Current methods of colorectal cancer screening. (Cover Story).

Cancer of the colon and rectum is a leading cause of cancer death in the United States. Since this disease has a well-defined progression, accurate patient screening can change the overall prognosis and outcome in individuals with early disease. Currently, the American Cancer Society recommends screening selected at-risk individuals with fecal occult blood testing, colon X-ray, and colonoscopy in an attempt to detect colon cancers at an early stage. During the past 20 years, there has been an overall improvement in colorectal cancer survival, thought to be due to early detection of disease. In fact, randomized trials have demonstrated a reduction in mortality of 15 percent to 33 percent over 10 years to 14 years, with fecal occult blood testing.

Despite these promising advances, the sensitivity and specificity of the available screening tests are relatively low. As colon cancer progresses, specific mutations in genes such as K-ras, APC, p53, and Bat 26 may occur. In the past seven years, new methods of detecting these DNA changes in ex foliated cancer cells have been proposed as a feasible option for early detection of colorectal cancer. New tests are being developed to detect these gene mutations, and initial studies demonstrate they have increased sensitivity and specificity over current methods. These new tests may hold the key to accurate early diagnosis in the future.

Introduction

Colorectal cancer is the second leading cause of cancer death in the United States, with more than 570,000 new cases reported worldwide each year. (1) Of the 131,000 cases diagnosed in the United States in 1998, only half of these people are predicted to survive five years." (2) An overall 6 percent to S percent improvement in survival has occurred over the past 20 years, which may be due to advances in prevention and diagnosis. (1) Since colorectal cancer follows a predictable natural history of disease from benign neoplastic polyps to cancer, it is particularly amenable to screening. The ability to detect a potential cancer in advance allows for treatment and cure in many cases. However, the implementation of screening tests has not been as widespread or accepted as the current recommendations suggest. Current screening practices are based on fecal occult blood testing, colon X-ray, and colonoscopy. Inadequacies of these tests include incomplete detection rates, in addition to the discomfort and inconvenien ce associated with some of the procedures.

Pathogenesis of colon cancer

Polyps are masses of tissue that protrude into the lumen of the intestine and can form throughout the length of the gastrointestinal tract. Polyps are significant lesions because some appear to progress to colon cancer. Non-neoplastic polyps represent 90 percent of the large intestine polyps and are mostly sporadic in their origin (Table 1). Adenomatous polyps (also called adenomas), however, are true neoplastic lesions and have the potential to develop into adenocarcinoma, or colorectal cancer. Adenomatous polyps range in size from small, often pedunculated (stalk-like) lesions, to large usually sessile (fiat) forms. This type of polyp is still considered sporadic in its origin, although there is some familial predisposition, and hereditary factors clearly play a role in inherited familial syndromes such as familial adenomatous polyposis. The subtypes of adenomatous polyps and their relationship to adenocarcinoma potential are shown in Table 2. (3)

Adenomatous polyps have been associated with adenocarcinoma of the colon by several lines of evidence: 1) adenomas are uncommon in areas where colon cancer is low; 2) the incidence of colonic adenomas increases with age in countries with a high or intermediate risk for colon cancer, occurring in 30 percent to 40 percent of individuals older than 60 years in the United States; and 3) the potential of adenomatous polyps to transform into adenocarcinoma increases with the size and histologic appearance of the polyps. (2)

Adenomatous polyps are considered invasive adenocarcinoma when the lesion invades into the submucosa of the bowel wall. In the submucosa, the carcinoma gains access to lymphatic channels and vessels through which it can travel to other areas of the body (metastasize). The majority of the colonic adenocarcinomas originate in the cecum or ascending colon (38 percent), and the sigmoid colon (35 percent). The lesions are usually single, and the macroscopic morphology, along with the patient's symptomatology, reflects the lesion's location. Lesions of the proximal colon are often protruding masses that expand into the gut lumen and can bleed because of their size. They usually cause vague symptoms that often go unnoticed longer than the changes in bowel habits experienced by patients with distal colon carcinomas. Adenocarcinomas of the distal colon tend to grow as circular lesions that encompass the entire circumference of the colon. As these lesions expand, they cause narrowing of the gut lumen, which leads to ea rlier presentation with symptoms of constipation or diarrhea. These lesions also bleed, and because of their proximity to the rectum, typically cause noticeable bright red blood in the stool. Even though patients with distal colon carcinoma may present earlier, these lesions are often more infiltrative at the time of diagnosis, and thus they have a poorer prognosis.

