Colon Cancer Screening Guidelines 2005: The Fecal Occult Blood Test Option Has Become a Better FIT

In this issue of Gastroenterology Morikawa et al1 present important data on the performance characteristics of a new fecal immunochemical test (FIT), Magstream 1000/Hem SP (Fujirebio Inc, Tokyo, Japan), in a large, average risk Japanese population. They used the Magstream 1000/Hem SP automated system (Fujirebio) and determined sensitivity, specificity, and positive predictive value by performing colonoscopy on all patients regardless of FIT result. The overall positivity rate was 5.6%. The sensitivity of the FIT for colon cancer was 66% and 20% for polyps > 1 cm. A 66% sensitivity for cancer is a marked improvement over the colonoscopy confirmed 13%–39% sensitivity reported for the most commonly used guaiac test (GT) Hemoccult II (Beckman Coulter, Inc, Fullerton, CA).2, 3 It is remarkably similar to that reported for the FIT HemeSelect (69%).4 The FIT, FlexSure OBT, has been shown to have a better sensitivity than Magstream for left-sided cancers (82%) and left-sided polyps > 1 cm (25%) when patients with both positive and negative results were sigmoidoscoped.5

The problems with the study design are dealt with well in the Discussion section of the manuscript. They are all offset by the valuable information this study offers on the performance characteristics (verified by a “gold standard” endoscopic exam) of another FIT. Before FIT development only GT were available. GT detect any peroxidase activity in the feces. Heme either as intact hemoglobin or free heme has peroxidase-like activity therefore, in the presence of heme and a developer (hydrogen peroxide), guaiac acid is oxidized producing a blue color. Heme is present in red meat, and peroxidase activity is present in fresh fruits and vegetables such as radishes, turnips, and broccoli. These foods, therefore, have the potential to produce false-positive results. Although there are several available guaiac tests only 2, Hemoccult II and Hemoccult Sensa, have been extensively evaluated in large screening populations. Accurate interpretation of results for GT requires training and supervision especially when interpreting borderline results.6, 7, 8 Results are affected by vitamin C, which inhibits the guaiac reaction.9 The person screened is required to be on a red meat and peroxidase-containing, food-free diet and to collect the stool sample in the dry state sampling the feces with a wooden stick. All these requirements limit patient acceptance and make interpretation of test results questionable except in well-run laboratories. In a group of motivated volunteers in an Australian population where red meat consumption is relatively high, a restrictive diet reduced participation by 13%.10

The fecal immunochemical tests (FIT) address many of the weaknesses of the GT. They have superior sensitivity and specificity.4, 5 The application sensitivities for cancer are good (66%–82%) but for polyps > 1 cm they are more modest; however, the use of a program of fecal occult blood test (FOBT) screening (in which screening is performed yearly or every other year) may result in a cumulative sensitivity of the program that is competitive with a program of a more sensitive test performed less frequently. Specifically, a combined program of FOBT plus sigmoidoscopy may be as effective as or more effective than a program of colonoscopy screening every 10 years11 because improved FOBT tests such as FIT are more sensitive than the widely used Hemoccult II. The suggestion of some experts that FOBT screening should be seriously considered by physicians, patients, and policy makers12 should now make the case for FOBT even stronger. The FITs efficacy in decreasing colon cancer mortality and incidence has not been studied in randomized controlled trials but, Fletcher has pointed out that if new screening tests are truly more accurate than GT (Hemoccult II), their effectiveness need not be confirmed by randomized controlled trials because Hemoccult II’s ability to save lives from colorectal cancer has already been shown.13

FITs use antibodies specific for human globin and are, unlike the GT, specific for colorectal bleeding and not affected by diet or medications. Some can be developed by automated developers and readers. This innovation allows for managing large numbers of tests in a standardized manner with excellent quality assurance. There is evidence that FIT use improves patient participation in screening for colorectal cancer.14 Some new technology FITs are able to quantify fecal hemoglobin allowing for adjustment of sensitivity, specificity, and positivity rates when screening for colorectal neoplasia.15 The developing instrument for some FITs has the ability to read a bar code on the test. This ensures accurate identification of the person screened and allows for a print out of the result, as well as a reminder print- out for future compliance. When these innovations have been perfected and tested in large asymptomatic populations, government agencies or individual health plans will be able to decide what positivity rate their budget and manpower resources can accommodate and still have good sensitivity and specificity for advanced neoplasms in an annual screening program. Australia has initiated a colon cancer screening program using InSure and Magstream 1000/ Hem SP and the United Kingdom and Canada are considering the use of FITs in their screening programs. Japan has been using FITs successfully for years.

The new and improved FIT choices are now available and reimbursable by CMS at $22 per test (includes completed test card with 2 samples and analysis). In 2004, CMS concluded that adequate evidence existed to determine that the immunoassay fecal occult blood test (FIT) is an appropriate and effective colorectal cancer screening test for detecting fecal occult blood in Medicare beneficiaries aged ≥50 years. The CMS reimbursement decision has lead to the appearance of several FITs approved by the FDA for marketing in the United States which, in addition to InSure (Enterix USA, Inc, NJ), include Instant-View (Alpha Scientific Designs, San Diego, CA), immoCARE (Care Health Technologies, Ltd, Waterbury, CT), MonoHaem (Chemicon International, Inc, Temecula, CA), and Clearview Ultra-FOB (Inverness Medical, Newton, MA). FlexSure OBT (Beckman Coulter, Palo Alto, CA) will soon reappear as Hemoccult-ICT (Beckman Coulter Inc, Primary Care Diagnostics, Los Angeles, CA). Magstream 1000/Hem SP (Fujirebio Inc, Tokyo, Japan) is marketed in Australia and Europe (Bayer Diagnostics, Tarrytown, NY) as Bayer Detect and is also likely to be marketed in the US. These tests differ in the methods used for stool handling and sampling, how (automated or laboratory technician) and where (office or laboratory) the tests are developed. Only FlexSure OBT, HemeSelect, InSure, and now MagStream1000/Hem SP have been evaluated in large numbers (thousands) of average risk patients with results published in US peer-reviewed journals.

Despite all the data cited above, clinicians, patients, and the lay press may still dismiss choices other than colonoscopy for colon cancer screening because of statements that other tests may miss 25%–65% of advanced neoplasms. I would answer as follows:

It is not realistic to believe any colorectal cancer screening test will ever detect all advanced neoplasms. As Fletcher has pointed out, clinicians should be prepared to miss some cancers because many other factors such as complications, inconvenience, discomfort, cost effectiveness, and the workforce needed to perform procedures are also in the balance when deciding which screening policies make the most sense.40 The best screening test is the one that gets done.41 The choice of screening tests should be suited to the screening situation. All the recommendations for a FOBT option in colon cancer screening guidelines were made on the basis of findings from randomized controlled trials using GT. If, as it appears, the FIT has better performance characteristics than the GT, that is compelling evidence for recommending its use as the FOBT of choice in colon cancer screening programs.42 Screening programs involving new FOBTs deserve serious consideration as a part of our armamentarium of tests for CRC screening.