COLORECTAL CANCER, SERIES #3
`Series Editor: Melvin Schapiro, M.D.
`
`Fecal Immunochemical Tests (FIT)
`vs. Office-Based Guaiac Fecal
`Occult Blood Test (FOBT)
`
`Graeme P. Young
`
`Fecal occult blood tests (FOBT) continue to have an important place in screening for col-
`o rectal cancer as they serve to identify people who are more likely to have neoplasia and
`so direct them to colonoscopy. There are two main FOBT technologies: guaiac or
`gFOBT and fecal immunochemical tests or FIT. They are quite diff e rent from each other
`in their biological, behavioural, clinical and technological characteristics. The criteria
`for the ideal FOBT are best met by FIT. With FIT, the whole sampling process is sim-
`plified for the individual, especially if the brush-sampling technology is used. Clinical
`performance is also better with FIT as they have a better sensitivity:specificity ratio.
`I d e a l l y, sampling for FOBT are done at home at the convenience of the individual. In
`this setting, people are most willing to undertake FOBT of the brush-sampling FIT type.
`
`INTRODUCTION
`
`Fecal occult blood tests (FOBT) have a very real
`
`place in screening for colorectal cancer (CRC) (1).
`Their value is proven in randomized controlled tri-
`als at the population level. They meet WHO require-
`ments (2) in that they are simple tests which serve to
`select out those with a higher probability of having
`CRC (3) to whom diagnostic, perhaps therapeutic,
`
`Graeme P. Young, Professor of Gastroenterology,
`Head, Department of Gastroenterology and Hepatol-
`ogy, Flinders Medical Centre, Bedford Park, Adelaide,
`Australia.
`
`46
`
`PRACTICAL GASTROENTEROLOGY • JUNE 2004
`
`colonoscopy is then directed. Just as we use indicators
`of high risk to determine who gets surveillance
`colonoscopy, i.e. family history and past history of
`adenomas (3,4), the FOBT serves to profile risk. In
`fact, the person with a positive FOBT result is much
`more likely to have neoplasia than the person with a
`family history or past neoplasia.
`It is important that a screening test, which is
`directed at healthy people, have an impact measurable
`at the population level (4). People are inherently reluc-
`tant to undergo invasive and inconvenient tests for
`screening such as colonoscopy without strong motiva-
`(continued on page 49)
`
`Geneoscopy Exhibit 1044, Page 1
`
`

