Optimizing Colorectal Cancer Screening by Getting FIT Right

Abstract

See “Cost-effectiveness analysis of a quantitative immunochemical test for colorectal cancer screening” by Wilschut JA, Hol L, Dekker E, et al, on page 1648. See “Cost-effectiveness analysis of a quantitative immunochemical test for colorectal cancer screening” by Wilschut JA, Hol L, Dekker E, et al, on page 1648. Over the last 4 decades, there have been declines in US colorectal cancer (CRC) incidence from 59.5 per 100,000 population in 1975 to 44.7 in 2007 and in the CRC death rate from 28.6 per 100,000 population in 1976 to 16.7 in 2007.1Richardson L.C. Tai E. Rim S.H. et al.Vital Signs: colorectal cancer screening, incidence, and mortality—United States, 2002–2010.MMWR Morb Mortal Wkly Rep. 2011; 60: 884-889Google Scholar Not all Americans are benefiting equally from this decline, however. In a recent American Cancer Society analysis, CRC mortality rates significantly decreased in all states except Mississippi between 1990 and 2007. The decrease in death rates between 1990 and 1994 and 2003 and 2007 ranged from 9% in Alabama to >33% in Massachusetts, Rhode Island, New York, and Alaska; Mississippi and Wyoming showed no significant decrease. The highest CRC mortality rates shifted from the northeastern states during 1990 to 1994 to the southern states along the Appalachian corridor during 2003 to 2007. The decrease in CRC mortality rates by state correlated strongly with uptake of screening.2Naishadham D. Lansdorp-Vogelaar I. Siegel R. et al.State disparities in colorectal cancer mortality patterns in the United States.Cancer Epidemiol Biomarkers Prev. 2011; 20: 1296-1302Google Scholar What drives the variation in screening rates? Insurance coverage is strongly correlated with uptake of screening,1Richardson L.C. Tai E. Rim S.H. et al.Vital Signs: colorectal cancer screening, incidence, and mortality—United States, 2002–2010.MMWR Morb Mortal Wkly Rep. 2011; 60: 884-889Google Scholar but even with Medicare coverage of screening, there are gaps in use of screening tests.3O'Malley A.S. Forrest C.B. Feng S. et al.Disparities despite coverage: gaps in colorectal cancer screening among Medicare beneficiaries.Arch Intern Med. 2005; 165: 2129-2135Google Scholar Socioeconomic factors such as education and income, as well as racial and ethnic barriers play a role, as well.4Doubeni C.A. Laiyemo A.O. Reed G. et al.Socioeconomic and racial patterns of colorectal cancer screening among Medicare enrollees in 2000 to 2005.Cancer Epidemiol Biomarkers Prev. 2009; 18: 2170-2175Google Scholar, 5Doubeni C.A. Laiyemo A.O. Young A.C. et al.Primary care, economic barriers to health care, and use of colorectal cancer screening tests among Medicare enrollees over time.Ann Fam Med. 2010; 8: 299-307Google Scholar Having a usual source of primary care increases the uptake of CRC screening,5Doubeni C.A. Laiyemo A.O. Young A.C. et al.Primary care, economic barriers to health care, and use of colorectal cancer screening tests among Medicare enrollees over time.Ann Fam Med. 2010; 8: 299-307Google Scholar but access to specialty care is also important. Screening capacity and specialty access vary between urban and rural areas,6Benuzillo J. Jacobs E. Hoffman R. et al.Rural–urban differences in colorectal cancer screening capacity in Arizona.J Community Health. 2009; 34: 523-528Google Scholar and lack of access to colonoscopy may make CRC screening less likely. Patient preferences are an important factor, and if people are only offered colonoscopy they may be less likely to be screened, than if offered fecal blood testing.7Inadomi J.M. Vijan S. Janz N.K. et al.Method of recommendation for colorectal cancer screening strategies impacts adherence.Gastroenterology. 2010; 138: S-23Google Scholar Is noninvasive fecal blood screening inferior to colonoscopy? Many in the US gastroenterology community think of colonoscopy as the preferred test for CRC screening.8Rex D.K. Johnson D.A. Anderson J.C. et al.American College of Gastroenterology guidelines for colorectal cancer screening 2009 [corrected].Am J Gastroenterol. 