A genetic model has been proposed outlining the progression from normal epithelium to colorectal cancer in the sporadic (nonfamilial) forms of colorectal cancer. Morphological changes have been shown to parallel genetic changes in colorectal cancer. (4) It is proposed that the adenomatous polyposis coli (APC) gene is lost or mutated early in this process in normal epithelium, leading to hyperproliferative epithehum in the intestine. During this hyperproliferative phase, DNA methylation is lost, which leads to the formation of adenomas. It is thought that adenomas incur a mutation of the ras gene leading to their evolution into "intermediate" adenomas. Further loss of a tumor suppressor gene on chromosome 18 eventually leads to "late" adenomas. Finally, the loss of the p53 gene is proposed to cause the formation of carcinomas. (4) All of these mutated genes play an integral role in controlling the cell cycle and are shown in Table 3.

Current screening methodologies

Screening for colorectal cancer requires an identifiable marker of all neoplastic lesions (adenomatous polyps and adenocarcinoma). The American Cancer Society currently recommends that beginning at age 50 (average risk), a yearly fecal occult blood test and a flexible sigmoidoscopy should be performed, and repeated every five years. Other screening methods available include double contrast barium enema and colonoscopy. Other recommendations by the American Cancer Society can be found at www.cancer.org.

When screening is recommended, it generally starts with fecal occult blood testing (FOBT). Many colorectal cancers bleed into the intestinal lumen, and fecal occult blood tests can detect the presence of blood that is otherwise unapparent through simple inspection. As blood passes through the gastrointestinal tract, it becomes degraded, and depending upon the site at which the hemorrhage occurs, blood products in the stool will vary. For example, if a lesion is located in the proximal colon, hemoglobin will be completely digested and will be metabolized by bacteria into porphyrins. However, if the lesion is in the distal colon, hemoglobin and heme will remain mostly intact and unchanged. The design of the FOBT, therefore, must take these variations into consideration. (5)

There are several different FOBTs currently available on the market. The most commonly used is the Hemocult II test (SmithKline Diagnostics). (6) Stool is placed on guaiac impregnated test pads that detect the peroxidase-like activity of heme by changing the color of the pad from white to blue. (5) At least three samples must be collected since most colorectal lesions have an intermittent bleeding pattern. Red meat and certain fresh fruits and vegetables can lead to false positive results. These food sources, therefore, should be avoided for three to five days prior to and during testing. Additionally, vitamin C intake should be avoided over the same time period because it can inhibit the guaiac reaction leading to false negative results. Nonsteroidal anti-inflammatory drugs, alcohol, or any other gastric irritants should also be avoided for a three-day period prior to collection, as they may increase gastrointestinal bleeding. (5)

The physiologic loss of blood into the gastrointestinal tract of a normal, healthy individual is approximately 0.7 mL per day. The amount of additional blood loss by small colorectal cancers (< 2 cm), if they bleed at all, is usually only 1 mL to 2 mL, which may not be detected by FOBTs.

As a whole, FOBTs have a limited ability to decrease mortality by themselves. Three large randomized studies have demonstrated a reduction of mortality by 15 percent to 33 percent in colorectal cancer patients over a 10 year to 14 year period. (7-9) Despite these aggressive screening attempts, 67 percent to 85 percent of those with colon cancer who underwent FOBT died from the disease. This indicates that detection does not occur early enough to maximally affect the overall outcome of the disease. This also indicates that FOBT is not a sensitive test and apparently misses many early stage cancers and adenomas. (10)

