Colorectal cancer is one of the few preventable cancers. The removal of precancerous polyps during colonoscopy inhibits their progression to cancer. Widely available screening tests such as the fecal occult blood test (FOBT), sigmoidoscopy, and colonoscopy are associated with a decrease in colorectal cancer mortality. Unfortunately, less than 50% of eligible Americans have undergone screening, and more than 45% of cancers are diagnosed at a time when the cancer is incurable.

Definition

The colorectum is the portion of the gastrointestinal (GI) tract most commonly affected by tumors. Most colonic tumors are benign epithelial polyps. Polyp is an inexact term that indicates a protuberance of tissue into the colonic lumen. There are many histologic types of polyps. The best characterized and most common cancer precursor is the adenomatous polyp. The size, number of adenomas, and degree of villous features predict the future risk of advanced neoplasia, including malignancy in patients who harbor adenomas. An additional precursor to colorectal cancer called a sessile serrated adenoma has been identified. Persons who harbor such lesions should be managed similarly to those with adenomas.

This chapter focuses on the most common neoplastic epithelial tumors, including adenomatous polyps and adenocarcinoma of the colon and rectum.

Back to Top

Prevalence and Risk Factors

The incidence of colorectal carcinoma has been decreasing an average of 1.5% annually since the mid-1980s. Research has suggested that the decline may be caused by an increased use of screening and polyp removal, which prevents progression from polyp to cancer. Although almost 40% of Americans 50 years and older harbor adenomatous polyps, it is estimated that only 2% of adenomas will progress to cancer.

Adenocarcinoma of the colon and rectum is the third most common cancer and cause of cancer deaths in the United States. Both men and women face a lifetime risk of almost 6% for the development of invasive colorectal cancer (Table 1).¹ It is estimated that approximately 149,000 new cases of colorectal cancer were diagnosed in 2008 and that 50,000 deaths from colorectal cancer will occur. The only known race predilection is in African Americans, who have higher colorectal cancer incidence and mortality rates.

Epidemiologic studies have implicated a number of environmental cofactors in the development of colorectal cancer. They include advanced age, a diet high in red meat, a diet high in fat, smoking, alcohol consumption, and obesity. In approximately 30% of patients with colorectal cancer, risk factors have been identified (Box 1), and the remaining 70% of newly diagnosed colorectal cancers arise in patients without any identifiable risk factors.

Back to Top

Pathophysiology and Natural History

Well-established and newly discovered molecular and genetic pathways of colorectal carcinogenesis including chromosomal instability, microsatellite instability, and CpG island methylation (CIMP) have broadened our understanding of the precursors to colorectal cancer.¹ In approximately 80% of cases, colorectal carcinoma arises from an adenomatous polyp. Observational studies have suggested that the adenoma-to-carcinoma sequence takes approximately 10 years. This pathway traditionally was believed to result from multiple acquired genetic alterations in tumor tissue and is called the chromosomal instability pathway. The remainder of colorectal cancers arise from alterations in DNA repair genes, which are characterized by extensive instability in simple repeated nucleotide sequences (microsatellites) called microsatellite instability or by widespread hypermethylation in gene promoter regions (CIMP).² Convincing evidence shows that serrated polyps, in particular sessile serrated adenomas, are the precursor to CIMP cancers. These precursor serrated lesions are usually large and have high levels of CIMP, BRAF mutations, and a proximal colon predilection.

A personal history of adenomatous polyps or colorectal cancer increases the risk for metachronous colorectal cancer. First-degree relatives of patients with colorectal cancer have a two- to threefold increased risk for colorectal cancer and adenomatous polyps. A systematic review and meta-analysis has proved that first-degree family members of patients with adenomatous polyps also have an increased risk of colorectal cancer themselves, particularly when the relative’s adenoma is diagnosed before age 60 years.³

Patients with the highest risk for colorectal cancer are those who have a hereditary colorectal cancer syndrome. The dominantly inherited syndromes include familial adenomatous polyposis (FAP) and Lynch syndrome, which has also been known as hereditary nonpolyposis colorectal cancer (HNPCC). An autosomal recessive colon cancer syndrome has been discovered, MYH-associated polyposis (MAP).