2009; 104: 739-750Google Scholar The American Cancer Society and US Multi-Society Task Force have stated a preference for tests that prevent cancer, over those that merely detect it, stating that fecal blood screening is inferior for adenoma detection and cancer prevention.9Levin B. Lieberman D.A. McFarland B. et al.Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology.CA Cancer J Clin. 2008; 58: 130-160Google Scholar However, fecal screening is the most common form of CRC screening across the world.10Benson V.S. Patnick J. Davies A.K. et al.Colorectal cancer screening: a comparison of 35 initiatives in 17 countries.Int J Cancer. 2008; 122: 1357-1367Google Scholar The US Preventive Services Task Force considers annual, high-sensitivity fecal blood testing (guaiac or immunochemical) to be equivalent to a colonoscopy performed every 10 years,11US Preventive Services Task ForceScreening for colorectal cancer: U.S. Preventive Services Task Force Recommendation Statement.Ann Intern Med. 2008; 149: 627-637Google Scholar based on a simulation analysis showing nearly identical outcomes when both tests have equally high adherence.12Zauber A.G. Lansdorp-Vogelaar I. Knudsen A.B. et al.Evaluating test strategies for colorectal cancer screening: a decision analysis for the U.S. Preventive Services Task Force.Ann Intern Med. 2008; 149: 659-669Google Scholar In fact, adherence is the most important factor affecting screening effectiveness.12Zauber A.G. Lansdorp-Vogelaar I. Knudsen A.B. et al.Evaluating test strategies for colorectal cancer screening: a decision analysis for the U.S. Preventive Services Task Force.Ann Intern Med. 2008; 149: 659-669Google Scholar A fecal blood screening option serves 2 functions: It allows screening to occur in places that may be underserved by endoscopic specialists and it provides an important option for screening the large proportion of people who are likely to refuse screening if only offered a colonoscopy. Kaiser Permanente in Northern California has used the fecal immunochemical test (FIT) to deliver CRC screening through organized mailed outreach, and we have observed a substantial increase in CRC screening rates, as measured by publicly reported Healthcare Effectiveness Data and Information Set (HEDIS) rates (Figure 1) . FIT is part of a comprehensive screening program that includes organized, population-based, mailed outreach of FIT screening kits and opportunistic in-reach using electronic health record prompts to identify patients who are due for screening at the time of an office visit. In addition to opportunistic referral for screening colonoscopy, office-based support staff are employed to distribute screening kits at physician office visits or preventive health flu shot clinics.13Levin T.R. Jamieson L. Burley D.A. et al.Organized colorectal cancer screening in integrated health care systems.Epidemiologic Reviews. 2011; 33: 101-110Google Scholar From 2005, when HEDIS CRC screening rates were first publicly reported, to 2011, the commercial screening rate at Kaiser Permanente in Northern California has increased from 37% to 75%, and the Medicare population screening rate has increased from 41% to 85%. From a relative national position of being below the 50th percentile in 2005, our screening performance now ranks in the top 10% of HEDIS reporting health plans. This screening rate increase is associated with a change in the incidence of CRC in our population, and the stage at which we diagnose CRC.13Levin T.R. Jamieson L. Burley D.A. et al.Organized colorectal cancer screening in integrated health care systems.Epidemiologic Reviews. 2011; 33: 101-110Google Scholar The question facing clinicians and health care delivery system policy and decision makers is no longer whether to use FIT as part of the screening menu of options, but which FIT to use and how to optimize FIT use. There are no comparative effectiveness studies of different FIT platforms on the patient-centered outcomes of screening uptake, cancer incidence, or cancer mortality, but several trials have compared FIT to either high-sensitivity or standard guaiac testing in trials randomized at invitation,14van Rossum L.G. van Rijn A.F. Laheij R.J. et al.Random comparison of guaiac and immunochemical fecal occult blood tests for colorectal cancer in a screening population.Gastroenterology. 