`

`FIT vs. FOBT
`
`COLORECTAL CANCER, SERIES #3
`
`Test – FIT
`Antibody detects globlin.
`No dietary interference.
`Detects only colonic bleeding when
`occult.
`
`Test – gFBOT
`Guaiac detects peroxidase activity.
`Interfered with by plant peroxidases
`and red meat; vitamin C.
`Detects bleeding from entire GIT.
`
`Figure 1. Schema for hemoglobin breakdown in feces and associated issues with the two
`FOBT technologies.
`
`(continued from page 46)
`
`tion. Indeed, colonoscopic screening
`of everyone must be considered very
`carefully when a minority, only about
`4%–7%, will develop CRC.
`
`FECAL OCCULT BLOOD TEST
`(FOBT) TECHNOLOGIES
`There is a range of types of tests for
`blood products in feces collectively
`referred to as FOBT (3–5). FOBT
`should be really qualified, however,
`to more accurately reflect the actual
`technology employed since they are
`by no means similar. The main com-
`mercial FOBT technologies detect
`either of two classes of hemoglobin
`product in feces. The guaiac FOBT (gFOBT) detect
`heme while fecal immunochemical tests (FIT) detect
`globin (Figure 1).
`Detection of heme by gFOBT is dependent on the
`peroxidase activity of heme. Dietary peroxidases (found
`in a range of certain fruit and vegetables especially if
`raw) can cause false-positive results with gFOBT [see
`3]. Antioxidants such as vitamin C may interfere with
`the chemistry of the reaction to cause false-negative
`results (see 4). Dietary heme from red meat also causes
`false positives (6). Heme is also reasonably stable in the
`gut and gFOBT may detect bleeding from any site in the
`GI tract although they are more sensitive for lower GI
`bleeding (3). This means that gFOBT are not selective
`for bleeding from colon/rectum.
`Detection of globin is based on antibodies which
`are generally specific for human hemoglobin and some
`of its lumenally-derived degradation products (3-5).
`FITs are not subject to interference by diet or drugs
`and do not require proscription of any foods or drugs
`prior to sampling feces. As globin is rapidly digested
`in stomach and small intestine, FITs are much more
`selective for occult bleeding of colorectal origin than
`are gFOBT (3,5).
`
`THE IDEAL FOBT
`The primary use of gFOBT and FIT is in screening for
`colorectal cancer (CRC). They should therefore meet
`
`the requirements of WHO for tests used in population-
`based screening (2). Table 1 lists the ideal characteris-
`tics, considering what is involved in the screening
`process.
`Ability to select out those most likely to have neo-
`plasia is embodied in the pre-test/post-test likelihood
`ratio (3,4), or in more familiar terms, reflects the bal-
`ance of sensitivity and specificity. Sensitivity facili-
`tates detection of those with significant neoplasia
`while specificity effectively determines the number of
`colonoscopies needed.
`
`Table 1
`Characteristics of the ideal FOBT.
`
`Sampling
`• Convenient, without need to attend a physician.
`• Readily organizable, without need for diet and drug
`restrictions.
`• Acceptable, with easy and simple fecal sampling.
`
`Performance
`• Selective for colorectal bleeding.
`• Able to select out those most likely to have neoplasia and to
`whom diagnostic colonoscopic resources are applied.
`
`Tests development:
`• Suitable for mass development of large numbers in a short
`time.
`• Subject to ready quality control and objective reading of
`results with a stable, instrument-readable endpoint.
`
`PRACTICAL GASTROENTEROLOGY • JUNE 2004
`
`49
`
`Geneoscopy Exhibit 1044, Page 2
`
`

`

`FIT vs. FOBT
`COLORECTAL CANCER, SERIES #3
`
`Table 2
`Estimates of performance characteristics of different
`types of FOBTs.
`
`Test and type
`Rehydrated Hemoccult:
`gFOBT
`Hemoccult II: gFOBT
`
`Specificity
`for neoplasia*
`90%
`
`94%–98%
`
`HemoccultSENSA:
`gFOBT
`Heme Select variants:
`FIT
`
`88%-92%
`
`95%
`
`Sensitivity
`for cancer+
`90+% with repeated
`annual screening.
`35%–55% with
`once-off testing.
`Up to 80% with
`repeated annual
`testing.
`80% with once-off
`testing.
`70%-82% with
`once-off testing.
`
`Note: Adapted, updated and simplified from [whoguide].
`
`OVERVIEW OF gFOBT PERFORMANCE
`The role of gFOBT is clearly established in screening
`for CRC. A program of regular biennial screening with
`the Hemoccult II® gFOBT significantly reduces popu-
`lation mortality on an intention-to-screen basis by
`15%–18% (7,8). Once-off sensitivity of Hemoccult II®
`is generally in the order of 35%–50% [see 9] although
`repeated annual screening increases sensitivity to 80%
`(10).
`Several methods have been implemented in efforts
`to improve sensitivity. One is rehydration of fecal sam-
`ples prior to development (10). Rehydrated Hemoccult
`II achieved a more substantial (33%) reduction in mor-
`tality from CRC (10) and also reduced incidence in the
`long term by 20% (11).
`The performance characteristics of gFOBT are
`summarised in Table 2. In view of these trial results,
`many bodies have issued guidelines recommending
`that FOBT screening be undertaken, along with
`screening by other modalities (1,12).
`While rehydrated Hemoccult II is more sensitive, it
`has poor specificity, caused by activation of plant per-
`oxidases resulting from rehydration of fecal smears
`(13). It is therefore not recommended. Hemoccult II has
`been compared to rehydrated Hemoccult in two larg e
`studies: the Minnesota randomized controlled trial (10)
`
`50
`
`PRACTICAL GASTROENTEROLOGY • JUNE 2004
`
`Figure 2. Examples of test endpoints for Hemoccult, Hemoccult
`Sensa and InSure FOBT. The latter is an FIT with a stable endpoint
`while the endpoint with the two gFOBTs is often transient.
`
`and the Texas (MD Anderson Cancer Center) screening
`study (14). In the Minnesota study, the positivity rate of
`unhydrated Hemoccult was 2.4% and rehydration
`increased it to 9.8%. In the MD Anderson study, the
`positivity rates were 5% and 14.6% respectively.
`In practice, dietary restriction can be confined to
`red meat alone by waiting three days before develop-
`
`(continued on page 52)
`
`Geneoscopy Exhibit 1044, Page 3
`
`