FAP is caused by a mutation in the tumor suppressor gene APC on the long arm of chromosome 5. Germline mutations in APC can be detected in approximately 80% of patients with FAP. The APC mutation results in the development of hundreds to thousands of colonic adenomas, usually by the second decade of life. Colon cancer develops in all FAP patients by age 40 years if prophylactic colectomy is not performed. Attenuated FAP is a phenotypically distinct form of colonic polyposis. The number of colonic polyps in these patients usually numbers less than 100, and the age of onset of polyposis and cancer is shifted 1 to 2 decades later. Duodenal adenomas are common in either form of FAP, and periampullary cancer is the second leading cause of cancer death in this population. Gardner’s syndrome is a phenotypic variant of FAP. In addition to colonic polyposis, other manifestations may be seen, such as benign soft tissue tumors, osteomas, supernumerary teeth, desmoid tumors, and congenital hypertrophy of the retinal pigment epithelium.

MAP is associated with mutations of the MYH gene. Biallelic germline mutations in MYH are responsible for 10% of the APC-negative cases of FAP and up to 29% of patients with numerous adenomas (>15) during their lifetime. The affected patients usually have fewer than 100 colorectal adenomas and an increased risk of cancer.⁴

Lynch syndrome is caused by genetic alterations in one of the mismatch repair genes. Alterations in these genes prevent adequate repair of DNA, resulting in a change of length of nucleotide bases (MSI) when compared with those of normal tissue. MSI can be investigated in a colon cancer specimen in a molecular pathology laboaratory. If MSI is present, immunohistochemistry (IHC) should be performed to assess for the lack of protein expression from one of the mismatch repair genes. MSI or abnormal IHC can suggest the tumor has developed in a patient with Lynch syndrome. In Lynch syndrome, colorectal cancer occurs in up to 80% of those affected, usually by the age of 50 years, and is often right-sided and associated with unusual pathologic features. Extracolonic cancers, such as endometrial, ovarian, small bowel, transitional cell of the ureter or bladder, and gastric cancer, often occur in patients with Lynch syndrome. The risk of endometrial carcinoma has been reported in up to 60% and ovarian carcinoma in up to 20%. Therefore, aggressive gynecologic screening for endometrial and ovarian cancers is recommended in women in Lynch kindreds.

Clinical criteria have been developed to identify families with Lynch syndrome. The strictest criteria include the Amsterdam criteria:

• Three or more relatives with colorectal cancer, with one a first-degree relative of the other two
• At least two successive generations affected
• One cancer diagnosed before age 50 years

Many researchers have found the Amsterdam criteria neither sufficiently sensitive nor specific for use as the sole criterion for determining which families should undergo intensive surveillance or genetic evaluation. The Amsterdam II criteria broadened the original criteria to include other HNPCC-associated cancers, including colorectal cancer or cancer of the endometrium, small bowel, ureter, or renal pelvis.⁵

The revised Bethesda criteria were formulated to identify tumors that should be tested for MSI⁶:

• Colorectal cancer diagnosed before the age of 50 years
• Presence of synchronous or metachronous colorectal or other HNPCC-associated tumors, regardless of age
• Colorectal cancer with MSI diagnosed in a patient younger than 60 years
• Colorectal cancer diagnosed in a patient with one or more first-degree relatives with an HNPCC-related tumor, with one of the cancers being diagnosed before age 50 years
• Colorectal cancer diagnosed in a patient with two or more first- or second-degree relatives with HNPCC-related tumors, regardless of age

The chronic inflammatory colitides, ulcerative colitis and Crohn’s disease, are associated with an increased risk of colorectal cancer. The proximal extent of colonic involvement, duration of disease, and activity of disease stratify the level of risk. Risk is highest in patients with pancolitis and is negligible in patients with proctitis. After a decade of disease, the cancer risk increases yearly by 1% to 2%.

Back to Top

Signs and Symptoms

Colon polyps and early colon cancer are often asymptomatic until they are advanced. GI blood loss is the most common sign and can include a positive FOBT result, iron-deficiency anemia, or hematochezia. When tumors are advanced, unexplained anorexia, weight loss, or symptoms from obstruction or local invasion, such as a change in bowel habits, abdominal pain, or obstruction, can occur.

