Where does it FIT?: The roles of fecal testing and colonoscopy in colorectal cancer screening

While colonoscopy predominates among the selection of colorectal cancer screening tests, stool testing, particularly the fecal immunochemical test (FIT), has also carved out an important niche for itself. Its simplicity and low cost make it ideal for mass population screening both in the US and abroad, as well as an alternative choice for those reluctant to undergo endoscopy.

Colorectal cancer (CRC) incidence rates have been declining in the United States for the past few decades, due in large part to the salutary effects of screening.¹ Screening guidelines offer a menu of CRC screening test choices from which asymptomatic, average-risk adults age 50 years and older can choose that, over this time, have evolved from guaiac-based fecal occult blood tests, barium enema, and rigid sigmoidoscopies to more sophisticated fecal immunochemical tests to detect fecal hemoglobin, colonoscopy, virtual colonoscopy, and stool tests for DNA mutations.²

Most clinicians and policymakers at the present time seem to be pushing the use of colonoscopy as the screening tool of choice.³ In truth, recent studies have confirmed colonoscopy's superiority for the reduction of colorectal cancer (CRC) mortality.^4,5 But improved mortality does come at a price, starting with the high cost of the procedure (including the operator, anesthesia, and the facility). Colonoscopy is also accompanied by a risk of bleeding, perforation, anesthesia complications, and other side effects; the requirement of a difficult two-day bowel cleansing procedure (inclusive of the restrictive liquid diet preparation); the need to take off from work and have a companion in attendance; and a 20% poor prep rate necessitating repetition of the procedure.

In the US or other Western societies, there may be an acceptance of such negatives for colonoscopy as well as a higher threshold for the amount we are willing to spend on our healthcare. The colonoscopy, for a variety of reasons that includes a belief that its efficacy in reducing mortality is higher than the other acceptable tests, as well as financial incentives on the part of gastroenterologists, has become the “gold standard”.^6,7 Thus, it is often the only test offered to eligible individuals and certainly is the one that is most often encouraged. Furthermore, there will always remain a substantial fraction of the population who will be resistant to the luminal invasion that a colonoscopy represents, are not good candidates for this procedure due to various comorbidities, or who are unable to access services for financial or social reasons.

Historically, it is the detection of occult blood in the stool that was the forerunner of CRC screening. Greegor showed in the 1960s that guaiac-based testing for occult blood could potentially identify CRC in asymptomatic individuals.⁸ His work and that of others led to three randomized trials which all proved that guaiac-based fecal occult blood testing (gFOBT) could reduce CRC mortality on the order of 30%.⁹ Guaiac-based FOBT costs only a few dollars, is logistically easy, has virtually no complications, and requires no special facilities, making it a practical tool with which to screen asymptomatic, average risk individuals.

In 2008, several U.S. professional societies endorsed the adoption of the fecal immunochemical test (FIT) as a replacement for FOBT.^2,10 Unlike gFOBT that detects heme in the stool and is sensitive to exogenous hemoglobin derived from the diet, FIT uses antibodies specific to the globin portion of hemoglobin. Human globin does not survive in gastric juices; thus, FIT is specific for lower gastrointestinal bleeding. This test also has the advantage that it does not require prior dietary restriction, utilizes fewer stool samples, can be calibrated to assess a specific level of hemoglobin in the stool, is simpler for the patient to perform, and has higher sensitivity and specificity when compared to gFOBT. A recent meta-analysis examined the diagnostic accuracy of FIT for CRC in asymptomatic, average-risk adults and found FIT to be moderately sensitive and highly specific (sensitivity 0.79 (95% CI, 0.69 to 0.86); specificity 0.94 (95% CI, 0.92 to 0.95)), with an overall accuracy for CRC of 95% (95% CI, 93% to 97%).¹¹ Similar to gFOBT, FIT is recommended annually and, if positive, followed by a diagnostic colonoscopy.¹²

This relatively simple and cheap test has generally been proposed for mass population as opposed to individual screening.¹³ Thus, for individuals who seek their wellness care, including cancer screening, through an individual physician, colonoscopy has usually been the recommended CRC screening modality. However, when an organization or health agency seeks to screen a large population, particularly disadvantaged populations, who may or may not have had contact with a primary caregiver, the test of choice is a stool-based one, most recently the FIT.¹⁴ In the U.S., Kaiser Permanente of Northern California and the Veterans Administration have been utilizing FIT to send to their members who have not sought annual check-ups. Likewise, the Canadian Health Care System has provided FIT by mail for CRC screening to population groups in Ontario in a similar fashion.

