrant colonoscopy. A single nonrandomized study reported a 43% reduction in CRC-related deaths by this combination approach compared with sigmoidoscopy alone. In addition, this approach can increase the yield of adenoma detection fourfold compared with FOBT alone. However, more recent studies have not shown such promising results. The addition of FOBT to FS was estimated to increase the adenoma detection rate only from 70% to 75%. Additionally, undergoing a single FS and FOBT was not as effective as biennial FOBT in detecting CRC over a 16-year period. The lack of improvement in detection noted in this long-term case-control study may have been due to lack of compliance associated with undergoing FS.

Double-Contrast Barium Enema

DCBE has not been studied as a screening test for CRC in prospective randomized trials, and several recent studies have highlighted some major limitations associated with the examination. In a number of comparative studies of colonoscopy and DCBE in screening and surveillance populations, DCBE detected only 45–48% of polyps greater than 1 cm. In a large prospective study, the sensitivity for detecting CRC was 83% for DCBE versus 95% for colonoscopy. Overall, DCBE is less sensitive than colonoscopy for the detection of large lesions, and this examination has fallen out of favor among certain endorsing societies as a primary screening strategy in average-risk persons.

However, given the number of deterrents to the widespread use of colonoscopy, DBCE is considered an alternative CRC screening modality to be performed every 5 years. It is a radiologic examination of the colon that requires a cathartic preparation similar to that given prior to colonoscopy and requires the instillation of barium and air to define the contour of the colon. It is included as a screening option because it can evaluate the entire colon, is generally widely available, and has a low risk of serious adverse events. However, this procedure does not allow for polyp removal or biopsy of cancer if detected. Artifact, such as stool appearing as polyps, is common and causes many false-positive tests. All patients with an abnormal barium examination need to undergo colonoscopy.

Although DCBE is an acceptable option for CRC screening in average-risk adults, its utilization has declined over the past few years. This has had an impact on radiology training programs as fewer trainees are able to acquire the necessary skills needed to perform the labor-intensive procedure properly due to the low volume of DBCE studies being requested. This trend is expected to continue as increased efforts are being made to provide training related to computed tomography colonography (CTC) among radiologists.

Colonoscopy

Colonoscopy is considered by many to be the gold standard for CRC screening. Despite the lack of evidence from large-scale randomized trials of screening colonoscopy, the procedure has been shown in cohort studies to decrease the expected incidence of CRC through the removal of adenomatous polyps. The NPS demonstrated that the removal of adenomas led to the decreased incidence of CRC and that colonoscopy was the only single test that could provide screening, diagnosis, and treatment. In the NPS, a group of patients who underwent colonoscopy with clearing of colonic adenomas yielded a 76–90% reduction in CRC compared with reference populations. Given the potential benefits of colonoscopy, multiple gastroenterologic professional societies have recommended colonoscopy as the preferred modality among the menu of screening options.

However, several published studies have addressed the shortcomings of colonoscopy as currently practiced. Indirect evidence from several studies has questioned the accuracy of colonoscopy. In one study, an audit of CRC cases revealed that 5% of subjects with CRC had undergone screening colonoscopy within 3 years of diagnosis. Chemoprevention studies that included follow-up colonoscopy as a means of measuring the efficacy of chemoprevention on polyp recurrence have also demonstrated a higher rate of developing interval cancers (1.7–2.2 cases per 1000 person-years) when compared with NPS results (0.6 cases per 1000 person-years). Recent studies have also questioned the effectiveness of colonoscopy in detecting proximal colonic neoplasia. A large population-based, case-control study using administrative claims data from Ontario, Canada, evaluated the association between colonoscopy and CRC deaths and found that colonoscopy was associated with lower CRC mortality rates. However, the association was primarily limited to deaths from cancer developing in the left side of the colon. The question arises whether CRCs in the proximal colon are more frequently missed due to the variability in endoscopists' performance of colonoscopy or because these lesions have different biologic features compared to distal CRCs, including higher rates of microsatellite instability and the CpG island methylator phenotype (CIMP), and perhaps these biologic features are overrepresented in interval CRCs. The performance of colonoscopy has been shown to vary by endoscopist, and the individual endoscopist may be a powerful predictor of adenomatous polyps detected during colonoscopy. To improve the accuracy (sensitivity) of colonoscopy for adenoma and CRC detection and standardize its performance among endoscopists, the U.S. Multisociety Task Force on Colorectal Cancer and the American Society for Gastrointestinal Endoscopy/American College of Gastroenterology Taskforce on Quality in Endoscopy have proposed several measurable factors of routine colonoscopy to serve as quality indicators. They include appropriate cecal intubation rates, adequate colonic cleansing with minimal fecal residue, a minimum withdrawal time from the cecum of 6 minutes, and goal adenoma detection rates for men and women undergoing screening colonoscopy. It is anticipated that the accuracy of colonoscopy can be increased through quality improvement initiatives for enhanced performance of colonoscopy.

