Colorectal Cancer Screening

Burt, Randall W


Curr Opin Gastroenterol. 2010;26(5):466-470. 

In This Article

Issues with Present Tests and Approaches

There are a number of issues that have recently arisen concerning the efficacy of present screening tests and how they are applied. The following sections will present and discuss these issues.

Guaiac-based Fecal Occult Blood Testing

FOBT is done most commonly with a guaiac-based system that reacts to peroxidases, including hemoglobin. Large series have demonstrated decreased colon cancer mortality from annual or biannual testing with double window, three card tests. The decrease in mortality, however, is not large and in the range of 15–30%. Successful testing depends on a number of compliance factors including the patient being on a low peroxidase diet during collection, not taking aspirin or nonsteroidal antiinflammatory drugs (NSAIDs), returning the cards within 2 weeks for development, hydration of the card before the developer is applied, annual or biannual completion of the testing, and completion of colonoscopy for positive tests. The expense of the FOBT testing is minimal, but one substantial cost in the approach is frequent referral for colonoscopy based on false-positive tests.

In many large studies, over half of patients eventually undergo colonoscopy after a number of years of study. The question thus arises as to whether colon neoplasias (polyps and cancer) are detected because of true FOBT-positive tests or simply because of serendipity.[9] Analyses of studies suggest that about half of neoplasias are the result of true-positive FOBT tests, whereas the other half are detected because of frequent colonoscopies from false-positive tests. There is virtually no sensitivity for small adenomatous polyps (aside from serendipity), so FOBT cannot be regarded as preventive, but instead an early diagnostic test.

Fecal Immunochemical Test

Some improvement in specificity and possibly sensitivity is achieved with fecal immunochemical testing, which is based on an immune reaction to human hemoglobin.[10,11•,12] The need for a special diet during testing is also avoided. The test is a bit more expensive than guaiac-based tests, but a single annual test may be adequate (this is being evaluated). There is higher sensitivity for advanced adenomas compared to FOBT. As fewer false-positive tests are likely, savings would come mainly from fewer false-positive referrals for colonoscopy.

Fecal DNA Test

Fecal DNA testing is a relatively new development.[13•] It is based on finding several specific tumor-related DNA changes in cells shed from colonic neoplastic lesions into the bowel contents. Initial difficulty in separating human from bacterial DNA has been mostly overcome and the test is now commercially available. The test appears to be somewhat more sensitive than guaiac-based FOBT but there remains uncertainty if adenomas, even advanced adenomas, can be reliably detected. It is also uncertain how frequently the test should be done to adequately screen for colon cancer. Wide application has not yet been achieved, despite recommendation for use by the US Multi-Society Task Force on Colorectal Cancer, probably because of the considerable cost of the test.


Sigmoidoscopy has been shown to be highly effective by several case-control studies. Cancer in the distal colon was shown to be decreased 50% or more with these studies. There was, however, no effect on proximal colonic malignancy incidence. Some decrease in mortality extended to 10 years. Recommendations for colonoscopy as a screening tool are based in part on the results of these sigmoidoscopy investigations.

Sigmoidoscopy is the best investigationally supported luminal screening tool. The problem will always remain, however, that if endoscopy of the distal colon is so successful, then why not examine the entire colon. The combination of every-5-year sigmoidoscopy with annual FOBT is better than either test used alone.[14••]


Colonoscopy is considered the gold standard of colon cancer screening.[15] Evidence for its efficacy, however, is indirect, based on sigmoidoscopy results and improved cancer outcomes in many follow-up studies. A number of problematic issues have arisen concerning colonoscopy in the past several years, however.[16•] First is cost and preparation. The cost of screening colonoscopy, although now supported by Medicare and most insurance companies, is still substantial and limiting for many patients, even when only co-pays are required. Uncomfortable preparation may also be an issue of noncompliance.

Oversensitivity Oversensitivity is also perhaps an issue with colonoscopy as millimeter-sized lesions can now be detected.[17] This results in considerable follow-up that may not be necessary. Separation of 'advanced' adenomas from other adenomas has helped this issue.[18] Advanced adenomas are defined as those at least 1 cm, those containing villous or advanced histology and those that occur in the setting of three or more adenomas. Advanced adenomas have a higher correlation with adenoma recurrence and interval cancer but account for only a small fraction of adenomas found.

Effectiveness The most problematic issue with colonoscopy comes from recent investigations that question its effectiveness in preventing cancer. Several years ago interval cancer occurrence in polyp prevention studies was examined and found to be no different than expected cancer rates.[19] The comparison was not particularly valid as it compared a group with previous adenomatous polyps with an average-risk population. Nonetheless, the finding was in stark contrast to the national polyp study,[20] the original study that found follow-up colonoscopy in adenomatous polyp patients markedly reduced colon cancer incidence.

The study questioning colon cancer prevention success also examined factors associated with occurrence of colon cancer. The authors reported a number of issues likely related to decreased sensitivity including inadequate colon preparation before colonoscopy and cancers that were microsatellite unstable. Physician factors included too rapid withdrawal time, an exam done by an internist or family physician, and colonoscopy done in the office as opposed to endoscopy center. Other factors possibly associated with missed cancers included advanced patient age, the presence of diverticular disease, proximal colonic cancer and 'flat' adenomas. A large study then examined colonoscopy withdrawal times.[21] A withdrawal time of 16.8 min compared to 3.1 min resulted in more than three times the number of patients found to have adenomas. A withdrawal time of 6 min or greater compared to one less than 6 min resulted in more than double the number of adenomas, and double the numbers of advanced adenomas found.