2008; 135: 82-90Abstract Full Text Full Text PDF Scopus (619) Google Scholar, 15Hol L. van Leerdam M.E. van Ballegooijen M. et al.Screening for colorectal cancer: randomised trial comparing guaiac-based and immunochemical faecal occult blood testing and flexible sigmoidoscopy.Gut. 2010; 59: 62-68Google Scholar and in studies in which participants collected several side-by-side samples.16Allison J.E. Sakoda L.C. Levin T.R. et al.Screening for colorectal neoplasms with new fecal occult blood tests: update on performance characteristics.J Natl Cancer Inst. 2007; 99: 1462-1470Google Scholar, 17Allison J.E. Tekawa I.S. Ransom L.J. et al.A comparison of fecal occult-blood tests for colorectal-cancer screening.N Engl J Med. 1996; 334: 155-159Google Scholar, 18Park D.I. Ryu S. Kim Y.H. et al.Comparison of guaiac-based and quantitative immunochemical fecal occult blood testing in a population at average risk undergoing colorectal cancer screening.Am J Gastroenterol. 2010; 105: 2017-2025Google Scholar In these studies, the FIT was shown to lead to higher patient participation, higher diagnostic yield, and improved sensitivity with preserved specificity compared with guaiac testing or flexible sigmoidoscopy. All FITs use antibodies specific for human hemoglobin to detect blood loss from the colon, but there is considerable variation between tests, and very little comparative research evaluating different tests. Each test uses a different epitope for the immunoassay, and the US Preventive Services Task Force evidence review declared that there was insufficient evidence to declare 1 version of FIT superior to another.19Whitlock E.P. Lin J.S. Liles E. et al.Screening for colorectal cancer: a targeted, updated systematic review for the U.S. Preventive Services Task Force.Ann Intern Med. 2008; 149: 638-658Google Scholar FITs vary in terms of collection devices, (whether dry cards are used or samples are mixed with a liquid buffer), number of samples (some tests are approved as a 1-sample test, and others are approved using 2 or 3 samples), and how they are read (most FITs sold in the United States are read manually, and all are reported as a qualitative, positive/negative, result). In this issue of Gastroenterology, Wilschut et al,20Wilschut J.A. Hol L. Dekker E. et al.Cost-effectiveness analysis of a quantitative immunochemical test for colorectal cancer screening.Gastroenterology. 2011; 141: 1648-1655Abstract Full Text Full Text PDF Scopus (100) Google Scholar from Amsterdam, Rotterdam, and Nijmegen, The Netherlands, have used simulation analysis to determine the optimum cutoff for a quantitative FIT in population screening. Several versions of FIT have the key advantage of being read on an automated machine with a quantitative result, reporting the exact concentration of hemoglobin per milliliter of buffer. This allows clinicians to choose the best cutoff for their patient population and available clinical resources. Only 1 FIT sold in the United States (OC-Auto FIT Chek, Polymedco, Cortlandt Manor, NY) is read by an automated machine, allowing the reporting of a quantitative result. This test is licensed from Eiken in Japan, and is identical to the test used in the Dutch trials. In clinical practice, in part because of regulatory requirements, the test in only reported as a qualitative (positive/negative) result. The Wilschut study used the MIcrosimulation SCreening ANalysis (MISCAN)-Colon microsimulation model to explore the implications of varying FIT cutoffs between 50, 75, 100, 150, and 200 ng hemoglobin per milliliter of buffer.20Wilschut J.A. Hol L. Dekker E. et al.Cost-effectiveness analysis of a quantitative immunochemical test for colorectal cancer screening.Gastroenterology. 