In contrast to FOBTs, minimally invasive procedures effectively detect neoplastic lesions. Since more than 60 percent of early lesions were believed to arise in the rectosigmoid areas of the large intestine, rigid sigmoidoscopy was routinely used for screening in the past. (11) Over the past several decades, however, there has been an increase in the number of lesions arising from more proximal regions of the colon, necessitating the use of flexible, fiberoptic sigmoidoscopes. Although these methods are very effective and offer a means of removing neoplastic polyps, they still leave all lesions beyond the reach of the scope (estimated between 25 percent and 34 percent) undetected. (12) Colonoscopy, based upon the same principle as sigmoidoscopy, allows visualization of the entire colon. Although it is the gold standard in colorectal cancer screening, it is very expensive, requires cathartic preparation, and has an increased risk of morbidity and mortality. Double contrast barium enema radiographic studies dis play the entire colon, but, partly due to limitation of two-dimensional viewing, have a detection rate much lower than that of colonoscopy. Computed tomographical (CT) colography (virtual colonoscopy) offers greater sensitivity in comparison, (13) but has a false positive rate of 6 percent to 17 percent, necessitating further examination by colonoscopy. (14-16) Additionally, limited insurance reimbursement and exceedingly high cost of this procedure makes it prohibitive.

For detection of recurrent disease after surgical resection, it is recommended that patients follow-up with history, physical examination, and laboratory tests including liver enzymes every three months for three years. (2) Liver enzymes are followed because they increase in higher stage colon cancers that often metastasize to the liver. Carcinoembryonic antigen (CEA) and CA19-9 (colon cancer tumor markers) may rise before symptoms or these laboratory tests are evident. (2) The use of CEA and CA19-9 levels in conjunction with radiologic techniques after colorectal cancer resection has been performed has been recently reviewed. (17)

Stool screening of DNA mutations in cancer cells

Neoplastic tumors of the colon continuously release a great number of viable colonocytes. In contrast, non-neoplastic cells of the digestive tract release sparse numbers of cells that have undergone cell death (apoptosis). (18) In fact, whole colonocytes can be detected in stool and have been used as starting material to assay for typical genetic changes that are found in colon cancers. (19-20) Therefore, neoplastic cells can be assayed using reverse transcriptasepolymerase chain reaction (RT-PCR) methods to amplify and display specific genes that are overexpressed by adenomas and adenocarcinomas. (6)

Initial studies have shown that mutant DNA (K-ras) in stools of patients with cancer or large adenomas can be detected, and agreement between mutations detected in stool samples and tumor presence has been high. (6) Drawbacks in assaying for the k-ras gene mutation include its expression by fewer than half of large adenomas and cancers and its expression by non-neoplastic sources. (6)

These initial studies have made it obvious that detection of single mutations in DNA found in colorectal cancer limits the sensitivity and specificity. This has led to the search for other mutations found in colorectal cancer that may be better indicators. For example, recent studies have assayed for the adenomatous polyposis coil (APC) gene, which is associated with the initiation of colorectal neoplasia. (21, 22) Unlike other identified mutations such as p53 that are found in later stages of disease, this gene is present early in disease. Also, APC has the advantage of not usually being expressed in non-neoplastic lesions. Stool samples from 28 patients with non-metastatic colorectal tumors and 28 control patients were investigated for APC mutations. (22) APC mutations were found in 57 percent of patients with neoplasia and in none of the 28 controls, indicating its potential use as a screening method for early, potentially curable, colorectal cancers. (22)

Since colorectal cancer has such genetic heterogeneity, other investigators have developed assays that detect multiple mutations including both APC and K-ras. (23) A multicomponent prototype test that detects mutations on K-ras; APC, p53, Bat-26 (a microsatellite instability marker) genes, and "long" DNA has undergone preliminary clinical testing. (23) The "long" DNA represents DNA of nonapoptotic colonocytes characteristic of cancer cells exfoliated from neoplasms, but not normal apoptotic colonocytes. (18, 24, 25) In normal apoptotic colonocytes, DNA is cleaved by endonucleases into short fragments of approximately 180 bp to 200 bp, making "long" DNA a good marker for neoplasia.

A blinded study using this assay was performed on samples from 22 patients with colorectal cancer, 11 patients with large adenomatous polyps, and 28 patients with normal colons. (6) The sensitivity was 91 percent for cancer and 82 percent for large adenomas, with a specificity of 93 percent found throughout the entire colon, far exceeding current screening methods. When K-ras was excluded in the panel, the sensitivity for cancer did not change, but was slightly decreased for large adenomas (73 percent). The most useful of the markers was the "long" DNA. (6) Although these studies are promising, larger clinical trials are necessary to evaluate the efficacy of these tests in broader populations.