`

`FIT vs. FOBT
`COLORECTAL CANCER, SERIES #3
`
`(continued from page 50)
`
`Table 3
`Shortcomings of guaiac-based FOBT (gFOBT)
`
`• Restrictions on diet and drugs needed to optimize specificity
`and sensitivity, especially for the sensitive gFOBT.
`• Generally use non-preferred spatula-sampling of stools with
`stool needing to be kept clear of toilet bowl water.
`• Not selective for colorectal bleeding.
`• Office-based testing might lead to increased false-positive
`results and suboptimal number of samples.
`• Endpoint is transient and can be difficult to read.
`• Sensitivity for cancer is suboptimal for less sensitive
`versions.
`• More sensitive versions are subject to unpredictably high
`false-positive rates.
`• Not suitable for mass-development or reading.
`
`ing a guaiac test (15). One review suggested that
`restrictions were not needed (16) but this is not so for
`the more sensitive gFOBT (13). Restrictions of these
`interfering substances do need to be implemented three
`days prior to testing (6). Drugs such as aspirin may
`also cause false-positive results for neoplasia as aspirin
`can cause dose-related gastrointestinal bleeding (3).
`With the Hemoccult Sensa® test, a new developer
`has enhanced sensitivity and improved readability and
`stability of the endpoint (17). An example of Hemoc-
`cult and Hemoccult Sensa test endpoints are shown in
`Figure 2. Care needs to be taken when reading tests
`and training and quality assurance is desirable (see 1).
`While Hemoccult Sensa® is not subject to interference
`by plant peroxidases provided that development is
`delayed for 72 hours after sampling (15) it is more
`affected by red meat ingestion than Hemoccult (13).
`Comparison of Hemoccult Sensa® with Hemoc-
`cult II has been undertaken at five centers (see 9). Its
`positivity rate was always higher than Hemoccult II
`and generally twice as high. In some populations, e.g.
`California, there was an unacceptably low test speci-
`ficity for HOSENSA (18) with a positivity rate of over
`six times that of Hemoccult while detecting twice as
`many cancers. Clearly, attempts to improve sensitivity
`with gFOBT lead to unpredictably variable and some-
`times high positivity rates due to poor specificity.
`Fecal sampling for gFOBT has commonly
`employed the traditional wooden spatula method when
`undertaken outside the doctor’s office.
`
`52
`
`PRACTICAL GASTROENTEROLOGY • JUNE 2004
`
`SHORTCOMINGS OF gFOBT
`Even though of proven use, gFOBT are being used less
`often (1) for the reasons (Table 3) that they do not meet
`the criteria required of an ideal test (Table 1).
`Use of gFOBT in the office setting is also a concern.
`Usually only one sample rather than the recommended
`three is obtained. The patient has rarely undertaken
`dietary preparation and so increases the risk of false-
`positives. There is always the concern that digital rectal
`examination will generate minor trauma and so lead to
`a positive result.
`
`THE CHANGING FACE OF SCREENING
`Screening for colorectal cancer (CRC) has several per-
`spectives—that which applies to the individual and that
`which applies to the population (4). The imperatives for
`each are different.
`The traditional mode has been a face-to-face meet-
`ing between the individual and a health professional—
`sometimes referred to as case-finding or individualistic
`screening. Here, the person can be assessed for symp-
`toms and level of risk. What is offered is done in the
`context of counselling. Duty of care and what is best dri-
`ves the decision-making—cost-effectiveness tends not
`to be a prime issue.
`Population screening is becoming increasingly
`prominent. It seeks, through a common often imper-
`sonal approach, to engage individuals in at least some
`form of preventive activity—in effect, anything is better
`than nothing. Hence, if one seeks to have screening
`impact on CRC outcomes at the population level, sim-
`p l i c i t y, acceptability, feasibility and low initial cost with
`proven cost-effectiveness are needed. Many thousands
`will be tested in a short time-frame often without ability
`to ascertain presence of symptoms or to profile risk.
`Even the approach of population screening is chang-
`ing. Specificity itself has been a major consideration in
`the past (3), but now, as we see a trend to recommend
`screening by colonoscopy itself, specificity is seen as
`being less of an issue than sensitivity (1) and ability to
`detect advanced adenomas, not just cancer, is important.
`
`OVERVIEW OF FIT CHARACTERISTICS
`FITs appear well-placed to overcome the shortcomings
`of gFOBT and fit into this changing face of screening.
`
`Geneoscopy Exhibit 1044, Page 4
`
`