2011; 141: 1648-1655Abstract Full Text Full Text PDF Scopus (100) Google Scholar The MISCAN model has been refined and repeatedly validated over the last decade and was instrumental in the recent revision of the US Preventive Services Task Force guideline.12Zauber A.G. Lansdorp-Vogelaar I. Knudsen A.B. et al.Evaluating test strategies for colorectal cancer screening: a decision analysis for the U.S. Preventive Services Task Force.Ann Intern Med. 2008; 149: 659-669Google Scholar The number of different comparisons possible with a model like MISCAN is beyond the scope of any randomized trial, and the randomized trials of FIT screening that are now just underway at the US Veteran's administration or in Europe will not have results for a decade or longer. In the current analysis, MISCAN simulations were used to vary age ranges of starting and stopping screening and intervals between screening tests. An advantage to this approach is the ability to identify whether changing any of these variables has a meaningful effect on screening effectiveness. Like all simulation models, MISCAN is limited by the assumptions that are built into the model. For the current analysis, FIT performance was determined using adherence and test performance data from 4 studies performed in The Netherlands.14van Rossum L.G. van Rijn A.F. Laheij R.J. et al.Random comparison of guaiac and immunochemical fecal occult blood tests for colorectal cancer in a screening population.Gastroenterology. 2008; 135: 82-90Abstract Full Text Full Text PDF Scopus (619) Google Scholar, 15Hol L. van Leerdam M.E. van Ballegooijen M. et al.Screening for colorectal cancer: randomised trial comparing guaiac-based and immunochemical faecal occult blood testing and flexible sigmoidoscopy.Gut. 2010; 59: 62-68Google Scholar, 21Hol L. Wilschut J.A. van Ballegooijen M. et al.Screening for colorectal cancer: random comparison of guaiac and immunochemical faecal occult blood testing at different cut-off levels.Br J Cancer. 2009; 100: 1103-1110Google Scholar, 22van Rossum L.G. van Rijn A.F. Laheij R.J. et al.Cutoff value determines the performance of a semi-quantitative immunochemical faecal occult blood test in a colorectal cancer screening programme.Br J Cancer. 2009; 101: 1274-1281Google Scholar The sensitivity of the 100 ng/mL FIT was assumed to be 51% for CRCs long before they become clinically apparent and 83% near the time of clinical presentation, with a large adenoma sensitivity of 13%. For the 50 ng/mL cutoff, assumed sensitivity was 61% for long preclinical cancers, and 88% for nearly clinical cancers; large adenoma sensitivity was 16.7%. These trials used a 1-sample FIT, so these MISCAN results only apply to varying the hemoglobin cutoff for the 1-sample Eiken FIT. Other investigators have used multiple samples and achieved a higher sensitivity, but usually at the expense of lower specificity.18Park D.I. Ryu S. Kim Y.H. et al.Comparison of guaiac-based and quantitative immunochemical fecal occult blood testing in a population at average risk undergoing colorectal cancer screening.Am J Gastroenterol. 2010; 105: 2017-2025Google Scholar, 23Levi Z. Rozen P. Hazazi R. et al.A quantitative immunochemical fecal occult blood test for colorectal neoplasia.Ann Intern Med. 2007; 146: 244-255Google Scholar As the cost of treating CRC has increased, FIT-based CRC screening has increasingly been shown to be cost saving.24Lansdorp-Vogelaar I. van Ballegooijen M. Zauber A.G. et al.Effect of rising chemotherapy costs on the cost savings of colorectal cancer screening.J Natl Cancer Inst. 2009; 101: 1412-1422Google Scholar, 25Parekh M. Fendrick A.M. Ladabaum U. As tests evolve and costs of cancer care rise: reappraising stool-based screening for colorectal neoplasia.Aliment Pharmacol Ther. 2008; 27: 697-712Google Scholar Choosing a 50 ng/mL cutoff requires more colonoscopies per cancer or large adenoma detected, but the cancer treatment savings may offset the increased colonoscopy costs.20Wilschut J.A. Hol L. Dekker E. et al.Cost-effectiveness analysis of a quantitative immunochemical test for colorectal cancer screening.Gastroenterology. 