Conclusions

The American Cancer Society has established guidelines for screening average and high-risk patients. (26) These guidelines include the use of FOBT, colonoscopy, and double contrast barium enema. Current guidelines are often expensive and uncomfortable. Greater participation in screening is necessary to impact colorectal mortality. Current participation rates are less than 30 percent in both genders, compared to screening for breast and cervical cancer that have rates of 70 percent to 80 percent. (26) Participation could be enhanced by the use of DNA testing for gene mutations associated with colorectal cancer. They are less uncomfortable, less expensive, and offer greater accuracy. However, larger clinical studies need to be performed to corroborate initial results.

Table 1

Types of polyps found in the colon

Types                                      Malignant **

Neoplastic * (10 percent of all polyps)
[] Adenomas                                Yes
[] Adenocarcinoma                          Malignant **

Non-neoplastic (90 percent of all polyps)
[] Hyperplastic                            No

[] Inflammatory                            No
(Pseudopolyps)
[] Hamartomatous                           No

[] Lymphoid                                No

Types                                      Potential associations

Neoplastic * (10 percent of all polyps)
[] Adenomas                                Adenocarcinoma
[] Adenocarcinoma

Non-neoplastic (90 percent of all polyps)
[] Hyperplastic                            More common in patients in
                                           their 50's and 60's
[] Inflammatory                            Inflammatory bowel disease
(Pseudopolyps)
[] Hamartomatous                           Inherited polyposis
                                           syndromes
[] Lymphoid                                Often an incidental finding

* Neoplastic: Abnormal growth of tissue which may be cancerous.

** Malignant: Characterized by progressive and uncontrolled growth
(especially of a tumore) and tending to worsen and result in death.
Table 2

Classification and characteristics of adenomas

Type                    Tubular         Villous       Tubulovillous
Characteristics *  Small (often <1cm)  Large and  Mixture of tubular
                    and pedunculated    sessile       and villous

* Cancer is unlikely to be present in adenomas <1 cm diameter. The risk
of cancer is increased in adenomas with villous architecture (often >4
cm) and adenomas with advanced dysplasia (morphology indicating
progression to cancer).
Table 3

Gene/protein mutations in colorectal cancer

Gene    Associated Function of Protein Gene
        Encodes

APC     Plays a role in cell adhesion and
        migration. In its defective form, high
        levels of the protein b-catenin
        accumulate which accelerates cell
        growth.

K-ras   Plays a role in signal transduction from
        activated transmembrance receptors to
        downstream protein kinases, including
        ras and mitogen activated protein
        kinase (MAPK). (27,28)

p53     DNA damage leads to increased production
        aid activation of p53. In response,
        activated p53 initiates cell cycle
        arrest and DNA repair. If repair is
        unsuccessful, the p53 helps initiate
        apoptosis or cell death. Mutations in
        the p53 gene lead to decreased or
        absent p53 function. Thus, damaged
        cells do not repair their DNA and are
        free to divide, producing more damaged
        cells. (29)

Bat-26  Bat-26 is a marker of microsatellite
        instability marker (MSI). Throughout
        the genome, tandem, repeats of 1 to 6
        nucleotides (known as satellites) are
        scattered and fixed in number. With
        errors in DNA repair, the satellites
        expand. These expanded satellites can
        be found in tumor cells, indicating a
        defective mismatch repair. Ten percent
        to 15 percent of colon cancers have
        MSI. (30)
Table 4

Risk stratification of colorectal cancer

High Risk

* Personal history of colorectal cancer
* Inflammatory bowel disease
* Sporadic adenomas
* Breast cancer
* Ovarian cancer
* Endometrial cancer
* Family history of colorectal cancer
* Family history of a hereditary
   non-polyposis cancer syndrome
* Family history of Gardner's syndrome
Average Risk
* Age [greater than or equal to]50 (Asymptomatic)

References

(1.) Landis, S. H., T. Murray, S. Bolden, and P. A. Wingo. Cancer statistics, 1998. CA Cancer J Clin 48: 6-29., 1998.