Finally, one study[22] found a substantial decrease in colon cancer mortality following colon cancer screening in the distal colon, but not the proximal colon. Suggested reasons for this finding include differences in biologic behavior of proximal colonic lesions, flat adenomas and sessile serrated polyps that occur more commonly in the proximal colon, and too rapid withdrawal times.[15] Quality studies are now recommended as a method of developing better quality of colonoscopy.

Flat Adenomas Flat adenomas are a potential reason for neoplastic lesions missed at colonoscopy.[23] They are defined as an adenoma whose diameter is at least twice its height. They account for almost 10% of colonic neoplasms but are 10 times more likely to contain carcinoma than pedunculated adenomas, although the majority of cancer risk is found in 'depressed' lesions. It is likely that flat or sessile lesions are missed more frequently than pedunculated adenomas in view of their anatomy.

Sessile Serrated Polyps/sessile Serrated Adenomas A newly recognized lesion in the last several years is the proximal colonic hyperplastic polyp.[24] These polyps appear to have malignant potential similar to adenomas. They are often very sessile and covered with mucus because of the abundance of mucus secreting cells in the polyps. When the mucus is cleared off, the polyps are often very flat with sometimes-difficult-to-discern edges. The polyps are histologically quite different from typical hyperplastic polyps of the distal colon. They have a serrated appearance in the upper portions of the crypt and cells near the base have an adenoma-like appearance. There is now general agreement that these lesions should be called sessile serrated polyps or sessile serrated adenomas. The two terms are equivalent. It is likely that these lesions are another reason that colonoscopy of the proximal colon has not been as successful at preventing colon cancer as distal colonic examination.

Summary of Colonoscopy Issues The issues discussed above have led to present work to develop quality standards for colonoscopy. Withdrawal times should be lengthened to 6–8 min or more. Similarly, acceptable parameters for polyp detection, flat polyp detection and sessile serrated polyp detection should be established.[25••] Measurable parameters for these issues need to be developed and then be a part of training and practice certification. It then needs to be determined if adherence to quality guidelines and measurements leads to more successful cancer prevention.

Virtual Colonoscopy

Computerized axial tomographic (CAT) colography, commonly called virtual colonoscopy, is a promising approach to colon cancer screening.[26] A high-resolution abdominal CAT scan is processed by software to allow two-dimensional, three-dimensional and 'fly through' images of the colon. Centers that are highly specialized in this technology demonstrate sensitivity with virtual colonoscopy for colonic lesions at least 1 cm in diameter of 90%, which is near-equivalent to that of colonoscopy.[27] The sensitivity for lesions from 5 to 9 mm drops to 70% and for those below 5 mm, the sensitivity is less than 50%.

There are several issues keeping virtual colonoscopy from being a mainstream approach to colon cancer screening. The preparation is the same as for optical colonoscopy, which is the major complaint of many patients needing large bowel examination. A rectal tube with colonic inflation is also necessary and sedation is not part of the procedure. Thus abdominal discomfort similar to that of barium enema is often experienced. Optimal virtual colonoscopy examination requires a multidetector spiral CT and appropriate software. There is also a definite learning curve for interpretation after radiologic equipment is obtained. The cost is similar to screening colonoscopy. Additionally, reading often takes 15–20 min, whereas reimbursement is similar to abdominal CT, which can often be read in 5 min or less. More importantly, most insurance companies do not reimburse for virtual colonoscopy and rulings within the past year indicate that Medicare will likewise not support the procedure.

Several technical issues require further development. The time required for reading virtual colonoscopy could be assisted by computerized reading with radiologist review of lesions. This is under development. It is hoped that the technology for virtual colonoscopy will eventually allow 'electronic cleansing.' This means digital subtraction of bowel contents, which would preclude the necessity of actual bowel preparation. Patients would obviously see such a development as very attractive. Finally, a logistical issue is that colonoscopy must be later performed with a second bowel preparation when lesions are found. Studies demonstrate that lesions that must be addressed by optical colonoscopy are found in 5–30% of those examined. This is clearly a major inconvenience. To address this issue some centers offer colonoscopy directly after a 'positive' virtual colonoscopy, thus avoiding a repeat appointment and preparation.

As virtual colonoscopy misses many lesions less than 1 cm in diameter, a few cancers may be missed, but more importantly many small polyps that might progress to cancer would be left. Furthermore there is debate as to whether all polyps from 5 to 9 mm should have immediate colonoscopy vs. a 3-year virtual colonoscopy follow-up. One study examined immediate colonoscopy following virtual colonoscopy for polyp removal, vs. a 3-year virtual colonoscopy follow-up in persons with 6–9 mm polyps found by virtual colonoscopy. It was estimated that 33% of high-risk adenomas would have been missed.[28] A modeling study also examined immediate colonoscopy following virtual colonoscopy for polyp removal, vs. a 3-year virtual colonoscopy follow-up in persons with 6–9 mm polyps.[29] The study modeled a cohort of 100 000 persons with polyps of that size. The immediate colonoscopy group was predicted to have or develop 39 colon cancers and experience 14 colon cancer deaths after 3 years. The group given a 3-year virtual colonoscopy follow-up to re-examine the polyps was predicted to have or develop 773 colon cancers and experience 79 colon cancer deaths. Thus a 3-year virtual colonoscopy follow-up in those with 6–9 mm polyps rather than immediate colonoscopy would be predicted to result in a substantial number of unnecessary colon cancers and deaths.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: