Navigation

Colon Cancer Images

Colorectal cancer (CRC) is the third leading cause of cancer and the second leading cause of cancer mortality in the United States, responsible for almost 150,000 cases and 57,000 deaths in 2003.1 Average-risk individuals have a 5%-6% lifetime risk of developing CRC.2 Survival is directly related to the stage of cancer at the time of diagnosis. The 5-year survival rate is 92% with localized disease; less than 60% with spread through the bowel wall; and 7% in the presence of distant metastases.2 There is strong evidence that population-based screening can reduce the mortality from CRC by detection of CRC at early stages.2

Progression from normal colonic mucosa to adenomatous polyp, then early invasive carcinoma into symptomatic CRC occurs over a period of years.3 This orderly progression offers an excellent opportunity to prevent CRC through the detection and removal of adenomatous polyps, and to decrease CRC mortality through its detection at early stages.

Evidence-based guidelines strongly recommend population-based screening for CRC.2, 4, 5 This paper will address the role of colonoscopy, and the promising new tests of CT colonography or "Virtual colonoscopy," and fecal DNA testing, in screening for CRC.

COLONOSCOPY

Colonoscopy permits visualization of the entire colon directly, detection and removal of polyps, and biopsy of CRC anywhere within the colon and rectum. It requires adequate bowel preparation using laxatives or large volumes of an oral cathartic solution. IV sedation minimizes pain and discomfort. The endoscope is maneuvered within the bowel and bowel distention with air is required for adequate evaluation of the colonic mucosa. Most polyps can be removed by electrocautery techniques.

Almost 40% of CRC arise proximal to the splenic flexure.6 Studies in patients with CRC proximal to the splenic flexure have found that a least two thirds of these patients have no CRC or adenomatous polyp distal to the splenic flexure.7, 8 The cecum is reached in 80% to 95% of procedures.9 Incomplete colonoscopies require either a repeat colonoscopy or supplemental barium enema to clear the proximal colon and cecum.

There are no studies evaluating whether screening colonoscopy alone reduces the incidence or mortality from CRC in the average risk patient population.2 There is, however, considerable indirect evidence of this. Data from the clinical trials of Fecal Occult Blood Test (FOBT) screening demonstrated reduced CRC mortality, with colonoscopy and polypectomy representing an integral part of the evaluation of a positive FOBT.10, 11 There is direct evidence that screening sigmoidoscopy reduces CRC mortality.12, 13 These studies emphasize the effectiveness of endoscopy and polypectomy, and can be extrapolated to establish the effectiveness of colonoscopy with polypectomy, which offers more complete examination of the colon.14 The National Polyp Study demonstrated that patients who had colonoscopy with complete polypectomy had a 76% to 90% reduction in the expected rates of CRC over the subsequent 6 years and there were no deaths from CRC.15 This reduction in expected rates and mortality of CRC occurred in a population with adenomatous polyps, a population at much higher risk for developing CRC compared to a screening population.

Colonoscopy is often referred to as the gold standard for the presence or absence of polyps and CRC.9 Colonoscopy, however, even in experienced hands, can miss lesions. Hixson et al.16 evaluated the miss rate of 2 colonoscopists by performing 2 consecutive same-day colonoscopies on 90 patients. They found a miss rate of 15% for polyps 1 cm. Rex et al.17 performed 2 consecutive same-day colonoscopies on 183 patients. The overall miss rate for adenomas was 24%, 27% for adenomas 1 cm. Patients with 2 or more adenomas at the first examination were more likely than patients with no or 1 adenoma at the first examination to have 1 or more adenomas at the second examination.

Adenoma miss rates correlated with withdrawal technique.18 Examinations of the proximal sides of flexures, folds, and valves; cleaning and suctioning; adequate distention; and adequate exam time were important factors in achieving a lower adenomatous miss rate.