(2.) Levin, B. Neoplasms of the large and small intestines. In: Cecil Textbook of Medicine (21st ed.), edited by G. L Cecil RL, Bennett JC. Philadelphia: W. B. Saunders, 2000, p. 741-790.

(3.) Crawford, J. The Gastrointestinal Tract. In: Robbins Pathologic Basis of Disease (6th ed.), edited by Cotran. St Louis: W.B. Saunders Company. 1999, p. 779-844.

(4.) Fearon, E. R., and B. Vogelstein. A genetic model for colorectal tumorigenesis. Cell 61: 759-67., 1990.

(5.) Henderson, A. R., Rinker, A.D. Gastric, Pancreatic, and Intestinal Function. In: Tietz Textbook of Clinical Chemistry (3rd ed.), edited by A. E. Burtis CA. Philadelphia: W.B. Saunders Co., 1999, p. . 1271-1327.

(6.) Ahlquist, D. A., and A. P. Shuber. Stool screening for colorectal cancer: evolution from occult blood to molecular markers. Clin Chim Acta 315: 157-68., 2002.

(7.) Hardcastle, J. D., J. O. Chamberlain, M. H. Robinson, S. M. Moss, S. S. Amar, T. W. Balfour, P. D. James, and C. M. Mangham. Randomised controlled trial of faecal-occult-bleed screening for colorectal cancer. Lancet 348: 1472-7., 1996.

(8.) Mandel, J. S., J. H. Bond, T. R. Church, D. C. Snover, G. M. Bradley, L. M. Schuman, and F. Ederer. Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota Colon Cancer Control Study. N Engl J Med 328: 1365-71., 1993.

(9.) Kronborg, O., C. Fenger, J. Olsen, O. D. Jorgensen, and 0. Sondergaard. Randomised study of screening for colorectal cancer with faecal -occult-blood test Lancer 348: 1467-71., 1996.

(10.) Ahlquist, D. A. Focal occult blood testing for colorectal cancer. Can we afford to do this? Gastroenterol Clin North Am 26: 41-59., 1997.

(11.) Meyer, R. J. Gastrointestinal tract cancer. In: Harrison's Principles of Internal Medicine (14th ed.), edited by B. E. Fauci AS. Isselbacher KJ, Wilson JO, Martin JB, Kasper DL. Hauser SL Longe DL New York: McGraw-Hill, 1998. p. 568-578.

(12.) Bawley, O. W., Kramer, B.S. Prevention and early detection of cancer. In: Harrison's Principles of Internal Medicine (14th ed.), edited by B. E. Fauci AS, lsselbacher KJ, Wilson JD, Martin JB, Kasper DL, Hauser SL, Lange DL New York: McGraw-Hill, 1998, p. 499-509.

(13.) Johnson, C. D., and D. A. Ahlquist Computed tomography colonography (virtual colonosopy): a new method far colorectal screening. Gut 44: 201-9., 1999.

(14.) Hare. A. K., O. Johnson, J. E. Reed, D. A. Ahlquist, H. Nelson, R. L. Ehman, C. H. McCollough, and D. M. UT Ilstrup. Detection of colorectal polyps by competed tomographic colography: feasibility of a novel technique. Gastroenterology 10: 284-98., 1996.

(15.) Fenlon, H. M., D. P. R Nunos, P. C. Scbrey, 3rd, M. A. Barish, P. D. Clarke, and J. T. Ferrucci. A comparison of virtual and conventional colonoscopy for the detection of colorectal polyps. N Engl J Med 341: 1496-503., 1999.

(16.) Rex, D. K., D. Vining, and K. K. Kopecky. An initial experience with scroeoing for colon polyps using spiral CT with and without CT colography (virtual colonoscopy). Gastrointest J., M Endosc 58: 369-13., 1899.

(17.) Watine, J., M. Miedouge, and B. Friedberg. Carcinoembryonic antigen as an independent prognostic factor of recurrence and survival in patients resected for colorectal liner metastases: a systematic review. Dis Colon Rectum 44: 1791-9., 2881.