Back to Top

Diagnosis

The diagnosis of colorectal polyps and cancer is most often made during a colonic evaluation performed for GI symptoms, for colorectal cancer screening, or as part of endoscopic surveillance.

Screening

National organizations, including the American Cancer Society, United States Preventive Services Task Force, and U.S. Multi-Society Task Force on Colorectal Cancer, have established guidelines for colorectal cancer screening and surveillance. Various options are recommended for average-risk patients (Table 2).⁷ Colonoscopy is recommended by the American College of Gastroenterology as the preferred colorectal cancer screening strategy. The evidence to support colonoscopy has been derived from data showing a decreased incidence of colorectal cancer mortality in subjects who have undergone colonoscopic adenoma removal. Additionally, colonoscopic screening has been shown to have favorable cost effectiveness when compared with other screening strategies. Unfortunately, lesions may be missed on colonoscopy, and up to 5% of patients in whom colorectal cancer is diagnosed had undergone colonoscopy within the previous 3 to 5 years.⁸

Patients with symptoms of colorectal cancer and those with more than a moderate to high risk for colorectal cancer should undergo colonoscopy (Table 3). Surveillance intervals differ based upon risk factors including adenoma characteristics of size, multiplicity, histology, and family history.

Fecal Occult Blood Test

Randomized trials have found that the use of annual guaiac based FOBTs (g-FOBT) has decreased the mortality from colorectal cancer by up to 33%. Traditional guaiac FOBTs take advantage of the peroxidase activity of hemoglobin from bleeding colonic lesions that can be detected in the stool by a color change when it catalyzes the oxidation of guaiac by a peroxide reagent. A special diet (e.g., a meat-free, high-residue diet without vegetables that have peroxidase activity, such as turnips and horseradish) is recommended for at least 24 hours before three separate stool specimens are collected at least 1 day apart. Unrehydrated test sensitivity is low at approximately 80%, with a specificity of up to 98%.

Immunochemical methods for FOBT (i-FOBT) use antibodies to human globin. The potential advantages of the i-FOBT include automation of test results and a potential increase in compliance, because no dietary restrictions are needed because the antibodies do not cross-react with nonhuman hemoglobin or peroxidases from food sources and the collection method is simpler, requiring only one or two samples. In addition, because globin does not survive passage through the upper GI tract, the test is specific for bleeding in the colon and rectum. A distinct advantage of i-FOBT is that the clinician can determine the fecal hemoglobin threshold to determine test characteristics and the level at which colonoscopy should be performed. When Levi and colleagues set the hemoglobin threshold at a cutoff of greater than 75 ng/mL, they found a sensitivity and specificity for cancer of 94% and 87.5%, respectively. The corresponding accuracy for all clinically advanced neoplasia was 67% and 91.4%, respectively.⁹

Although i-FOBT has not been studied in widespread randomized, controlled screening trials, in case-control studies a reduction of colorectal cancer mortality of up to 80% has been shown.^10-12

The accuracy of a one-time brush-sample i-FOBT has been studied and shows improved test characteristics compared with g-FOBT. In a study of more than 2500 subjects, i-FOBT was positive significantly more often than g-FOBT in subjects with cancer (87.5% vs. 54.2%) and with significant adenomas (42.6% vs. 23.0%).¹³Beneficial results were also confirmed for a study of nearly 6000 subjects. The sensitivity for detecting cancer was 82% for i-FOBT and 64.3% for g-FOBT. The sensitivity for detecting advanced neoplasia was 41% and 29%, respectively. The specificity of each modality was greater than 98%.¹⁴ The 2008 US PSTF guidelines for CRC screening recommend using an FOBT with a high sensitivity for the detection of colorectal cancer.