In a recent study of CT colonography (CTC), sensitivity and specificity of colonoscopy were assessed.19 Same-day CTC and conventional colonoscopy were performed. CTC was the initial test, and lesions were reported for each segment of the colon. Colonoscopy was then performed, the colonoscopist unaware of the findings of CTC. After completing the exam of a given colonie segment by colonoscopy, the results of the CTC for that segment were revealed, referred to as segmentai unblinding. If CTC revealed pathology, but that pathology was not present on colonoscopy, there was a second colonoscopic examination of that segment performed. If pathology was identified on the second examination, the colonoscopy was considered a false negative for that segment. The sensitivity of CTC for adenomatous polyps was 93.8% for polyps> 1 cm, with 96% specificity, compared to 87.5% sensitivity with colonoscopy. Most of the clinically significant adenomas missed prospectively on conventional colonoscopy were located on a fold, especially on the backside of a fold, or near the anal verge.20

Colonoscopy is associated with the greatest risk of complications among the screening tests. Complications include perforation, hemorrhage, respiratory depression due to sedation, arrhythmias, ileus and nosocomial infection. Approximately 1/1000 patients have perforation, 3/1000 have major hemorrhage, and 1-3/10,000 die as a result of the procedure.9 Serious disorders of sodium balance were reported after use of a Polyethylene Glycol colon preparation, including seizures and death.21 Cases of hyponatremia with encephalopathy have been reported after use of Visicol tablets.22, 23

Expert panels recommend that colonoscopy screening be performed at 10-year intervals if the initial exam is negative.2, 19, 18 This recommendation is based on the dwell time from the development of adenomatous polyps to transformation into CRC, estimated to be-10 years,15 and on a case control study of screening rigid sigmoidoscopy which found a protective effect from death due to CRC for up to 10 years in that segment of colon examined.12 However, the interval at which screening colonoscopy should be performed in average-risk persons has not been determined by observational studies. The longest reported interval between an initial normal colonoscopy and a second colonoscopy in a group of asymptomatic average risk persons 50 or older is 5.5 years.24 Cancer incidence in this population was 0% at 5.5 years, and the incidence of adenomas with advanced pathology, defined as those> 1 cm in size, or containing villous tissue or high grade dysplasia, was

In a recent study by Schoen et al.25 6 of 1292 patients had CRC discovered in the distal colon 3 years after a baseline sigmoidoscopy; 72 patients had an advanced adenoma. Some of these lesions were likely missed on the baseline study because of the known inherent miss rate of colonoscopy. However, up to 15% of CRCs have microsatellite instability, associated with mutations of MMR genes. These patients may develop polyps which evolve more rapidly to CRC.26

Colonoscopy is one of the recommended screening tests for CRC for persons age 50 and older who are at average risk of developing CRC.2, 4 It is the preferred screening test for persons age 50 and older who are at average risk for developing CRC as recommended by the American College of Gastroenterology.14 It is the recommended screening test for people at an increased risk for developing CRC, including those with a family history of CRC or adenomatous polyps.2,4,14

Surveillance with colonoscopy is recommended for patients who are at increased risk because they have a prior history of CRC or prior adenomatous polyps, or have a disease that predisposes them to CRC, such as inflammatory bowel disease.2, 4, 14

Colonoscopy has been found to be cost-effective in comparison to other CRC screening strategies.27

CT COLONGRAPHY (CTC). OR "VIRTUAL COLONOSCOPY'

CTC is a new, non-invasive method for examining the colon. Its non-invasiveness, speed of performance, lack of requirement for patient sedation, and high predictive values makes it attractive.28

The technique, as described by Bruzzi et al.,28 involves the rapid acquisition of thin section CT slices of the prepared colon using a helical CT scanner, and the subsequent manipulation of data to produce 2 - Dimensional axial or 3 - Dimensional images that resemble an endoscopie view of the bowel lumen. Bowel cleansing regimens are commonly used. Residual fecal material can result in both falsepositive and false-negative findings. Fecal tagging methods allow CTC evaluation in the setting of a minimally prepared or unprepared colon. Fecal tagging involves the administration of small amounts of barium or watersoluble contrast material 1 day prior to CTC. Fecal material is, therefore, labeled with high- density contrast and identified as such rather than as polyps or mass lesions.