2011; 141: 1648-1655Abstract Full Text Full Text PDF Scopus (100) Google Scholar The Wilschut analysis compared many different strategies, and at all ages and intervals between tests, the 50 ng/mL cutoff was the most efficient strategy, but varying the interval between 1 year and 1.5–2 years made only a small, incremental impact on the number of life-years gained, when attendance was high, but led to a substantial increase in costs. The colonoscopy costs and the CRC treatment costs were derived from those in Holland, and may both be lower than US costs. The main limitation of this analysis is in the assumptions of test performance characteristics. If the assumed incremental increase in sensitivity between the 50 ng/mL cutoff and the 100 ng/mL cutoff is overestimated or the magnitude of reduced specificity between 50 and 100 ng/mL is underestimated, the 50 ng/mL cutoff will seem to be more cost effective than it really is. In other settings, with different patient populations, FIT at each cutoff has performed differently.18Park D.I. Ryu S. Kim Y.H. et al.Comparison of guaiac-based and quantitative immunochemical fecal occult blood testing in a population at average risk undergoing colorectal cancer screening.Am J Gastroenterol. 2010; 105: 2017-2025Google Scholar, 23Levi Z. Rozen P. Hazazi R. et al.A quantitative immunochemical fecal occult blood test for colorectal neoplasia.Ann Intern Med. 2007; 146: 244-255Google Scholar Choosing a lower cutoff for test positivity is associated with other costs other than just that of colonoscopy. In the MISCAN assumptions, the specificity of the FIT 50 was assumed to be 95.8% (3 points lower than the FIT 200, and 2 points lower than the FIT 100). This translates into an additional 2%–3% of the population with a false-positive test. The realities of managing a FIT screening program involve substantial effort spent reminding and encouraging primary care physicians to refer their patients with a positive FIT for colonoscopy, and reminding patients to have colonoscopy. This has a cost in terms of physician and support staff time and effort. In a Korean study, the specificity of the FIT 50 was reported to be as low as 88%,18Park D.I. Ryu S. Kim Y.H. et al.Comparison of guaiac-based and quantitative immunochemical fecal occult blood testing in a population at average risk undergoing colorectal cancer screening.Am J Gastroenterol. 2010; 105: 2017-2025Google Scholar reflecting a higher false-positive rate in populations with a less severe spectrum of disease. At 88% specificity, 12% of screenees would experience a false-positive test. Increasing the number of samples collected to 2 or 3 may also increase sensitivity,18Park D.I. Ryu S. Kim Y.H. et al.Comparison of guaiac-based and quantitative immunochemical fecal occult blood testing in a population at average risk undergoing colorectal cancer screening.Am J Gastroenterol. 2010; 105: 2017-2025Google Scholar with corresponding loss of specificity, but may adversely affect patient participation by making screening more complex. These issues aside, the Wilschut study provides important information about what matters in FIT screening. Based on these results, maximizing sensitivity is most important, if specificity can be reasonably preserved, and participation can be maintained. In populations with a different prevalence of disease, or with patients who have had multiple rounds of screening, going to a FIT 50 strategy may lead to an unacceptably high false-positive rate, which could strain colonoscopy resources in that particular setting. In the Dutch population that formed the basis of this study, FIT 50 looks like a good option. Other settings and populations should repeat this simulation exercise and consider screening uptake, colonoscopy adherence derived from their patients, and colonoscopy capacity in their setting, before drawing conclusions about how to use quantitative FIT screening. Cost-effectiveness Analysis of a Quantitative Immunochemical Test for Colorectal Cancer ScreeningGastroenterologyVol. 141Issue 5PreviewTwo European randomized trials (N = 30,000) compared guaiac fecal occult blood testing with quantitative fecal immunochemical testing (FIT) and showed better attendance rates and test characteristics for FIT. We aimed to identify the most cost-effective FIT cutoff level for referral to colonoscopy based on data from these trials and allowing for differences in screening ages. Full-Text PDF