Stool DNA Testing

Stool DNA testing takes advantage of the potential for exfoliated neoplastic cells to pass into stool in order to detect mutations from within the molecular pathway that leads from adenoma to carcinoma. The largest study that compared fecal DNA to g-FOBT (Hemoccult II) in a screening population who underwent colonoscopy used a fecal DNA panel consisting of 21 mutations (three in the Kras gene, 10 in the APC gene, and eight in the p53 gene), the microsatellite instability marker BAT-26, and a marker of long DNA thought to reflect disordered apoptosis of cancer cells sloughed into the colonic lumen. The fecal DNA panel detected 52% of the cancer cases vs. 13 % of the g-FOBT. The corresponding results for the detection of advanced neoplasia was 18% versus 11 %.¹⁵ The results were disappointing, and the poor accuracy of fecal DNA was thought to be due to problems with technologic aspects of the DNA testing.

A second-generation fecal DNA test, which stabilizes DNA for transport and allows better extraction of DNA from stool, was tested in a small study and showed an increase in sensitivity for cancer to 72.5% and a specificity of 89%.¹⁶ When a methylation marker of the vimentin gene was combined with the DNA integrity (DIA) assay, the performance of those two markers alone had an improved sensitivity of 87.5% and an 82% specificity.

Computed Tomography Colonography

Over the last decade, significant advances have occurred in computed tomography (CT) technology, and there is widespread interest in the use of CT colonography (CTC) for colorectal cancer screening. Many studies have been conducted to determine the accuracy of CTC for detecting colorectal neoplasia. The best accuracy was reported by Pickhardt and colleagues, who found the sensitivity of CTC comparable to colonoscopy for detecting adenomas larger than 6 mm (86%) and 92% for adenomas larger than 10 mm.¹⁷ Poorer results came from Cotton and colleagues, with a sensitivity 32% for lesions larger than 6 mm and 52% for lesions larger than 10 mm. Rockey and colleagues found a sensitivity of 47% for polyps 6 to 9 mm and 53% for polyps larger than 1 cm.^18-19 The exact reasons for the disparity in the results are unknown. The results of the National CT Colonography Trial (ACRIN) of 2531subjects has not been published yet; however, preliminary reports (auntMinnie.com 9/28/2007) found the sensitivity for detecting lesions larger than 5 mm was 65% and for lesions larger than 9mm was 90%. The specificity was approximately 86% for any size.

Advocates of CTC suggest that patient compliance with screening will improve if CTC is an option. The fact that it is less invasive, is not associated with sedation, has little procedure-related risk, and has the potential to assess for lesions outside the colon might make it attractive to patients. However, the detection of insignificant extracolonic lesions might result in unnecessary health care expenditures, and the lifetime radiation-induced risk is a consideration. Other drawbacks include lack of reimbursement for screening by third-party payers, need for bowel preparation, and lack of detection of polyps smaller than 5 mm. Multidisciplinary collaboration between gastroenterologists and radiologists regarding the delivery of care between the radiology and endoscopy suite, interpretation and reporting of intraluminal and extraluminal findings, and proper follow-up of patients with detected lesions are necessary before widespread use of CTC is implemented.

Sigmoidoscopy

Results of several case-control studies have shown a reduction in deaths from colorectal cancer in subjects who have undergone sigmoidoscopic examinations. The reported reduction in mortality varies between 59% and 80%. The best-known study compared the use of rigid sigmoidoscopic screening in 261 patients who died from cancer of the distal colon or rectum with 868 control subjects. Screening reduced the rectosigmoid cancer mortality rate by 60%, and the protective effect of sigmoidoscopy was noted to last for up to 10 years. This reduction in mortality may have resulted from earlier detection of cancer and removal of premalignant polyps.

Sigmoidoscopic screening allows the lower one third of the colorectal mucosa to be visualized directly and diagnostic biopsy to be performed at the time of examination. Sensitivity and specificity are high for detecting polyps and cancer in the segment of the bowel examined. Unfortunately, however, almost 50% of polyps and cancers are beyond the limits of detection of the longest (e.g., 60 cm) flexible sigmoidoscope.

Opinions vary regarding the need for colonoscopy for patients in whom a single small (<1 cm) adenoma is found on flexible sigmoidoscopy. Studies have shown that the prevalence of advanced proximal neoplasms in patients with distal adenomas is up to 9%. Therefore, the use of colonoscopy to detect proximal neoplasia in patients with distal adenomas continues to be recommended.