Adequate colonie distention is required, accomplished by placement of a soft-tipped enema tube within the rectum, followed by air or CO2 insufflation to the maximal limit tolerated by the patient. The patient is then scanned in both the supine and prone positions to redistribute the gas into segments of the colon that may have been collapsed.

Fenlon et al.29 studied 100 high-risk patients comparing CTC with conventional colonoscopy. CTC detected 3 of 3 cancers and 20 of 22 polyps> 1cm, a 91% sensitivity. There were 19 false- positives. Sensitivities for polyps 6-9 mm and for polyps

Pickhardt et al.19 performed a multicenter trial involving 1201 average risk patients who underwent same day CTC and conventional colonoscopy. Stool tagging and digital subtraction, 3-D endoluminal imaging, and segmentai unblinding of CTC results at colonoscopy, were performed. The sensitivity of CTC for adenomatous polyps was 93.8% for polyps at least 1 cm in diameter, 93.9% for polyps at least 8 mm in diameter, and 88.7% for polyps at least 6 mm in diameter. The sensitivity of colonoscopy for adenomatous polyps was 87.5%, 91.5%, and 92.3% for the 3 polyp sizes, respectively. Specificity of CTC for adenomatous polyps was 96.0% for polyps at least 1cm, 92.2% for polyps at least 8 mm, and 79.6% for polyps at least 6mm in diameter. Two polyps were malignant and both were detected on CTC. One of those malignant polyps was identified on colonoscopy only after the results of CTC were revealed on segmentai unblinding. As well, unsuspected extracolonic malignancies were identified in 4.5% by CT findings. Pickhardt et al. demonstrated that CTC compared favorably to conventional colonoscopy in detecting colorectal neoplasms in asymptomatic average risk adults.

Cotton et al.30 performed a multicenter study of 615 predominantly average risk patients age 50 and older referred for routine colonoscopy to assess the accuracy of CTC compared with colonoscopy. There was no stool tagging or use of 3-D rendering as a primary review. Findings at colonoscopy were reported before and after segmentai unblinding to the CTC results. The sensitivity of CTC for detecting patients with one or more lesions at least 6 mm was 39%; for lesions at least 1 cm, it was 55%. This was significantly lower when compared to conventional colonoscopy, with sensitivities of 99% and 100%, respectively. CTC missed 2 of 8 CRCs.

There is, therefore, wide variability in the sensitivity and specificity of CTC. Several factors contribute to this inconsistency, as noted by Van Dam et al." First, different technologies have been used in various studies. The data suggest that multi-detector scanners allow more accurate detection of smaller lesions than single slice scanners. Hardware and software used to analyze images varies widely, as does bowel preparation methods, and the presence or absence of stool tagging. Finally, there is variation in the study populations' relative risk for neoplasia

CTC potentially offers the advantage of identifying CR neoplasms that may not be adequately identified by conventional colonoscopy. It has the capacity of imaging the colon proximal to obstructing lesions and it can serve to complete examination of the colon after an incomplete colonoscopy.

Levin et al.32 described the limitations of CTC:

1) False positive readings in approximately 15%

2) An unknown ability to detect flat adenomas, which may be a more aggressive from of neoplasia than the typical adenomatous polyp;

3) Lack of performance and training standards;

4) CTC is a diagnostic tool only, with no capability of polyp removal during the procedure;

5) The concern for cumulative radiation doses with serial screening examinations;

6) The cost of CTC may be higher than that for conventional colonoscopy.

Van Dam et al.31 identified fundamental questions that remain unanswered. Is there a minimum polyp size detectable by CTC for which patients should be referred for endoscopic polypectomy? What polyp size, if any, can remain in situ and undergo CTC surveillance rather that immediate polypectomy?

FECAL DNA TESTING

Our understanding of the molecular biology of colorectal carcinogenesis forms the basis for detecting CRC by detection of different mutations in DNA exfoliated into the stool. Neoplasm-specific DNA mutations are released into the bowel lumen continuously via exfoliation, rather than intermittently via bleeding. DNA is stable in stool and amplification techniques, such as polymerase chain reaction allow its detection in minute amounts. The stool analysis involves the collection of an entire bowel movement. Dietary restrictions are not required before testing.