Additionally, characteristics of the baseline colonoscopy are an important predictor for subsequent neoplasia. A population-based, case-control study using administrative claims data reports that the risk of developing CRC in patients following a negative result from a colonoscopy performed in the usual clinical practice is at most 60–70% of the risk of developing CRC in the general population and the duration of decreased CRC risk persists for more than 10 years. A recent prospective study examining the incidence of advanced neoplasia within 5.5 years of screening colonoscopy revealed a strong association between baseline colonoscopy results and the rate of serious incident lesions detected during the surveillance period. It has been shown that the baseline colonoscopy is the most beneficial in that it is responsible for the major benefit of polypectomy and that subsequent examinations may not add significant benefit except in people at high risk for future advanced adenomas. This supports the need to ensure that the baseline colonoscopy is of the highest quality in order to best detect adenomas.

Even though colonoscopy can fail to detect lesions, it is still considered the most sensitive of the screening methods. The advantages of this approach, in addition to the diagnosis and treatment of a lesion in a single session, include a long interval of "protection" in the setting of polypectomy and improved patient satisfaction due to the use of intravenous conscious sedation. However, colonoscopy involves greater cost and risk than other screening modalities. Serious complications associated with colonoscopy include bleeding and perforation and, while extremely rare, these occur at higher rates than with other screening options.

Colonoscopy may be the "preferred" screening modality for some, but it is important to emphasize that the evidence for its efficacy is indirect and that there are no randomized controlled trials demonstrating the superiority of colonoscopy to other screening modalities. In addition, no direct information indicates that a program of periodic colonoscopic surveillance leads to a reduction in CRC incidence or mortality in average-risk individuals. Large-scale, comparative effectiveness studies are needed to determine the impact of colonoscopy and other screening modalities on CRC incidence and mortality and whether the burden of right-sided neoplasia can be affected by colonoscopy to justify the associated extra cost and morbidity.

Noninvasive procedures to screen for CRC have been promoted as possible ways to increase compliance with CRC screening strategies. The 2008 consensus guidelines from the American Cancer Society, the U.S. Multisociety Task Force on Colorectal Cancer, and the American College of Radiology include CTC and fecal DNA in the testing options for CRC screening in asymptomatic, average-risk adults aged 50 years and older.

Computed Tomographic Colonography

Over the last decade, CTC (or virtual colonoscopy) was introduced as a possible alternative noninvasive test for the examination of the colon. It provides a two- and three-dimensional image of the colon using computer programming to combine data from multiple helical CT scans. CTC requires a colon preparation similar to that used for conventional colonoscopy and involves air insufflation through a rectal tube to distend the colon to enhance imaging. After the examination, a radiologist examines the scans; if an abnormality is noted, patients should undergo colonoscopy.

Results from early studies assessing the accuracy of CTC have found it to be comparable to that of conventional colonoscopy for the detection of polyps greater than 1 cm, with a high sensitivity and specificity of 94% and 96%, respectively. Similar results have not been reproduced in other studies, and subsequent sensitivities of approximately 50% have been reported. The inconsistency in the examination's efficacy is likely related to the selection bias of the population studied, because most studies on CTC have been performed in patient populations at increased risk for colorectal neoplasia and more likely to have occult disease. In addition, only highly trained radiologists were involved in interpreting the results of CTC in the majority of these studies. Therefore, the generalizability of the results of these studies remains unclear.

The technology related to CTC has improved over the past few years, and several screening trials using advanced CTC techniques have demonstrated its favorable performance. In a meta-analysis reviewing the cumulative published CTC performance data (including high-risk and screening cohorts), the pooled sensitivity and specificity for large (>1 cm) polyps were 85–93% and 97%, respectively, and for small polyps (6–9 mm) were 70–86% and 86–93%, respectively. The overall sensitivity for invasive CRC was comparable to that of conventional colonoscopy (96%).