Barium Enema

Barium enema has the advantage of imaging the entire colon. However, recent evidence has suggested that it is inaccurate for detecting polyps and early cancers and suboptimal for colorectal cancer screening or surveillance. In a prospective study comparing the use of double-contrast barium enema and colonoscopy, barium enema missed 52% of polyps larger than 1 cm.²⁰ If barium enema is the only option for screening or surveillance, it should be coupled with flexible sigmoidoscopy. The use of flexible sigmoidoscopy allows visualization of the rectosigmoid, which might not be well seen on barium enema because of the overlapping loops of bowel. Lesions detected on barium enema warrant colonoscopic evaluation.

Colonoscopy

Colonoscopy is the gold standard for detecting colonic neoplasms and is the preferred colorectal cancer screening strategy. The incidence rate of colorectal cancer has been shown to be reduced up to 90% in subjects who had polypectomy versus patients in three reference groups, including two cohorts in whom colonic polyps were not removed and one general population registry. Colonoscopy can be completed in more than 95% of examinations with negligible risk. Colonoscopic screening in average-risk patients has been found to be cost effective, and it is similar to cervical or breast cancer screening techniques in cost effectiveness per life-year saved. Medicare has approved the use of screening colonoscopy in average-risk beneficiaries. Unfortunately, not all Americans younger than 65 years have health care benefits that cover the charges for colonoscopy, which could affect patient compliance with screening colonoscopy.

The stage of colorectal cancer is the most important feature predicting curability and survival in colorectal cancer. The depth of tumor invasion and lymph node involvement are the two major components constituting the basis for colorectal cancer tumor staging. The first colorectal cancer staging system was developed in 1932 and is known as the Dukes’ classification. Since then, many modifications in the Dukes’ scheme have been made. In an effort to minimize confusion over which Dukes’ stage is implied and to standardize all organ system cancer staging, the TNM (tumor, nodal status, and presence of metastases) system has been adopted. Stages II and III have been further stratified based on tumor size. In addition, investigators used the surveillance, epidemiology, and end results (SEER) data to calculate 5-year survival rates, and patients with stage IIIA colon cancer were found to have significantly better survival than patients with stage IIB disease.

Back to Top

Treatment

The primary treatment of colorectal cancer is surgical resection of the primary tumor and regional lymph nodes. Surgery is curative for most early-stage colorectal tumors. For more advanced stages, surgery and adjuvant therapy are recommended to prevent recurrence and prolong survival.^20-22

Colon Cancer

Studies have proved the benefit of adjuvant chemotherapy in prolonging disease-free and overall survival in patients with stage III colon cancer compared with patients who received surgery alone.²⁰ Patients with stage III colon cancer who undergo surgery alone have a 50% risk of relapse; these patients benefit from 5-fluorouracil (5-FU)-based adjuvant therapy, with a 30% reduction in risk of death. Despite this recommendation for postoperative chemotherapy, there is documented underuse of adjuvant therapy in patients with stage III colon cancer.

Use of adjuvant therapy in patients with stage II colon cancer has been actively debated. Currently, the American Society of Clinical Oncologists does not recommend routine use of adjuvant chemotherapy in patients with stage II colon cancer. Certain patient populations, however, may be considered for adjuvant treatment, including patients with inadequately sampled nodes, T4 lesions, perforation, or poorly differentiated histology. Additionally, patients with stage II disease should be encouraged to participate in clinical trials aimed at determining the true benefit, if any, of adjuvant therapy in this patient population.

Rectal Cancer

The major limitation of rectal cancer surgery is the inability to obtain wide margins because of the confined space of the bony pelvis. Adjuvant radiation therapy decreases the rate of local recurrence, whereas the addition of systemic chemotherapy further enhances local control and improves survival. A 1990 NIH Consensus Development Conference recommended that postoperative 5-FU–based chemotherapy combined with irradiation should be the standard clinical practice in stages II and III rectal cancer because of its proven decrease in local recurrence, cancer-related deaths, and overall mortality.²¹

In the last decade, subsequent randomized trials have been challenging that recommendation. The debate is fueled, in part, by refinements in surgical techniques. Total mesorectal excision (TME) is one of the most exciting recent developments in surgical oncology. The sharp dissection follows along the mesorectal fascia, with removal of the rectum and of all tissue invested by the adjacent visceral fascia, including fatty tissue, lymph nodes, and lymphatic vessels. This technique has been associated with a reduction in local recurrence rates from 39% to less than 10%. These low rates have led some investigators to question the routine use of adjuvant radiation therapy. One study has compared the use of TME plus preoperative radiotherapy versus TME alone.²⁵ Short-term survival was no different in the two groups, but the rate of local recurrence was 2.4% for TME with preoperative radiotherapy versus 8.2% for TME alone.