Sporadic CRC is divided into those demonstrating chromosomal instability, and those with an impaired mismatch repair (MMR) mechanism.33 CRCs with chromosomal instability are characterized by the progressive accumulation of mutations in several genes, including the tumor suppressor genes, APC and k-Ras, and the p-53 oncogene. These constitute approximately 85% of sporadic CRCs.33 Both pathways are associated with characteristic DNA alterations, which may be detected in stool.

Since no single mutation has been found that is expressed in all CRCs, panels of various markers are required.33 A commercially available panel from EXACT Sciences Corporation includes 15 specific point mutations on APC, k-RAS, and p-53; gene mutations on BAT-26, a marker of microsatellite instability; and long DNA, a marker of DNA not degraded by apoptosis.32 If any component of the panel is positive, the result is characterized as positive.

Four studies 34,35,36,37 have reported using the multi-target assay panel. Overall, 99 of the 146 patients with CRC were successfully detected by stool DNA analysis, a 67.8% sensitivity for CRC. A total of 240 patients without colonoscopic abnormalities demonstrated the presence of mutations in only 10, for a specificity of 95.8%. Brand et al.35 demonstrated no advantage to more than 1 sample per patient for stool DNA testing. Syngal et al.37 analyzed fecal DNA from 56 patients whose CRCs had been diagnosed at colonoscopy. A 62% sensitivity for invasive cancer was reported. Stools were also obtained following surgical resection of the primary CRC. By 6 months post- operatively, the previously found stool DNA mutations were no longer detectable.

The results in these studies in patients with adenomas> 1 cm in size have been quite variable, ranging from a high of 73% 34 to a low of 27%.37

Dong, et al.38 were able to detect the majority of CRCs by analyzing stool DNA for just 3 genetic markers- p -53, BAT -26, and k-Ras. 51 patients who had CRC diagnosed at colonoscopy were evaluated. Prior to surgery, stool samples were collected and matched with each patient's tumor tissue. The stool was analyzed for the 3 genetic markers. Thirty patients demonstrated p-53 gene mutations in tumor DNA, and identical mutations were found in their stools. In 3 patients, mutations at the BAT-26 locus was identified in tumor, and also in each of the patient's stools. Nineteen patients demonstrated a k-Ras mutation in tumor tissue, identical to those detected in their stools. In no case was a mutation found in the stool that was not present in the primary tumor tissue.

The feasability of detecting APC mutations in fecal DNA was studied,39 employing a novel assay called digital protein truncation. Stool samples from 28 patients with non-metastatic CRCs, 18 patients with adenomas that were at least 1 cm, and 28 control patients without neoplasm were studied. APC mutations were identified in 17 of the 28 patients with Duke's stage B2 cancer (61%), 9 of the 18 patients with adenomas at least 1 cm (50%), and in none of the 28 control patients. They concluded that APC mutations can be detected in fecal DNA from patients with relatively early CR tumor.

Traverso, et al.40 also demonstrated the ability of fecal DNA to detect proximal CRC. Using a method for microsatellite mutation detection, 18 of 46 proximal CRCs had microsatellite alterations detected. The identical mutations were identified in the fecal DNA of 17 of these 18 cases, with a zero% false positivity among 69 individuals with normal colonoscopies, or among 19 individuals with adenomas. This demonstrates that DNA is not degraded as it passes through the length of colon.

The American Cancer Society's CRC Advisory Group concluded that questions related to the most appropriate markers for DNA detection of CRC, on the best combination of markers, and on the results of studies in populations at average risk for CRC need to be answered before DNA stool testing can be recommended as a screening test for the average risk adult.32

CONCLUSION

The emerging competitive techniques to diagnostic colonoscopy of CTC and fecal DNA testing, though promising and innovative technologies, at this time remain unproven as screening options for the average risk population of adults who are 50 or older. These technologies should be re-visited in the near future as additional data become available.