There are a number of limitations and unresolved issues related to CTC. The examination is not widely available and most often is performed at large, academic centers, usually in the setting of research studies. In addition, the false-positive rate is estimated to be about 15% and is related to retained stool mistaken for polyps, diverticular disease that limits colon distensibility, and thickened colonic folds mistaken for lesions. CTC is also not therapeutic, as polyps cannot be removed and masses cannot be sampled for tissue diagnosis during the procedure. Lastly, it is estimated that 27% of patients undergoing CTC will have an incidental finding warranting further diagnostic evaluation. This may increase patient anxiety and lead to an increase in medical costs, because most incidental findings are expected to be benign. It is estimated that there is only a 2% potential clinical benefit to an extensive evaluation for incidental findings noted on CTC.

Although there is variability regarding the optimal manner in which to perform the examination, as well as inconsistent radiology training and competency, the available data supporting its efficacy have rendered CTC an acceptable option for CRC screening in average-risk adults age 50 years and older. However, the interval for repeat examinations is uncertain, and future studies are warranted to determine the appropriate surveillance interval following an initial negative CTC. Presently, there is consensus that all patients with one or more polyps greater than 1 cm or three or more polyps greater than 6 mm should be referred for conventional colonoscopy. The management of patients with fewer than three polyps or polyps of 6–9 mm remains controversial. Research studies using CTC surveillance to evaluate the natural history of polyps in this size range are ongoing. At this time, a reasonable approach for patients with 6–9 mm polyps identified by CTC is to offer therapeutic colonoscopy. There is also a need for multidisciplinary consensus regarding the management of polyps smaller than 6 mm.

For effective implementation of CRC screening, the evolving technology needs to be standardized along with the reporting of findings. Additional information is also needed regarding the safety profile of CTC, including the radiation dose effects for computed tomography examinations for average-risk CRC screening. At present, experts have differing opinions about when and how CTC should be used. The U.S. Preventive Services Task Force decided that there was not enough evidence to recommend it as an option for CRC screening, and the federal Medicare program does not currently provide reimbursement for the use of CTC for routine screening. In a recent cost-effectiveness analysis of CRC screening tests that have been recommended by the U.S. Preventive Services Task Force, American Cancer Society, U.S. Multisociety Task Force on Colorectal Cancer, American College of Radiology, and American College of Gastroenterology, CTC was not deemed a cost-effective strategy for average-risk CRC screening.

Fecal DNA Testing

Testing to detect genetic abnormalities (somatic mutations) in fecal samples has been proposed as a potential screening test for CRC screening. Colorectal epithelial cells are shed in stool, and stable DNA can be extracted from stool and amplified using polymerase chain reaction (PCR). This process allows for the detection of mutations in several genes, including K-ras, APC, BAT-26, and tp53.

Early data from a small cohort revealed a sensitivity of 91% for cancer, sensitivity of 82% for adenomatous polyps greater than 1 cm, and specificity of over 90% for either, whereas other studies have reported sensitivities between 52% and 91% for cancer and 27% and 82% for adenomatous polyps. Variable results assessing the sensitivity of fecal DNA testing have been related in large part to the test's evolving technology. In a recent study using a newer fecal DNA test aimed at detecting methylated genes in CRC (Cologuard), 1187 specimens from patients with and without colorectal neoplasia were blindly evaluated, and the accuracy of the fecal DNA test was 85.3% for CRC and 63.8% for adenomas greater than 1 cm, with a specificity of 88% for either. CRCs and adenomas were detected equally well in the proximal and distal colon with higher accuracy for early-stage CRCs and adenomas of increasing size. In addition, the sensitivity of the test was unaffected by patient gender, age, or race.

As fecal DNA testing is able to identify the majority of CRC cases and large adenomas, there has been an expert consensus to include this modality as an acceptable option for CRC screening in average-risk adults. However, newer fecal DNA tests (whether they involve technologic advances made on existing tests or are new variants) warrant careful evaluation in future studies evaluating screening cohorts. At this time, the interval for routine screening between normal examinations is unclear, and there is no societal recommendations given the insufficient data. Nevertheless, the effectiveness of any screening program utilizing fecal DNA will depend on quality, adherence, and cost, as well as the sensitivity and specificity of the test.

CRC is a preventable cancer. The appropriate identification of individuals at risk dictates the necessary screening and surveillance approach and plays an important role in decreasing CRC mortality. There are several acceptable options presently available for CRC screening, and they provide the means to improving screening compliance. Although a number of randomized trials provide evidence to support the use of FOBT and FS for CRC screening, high-quality colonoscopy is considered the current gold standard given its ability to detect and remove adenomatous polyps, which in turn decreases CRC incidence and related mortality.