`

`They have an inherent biological advantage with
`their selectivity for colorectal bleeding (3,5).
`They are not subject to exogenous influences by
`diet and drugs and this provides a behavioural advan-
`tage for better particiaption. Removal of typical
`dietary restrictions for a guaiac test (see 3) can
`increase participation significantly in a country where
`red meat consumption is relatively high (19).
`Stool-sampling processes have also evolved with
`FIT. The original wooden spatula used with early
`gFOBTs required multiple sampling from the surface
`of the stool which was ideally kept clear of the toilet
`bowl water. FITs have incorporated newer approaches.
`Some require a probe to be inserted into the stool (e.g.
`Bayer detect™ version of Fujirebio’s Magstream®
`test) while others simply sample toilet bowl water
`from around the immersed stool (InSure®/InForm®,
`Enterix Inc.). Such new approaches may provide
`behavioural advantages if they are preferred over the
`older methods. But they also require validation as reli-
`able means of achieving a representative sample.
`These points are further discussed below.
`gFOBT are designed for small-scale in-office use;
`population screening requires rapid processing and
`development of many samples. With several commer-
`cially-available types, automated development is possi-
`ble (e.g. Bayer detect™, Enterix’s InSure). It is desirable
`to automate the reading of test end-points as well and this
`is also possible for several FIT. Some can give quantified
`endpoints although none is FDA-approved for this at
`present. Quantification facilitates standardization of
`methodology and maintenance of quality control. It also
`allows for adjusting the cut-off level for detecting fecal
`occult blood and deciding on who to colonoscope (3,20).
`
`COMPARISON OF FIT WITH gFOBT
`While FITs have not been compared to gFOBT in con-
`trolled trials of screening using CRC mortality as the
`end-point, several studies using informative surrogate
`end points have compared earlier commercially-con-
`figured FITs to several versions of gFOBT. This has
`been critically reviewed in detail and a full discussion
`is beyond the scope of this review (see 9).
`H e m e S e l e c t® is a stick-sampling FIT that was orig-
`inally developed as Immudia®HemSp by Fujirebio,
`
`FIT vs. FOBT
`
`COLORECTAL CANCER, SERIES #3
`
`( Tokyo, Japan) and has now evolved into the commer-
`cial tests Magtream HemSp and Bayer detect™. It has
`been extensively studied and shown in screening studies
`to detect more neoplasms than Hemoccult (17,18,21).
`Although it does not obviously appear to yield more
`neoplasms than HemoccultSensa, it provides an
`improved balance of sensitivity to specificity in that it is
`as sensitive but considerably more specific (18).
`FlexSure OBT™ is a spatula-sampling FIT that
`has been accepted by FDA as a reference point (22)
`but it has not remained commercially available. A new
`brush-sampling FIT—InSure—compares well with it
`for sensitivity and specificity (22).
`Overall, FIT with published data to support per-
`formance can be expected to have a better sensitiv-
`ity:specificity balance than do gFOBT and so perform
`better in selecting out those who are more likely to
`have neoplasia.
`
`BRUSH-SAMPLING FIT
`Most FIT use a variation of stick-based sampling of
`the stool although in most instances this has evolved
`from the wooden spatula used with the commonest
`gFOBT to a simple probe that once used to sample the
`stool is placed into a plastic tube with preservative.
`In an effort to develop a more acceptable and sim-
`pler sampling methodology, a brush-based sampling
`technique has been developed. The participant is asked
`to sample toilet bowl water from the surface of the
`immersed stool by swishing the brush in the bowl.
`This has been combined with an immunogold mem-
`brane test which uses a dual antibody system specific
`for human hemoglobin. The resultant InSure test
`(Enterix Inc., also known as InForm in Australia) pro-
`vides an endpoint which is stable (Figure 2) and highly
`readable by eye as well as by optical technology that
`allows quantification (23). Sample card development
`can be done completely by robot.
`Initial pre-screening evaluations of this brush-
`sampling FIT showed it to have similar specificity and
`sensitivity to the FIT FlexSure OBT (22).
`In an evaluation of the acceptability of the brush-
`sampling methodology to the general population, three
`randomly selected cohorts in urban Adelaide were allo-
`cated to a mail offer of either Hemoccult (spatula-sam-
`
`PRACTICAL GASTROENTEROLOGY • JUNE 2004
`
`53
`
`Geneoscopy Exhibit 1044, Page 5
`
`