BIBLIOGRAPHY

1. Jemal A, Murray T, et al. CA Cancer J Clin 2003; 53:5-26.

2. Winawer S, Fletcher R, et al. Gastroenterol 2003: 124;544-60.

3. Bond JH. Semin Gasrointes Dis 2000; 11:176-84.

4. Pigone M, Rich M, et al., Ann intern Med 2002; 137:132-41.

5. Smith RA, von Eschenbach AC, et al. CA Cancer J Clin 2001; 51:38-75.

6. Rex DK, Rahmani EY, et al. Gastroenterol 1997; 112:17-23.

7. Shanafelt MJ, Esber EJ. Gastrointest Endose 1997; 45:AB118 (abstract).

8. Rex Dk, Chak A, et al. Gastrointest Endosc 1999; 49:727-30

9. Winawer SJ, Fletcher RH, et al. Gastroenterol 1997; 112: 594-642.

10. Mandel JS, Bond JH, et al. NEJM 1993; 328: 1365-71.

11. Mandel JS, Church TR, et al. NEJM 2000; 343:1630-7.

12. Selby JV, Friedman GD, et al. NEJM 1992; 326:653-7.

13. Newcomb PA, Norfleet RG, et al. J Nad Cancer Inst 1992; 84:1572-5.

14. Rex DK, Johnson DA, Lieberman DA, et al. Am J Gastroenlerol 2000; 95:868-77.

15. Winawer SJ, Zauber AG, et al. NEJM 1993; 329:1977-81.

16. Hixson LJ, Rennerty MB, et al. J Natl Cancer Inst 1990; 82:1769-72.

17. Rex DK, Culter CS, et al. Gastroenterol 1997; 112:24-8.

18. Rex DK Gastrointest Endosc 2000; 51:33-6.

19. Pickhardt PJ, Choi JR, et al. NEJM 2003; 349:2191-200.

20. Pickhatdt PJ, Nugent PA, et al. Ann Intern Med 2004; 141:352-9.

21. Ayus JC, Levine R, Arieff AI. Brit Med J 2003; 326:382-4.

22. Mackey AC, Shaffer D, Prizont R. NEJM 2002; 347:295-6.

23. Rose M, Jocob LS. NEJM 2002; 347:295-6.

24. Rex DL. Cummings OW, et al. Gastroenterol 1996; 111: 1178-81.

25. Schoen RE, Pinsky PF, et al. JAMA 2003; 290:41-8.

26. Lieberman D. Practical Gastroenterol 2004: 27-42.

27. Lieberman DA. Gastroenterol 1995; 109:1781-90.

28. Bruzzi J, Brennan D, Fenlon H. Current Gastroenterol Reports 2001; 3:437-45.

29. Fenlon HM, Nunes DP, et al. NEJM 1999; 341:1496-503.

30. Cotton PB, Durkalski VL, et al. JAMA 2004; 291:1713-9.

31. Van Dam J, Cotton P, et al. Gastroenterol 2004; 127:970-84.

32. Levin B, Brooks D, et al. CA Cancer J Clin 2003;53: 44-55.

33. Ross M. Practical Gastroenterol 2004; 28-34.

34. Ahlquist D, Skoletsky J, et al. Gastroenterol 2000; 119:1219-27.

35. Brand R, Schuber A, et al. Gastroenterol 2002; 122:A479.

36. Tagore KS, Lawson MJ, et al. Clin Colorectal Ca 2003; 3:47-53.

37. Syngal S, Chung D, et al. Gastroenterol 2003; 124:A5.

38. Dong SM, Traverso G, et al. J Natl Cancer Inst 2001; 93:858-65.

39. Traverse G, Schuber A. et al. NEJM 2002; 346:311-20.

40. Traverse G, Schuber A, et al. Lancet 2002; 359:403-4.

SHELDON LIDOFSKY, MD

Sheldon Lidofsky, MD, is Clinical Assistant Professor of Medicine, Brown Medical School.

CORRESPONDENCE:

Sheldon Lidofsky, MD

33 Staniford St.

Providence, RI 02905

Phone: (401) 421-8800

Fax: (401) 751-4029

Copyright Rhode Island Medical Society Mar 2005
Provided by ProQuest Information and Learning Company. All rights Reserved