(18.) Ahlquist, D. A., J. J. Harrington, L. J. Burgart, and P. C. Roche. Morphometric analysis of the "mucocellular layer" overlying colorectal cancer and normal mucosa: relevance to exfoliation and stool screening. Hum Pathol 31: 51-7., 2000.

(19.) Albaugh, G. P., V. Iyengar, A. Lohani, M. Malayeri, S. Bala, and P. P. Nair. Isolation of exfoliated colonic epithelial cells, a novel, non-invasiva approach to the study of cellular markers. Int J Cancer 52: 347-50., 1992.

(20.) Loktionov, A., I. K. O'Neill, K. R. Silvester, J. H. Cummings, S. J. Middleton, and R. Miller. Quantitation of DNA from exfoliated colonocytes isolated from human stool surface as a novel noninvasive screening test for colorectal cancer. Clin Cancer Res 4: 337-42., 1998.

(21.) Tsao, J., and D. Shibata. Further evidence that one of the earliest alterations in colorectal carcinogenesis involves APC. Am J Pathol 145: 531-4., 1994.

(22.) Traverse, G., A. Shuber, B. Lenin, C. Johnson, L. Olsson, D. J. Schoetz, Jr., S. R. Hamilton, K. Boynton, K. W. Kinzler, and B. Vogelstein. Detection of APC mutations in fecal DNA from patients with colorectal tumors. N Engl J Med 346: 311-28., 2002.

(23.) Ahlquist, D. A., J. E. Skoletsky, K. A. Boynton, J. J. Harrington D. W. Mahoney, W. E. Pierceall, S. N. Thibodeau, and A. P. Shuber. Colorectal cancer screening by detection of altered human DNA in stool: feasibility of a multitarget assay panel. Gastroenterology 119: 1219-27., 2000.

(24.) Bedi, A., P. J. Pasricha, A. J. Akhtar, J. P. Barber, G. C. Bedi, F. M. Giardiello, B. A. Zehnbauer, S. R. Hamilton, and R. J. Jones. Inhibition of apoptosis during development of colorectal cancer. Cancer Res 55: 1811-6., 1995.

(25.) Strater, J., U. Wedding, T. F. Barth, K. Koretz, C. Elsing, and P. Moller. Rapid onset of apoptosis in vitro follows disruption of beta 1- integrin/matrix interactions in human colonic crypt cells. Gastroenterology 110: 1776-84., 1996.

(26.) Smith, R. A., A. C. von Eschenbach, R. Wender, B. Levin, T. Byers, D. Rothenberger, D. Brooks, W. Creasman, C. Cohen, C. Runowicz, D. Saslow, V. Cokkinides, and H. Eyre. American Cancer Society guidelines for the early detection of cancer: update of early detection guidelines for prostate, colorectal, and endometrial cancers. Also: update 2001 -- testing for early lung cancer detection. CA Cancer J Clin 51: 38-75: quiz 77-80., 2001.

(27.) Lowy, D. R., and B. M. Willumsen. Function and regulation of ras. Annu Rev Biochem 62: 851-91, 1993.

(28.) Bourne, H. R., D. A. Sanders, and F. McCormick. The GTPase superfamily: a conserved switch for diverse cell functions. Nature 348: 129-32., 1990.

(29.) Bargonetti, J., and J. J. Manfredi. Multiple roles of the tumor suppressor p53. Curr Opin Oncol 14: 86-91., 2002.

(30.) Slattery, M. L., K. Anderson, K. Curtin, K. N. Ma, D. Schaffer. and W. Samowitz. Dietary intake and microsatellite instability in colon tumors. Int J Cancer 93: 601-7., 2001.

RELATED ARTICLE: LEARNING OBJECTIVES

Upon completion of this article the reader will be able to:

1. Describe the development of colorectal tumors.

2. Discuss symptoms associated with colorectal cancer.

3. List three things that will place an individual at high risk for colorectal cancer.

4. Discuss the fecal occult blood test and its limitations.

5. Differentiate between flexible sigmoidoscopy and colonoscopy.

6. Discuss the use of carcinoembryonic antigen (CEA) in the management of colorectal cancer.