The incidence of CRC in Asian countries, while traditionally low, has been rapidly rising over the past two decades.^15,16 Between 1971-1975 and 2006-2010, the age standardized mortality rates for CRC in Taiwan doubled for men and increased 1.5-fold for women.¹⁷ This upsurge in CRC incidence and mortality in Taiwan has been attributed to rapid westernization in socioeconomic status, environment, and lifestyles in the post-Japanese period¹⁸ and endocrine-related changes in women related to the use of oral contraception and hormone replacement therapy.¹⁷

Although colonoscopy is the generally preferred screening method in Taiwan, the Health Promotion Administration initiated a national population CRC screening program in 2004 with biennial FIT plus diagnostic colonoscopy follow-up.¹⁹ Initially available to adults 50-69 years of age, this was later expanded to include individuals up to 74 years of age. The advantage of FIT over guaiac-based fecal occult blood testing (gFOBT) included lack of dietary restrictions, fewer false positive test results, improved participation in screening programs, and comparability to colonoscopy in detecting lesions of interest among Asian populations.

In this issue of Cancer, Chiu et al., evaluate the effectiveness of the nationwide FIT screening program in Taiwan.²⁰ The records of approximately 1.16 million adults aged 50-69 years who completed FIT screening between 2004 and 2009 were examined. Individuals were followed for a maximum of 6 years to evaluate repeat biennial testing, with participants completing 1-3 rounds of screening with FIT. Outcomes that were evaluated included CRC diagnosis and death ascertained from the program database with linkages to the National Cancer Registry and the National Death Registry in Taiwan. In the first round of screening, 1,160,895 (21.4% of the total eligible population) participated in FIT screening; of these, 329,042 (28.3%) went on to further screening. A positive FIT was found in 4.1% and 3.8% for the first and second screening rounds, respectively, and the majority of individuals (80% and 85% in rounds 1 and 2) with a positive FIT completed diagnostic testing (85% with colonoscopy). Overall, adenomas were detected in more than 19,000 individuals and 2,805 colorectal cancers were found. The investigators believed that the screening program resulted in a significant impact on CRC mortality in the screened population.

The researchers surpassed their goal to achieve 20% coverage in their target population during this inaugural round of screening and expect to screen up to 60% by 2014 but the calculation of CRC mortality reduction based on a small, non-representative fraction of the population raises concerns. Although Chiu et al. report 85.5%, 10.2%, and 4.3% (total 100%) completed diagnostic colonoscopy, sigmoidoscopy plus a barium enema, and an unspecified procedure, respectively, following a positive FIT, it is not clear what proportion of individuals who tested positive completed diagnostic testing. Percentages reported appear to represent the relative frequency of the various diagnostic tests used. Another issue that was not elucidated by the authors, that typically receives little attention when discussing population-based FIT screening programs, is the rate of false negative tests. A negative FIT result often confers a sense of reassurance regarding risk and may result in longer intervals between screening rounds, and thus delayed detection of advanced lesions and poorer health outcomes.²¹ Despite several limitations, this study confirms the ability of mass population CRC screening to engender large-scale compliance with simple inexpensive screening tests.

How does all of this resonate in terms of how the list of CRC screening options included in the current American Cancer Society guidelines² should be weighed? Clearly, if physician behavior and other indicators are to be believed, very few believe that these screening modalities share equal preference. The barium enema has virtually no evidence of efficacy to support its use and the United States Preventive Services Task Force no longer includes barium enema among its CRC screening recommendations.¹⁰ Insofar as current evidence and practice is concerned, the use of colonoscopy is dominating sigmoidoscopy, at least in the US.³ So how do we weigh endoscopy versus fecal blood testing, as represented by FIT? As suggested by others, colonoscopy is the screening modality of choice in the individual physician-patient encounter, with other modalities for those who refuse. However, in mass public health CRC prevention efforts, FIT is a more practical and logistically and financially sound approach.^9,13

Successful models of organized CRC screening exist in the United Kingdom, Canada, and the United States using varied approaches to the delivery of fecal-based CRC screening to their populations.⁹ Reaching and screening far more new screening-eligible individuals, organized programs offer several advantages over traditional opportunistic screening in which a patient requests testing or is offered screening during a healthcare encounter for another purpose. Organized programs efficiently and accurately identify eligible individuals and offer uniformly delivered screening services to the target population. With explicit policies that govern the age at which to begin and terminate CRC screening, organized fecal-based CRC screening programs define the terms of implementation, standardize testing and employ a defined quality assurance structure that is also capable of tracking CRC outcomes.⁹ Because organized CRC screening programs provide the fecal test via the postal mail, they do not require contact with the healthcare system and are well-suited to populations lacking or having limited access to medical facilities. They have been particularly adept at reaching poor and geographically and linguistically isolated populations and attaining high levels of compliance and repeated testing over time.^14,22 The effectiveness of such programs in screening large numbers of individuals who might otherwise never be screened and which facilitates repeated testing over time challenges the notion that the highly sensitive colonoscopy is the “only” test with which to screen for CRC on a population basis.

Improvements to the use of FIT to screen large numbers of individuals continue to evolve. Some have suggested, to increase FIT sensitivity, a risk stratification algorithm incorporating the FIT result, family history, age, and total calcium intake to guide referral to diagnostic colonoscopy.²³ New generations of stool-based screening tests are on the way, with the recently approved Cologuard in the vanguard.²⁴ But cost and simplicity will keep FIT as a major player in the CRC screening arena for some years to come. An appreciation of its performance characteristics is critical to be able to decide its role in various scenarios relative to other screening modalities.