There is also controversy over the best timing of radiotherapy. Compared with postoperative irradiation, preoperative treatment might have lower toxicity and might increase the chance of preserving the sphincter. The information from all controlled trials reported so far shows that the fraction of local recurrences is reduced to less than 50% when radiotherapy, up to moderately high doses, is given preoperatively. This reduction is smaller with postoperative radiotherapy, even if higher doses are used. Improved survival has been seen in trials using postoperative radiotherapy, but only when combined with chemotherapy. In one trial, a survival benefit was incurred with preoperative radiotherapy versus surgery alone. Randomized trials comparing preoperative and postoperative combined modality therapy are in progress.

Outside of clinical trials, curative-intent surgery combined with radiochemotherapy remains the recommended standard for treatment of stages II and III rectal cancer.

Back to Top

Prevention

Epidemiologic studies have found a modest decrease in colorectal cancer in patients using nonsteroidal anti-inflammatory drugs, particularly aspirin. Two randomized, controlled trials have found that aspirin substantially reduces the risk of recurrent and advanced adenomas in high-risk populations.^23,24 Two randomized, placebo-controlled trials have found a moderate reduction in recurrent adenomas with calcium supplementation.^25,26

A 6-month placebo-controlled trial of a selective cyclooxygenase (COX)-2 inhibitor, celecoxib, in FAP patients resulted in a 28% reduction in polyp size and number.²⁷The U.S. Food and Drug Administration has approved the use of celecoxib as an adjunct for managing colorectal adenomas in patients with FAP. The value of COX-2 inhibitors celecoxib and rofecoxib in the sporadic adenoma population has been investigated in three large multicenter studies of patients with a history of colorectal adenoma. The efficacy results of these 3-year trials all demonstrated a significant reduction in the recurrence of both adenomas and advanced adenomas. Unfortunately the studies were stopped prematurely because of adverse cardiovascular events.^28-30

The increased cardiovascular risk of COX-2 inhibitors appears to outweigh the potential for preventing colorectal neoplasia in patients with sporadic colorectal adenomas.³¹The effectiveness and cost-effectiveness of chemopreventive agents in different risk populations needs to be confirmed before widespread recommendations for their use can be given.

Back to Top

Conclusions

Colorectal cancer is one of the leading causes of cancer and death from carcinoma in the United States. Increasing awareness regarding the preventable nature of this disease, along with widespread use of screening, should favorably affect the incidence of colorectal cancer. Colorectal cancer screening and polyp removal can save lives, and the most exciting area of future research will be the primary prevention of adenomas and colorectal cancer through chemoprevention.

Back to Top

Summary

• Adenocarcinoma of the colon and rectum is the third most common cancer and cause of cancer deaths in the United States, but studies have suggested that the incidence is declining because of the increased use of screening and polyp removal.
• Colon cancer screening and surveillance strategies must be individualized based on a patient’s risk factors, including personal history of adenomas or colorectal cancer, family history of adenomas or colorectal cancer, family history of an inherited colorectal cancer syndrome, and personal history of inflammatory bowel disease.
• The stage of colorectal cancer is the most important feature predicting curability and survival in colorectal cancer.
• Continued investigations of chemopreventive agents are needed to identify a means of primary prevention of adenomas and colorectal cancer.

Back to Top

Suggested Readings

• Giardello FM, Brensinger JD, Petersen GM. AGA technical review on hereditary colorectal cancer and genetic testing. Gastroenterology. 2001, 121: 198-213.
• Lieberman DA, Weiss DG. One-time screening for colorectal cancer with combined fecal occult-blood testing and examination of the distal colon. N Engl J Med. 2001, 345: 555-560.