`

`FIT vs. FOBT
`COLORECTAL CANCER, SERIES #3
`
`Figure 3. Effect of test technology on participatory behaviour in
`population screening for colorectal cancer (23).
`
`pling of stool with diet restriction), F l e x S u re OBT ( s p a t-
`ula-sampling of stool without diet restriction) or InSure
`(brush-sampling of stool without diet restriction) (23).
`As can be seen from Figure 3, population participation
`increased with removal of diet restrictions and further
`increased with introduction of the brush-sampling
`
`Figure 4. Fecal hemoglobin concentrations in patients with neoplasia.
`Concentrations are expressed as intensity ratio relative to control
`line, measured using an adaptation of the InSure test where the
`endpoint is quantified by an optical device. Normal n = 166, cancer
`n = 22, adenoma n = 18. Each group is significantly different from
`the other, p < 0.05.
`Quantification of the fecal immunochemical test strips was deter-
`mined using a prototype machine reader and computer software
`developed by Larry Lapointe and Howard Chandler. After digital
`image acquisition and processing, each immunochromatographic
`result was quantified based on the ratio of the colloidal gold signal
`found at the hemoglobin test line to the corresponding signal of the
`internal control line. For quantification purposes, this ratio is
`referred to as the Intensity Ratio.
`
`54
`
`PRACTICAL GASTROENTEROLOGY • JUNE 2004
`
`method. By 12 weeks, participation with InSure testing
`was almost double that of Hemoccult (40% vs 24%).
`Because the sampling method is novel, it was
`evaluated for its ability to provide quantified results
`that differentiated between those who had cancers or
`adenomas and those who were normal at colonoscopy.
`A novel optical method for digital image acquisition
`and processing of the immunochromatographic result
`was specifically developed for this purpose (Larry
`La Pointe, Howard Chandler, personal communica-
`tion). The results, shown in Figure 4, clearly show
`good differentiation between these three clinical
`groups. They also show that people with adenomas
`may bleed.
`A direct within-individual comparison of InSure
`with Hemoccult Sensa is now underway and has been
`reported in abstract form (24). Two populations were
`asked to sample two stools using the brush-sampling
`technique of InSure and three with the Hemoccult
`device, prior to colonoscopy: a) Community screening
`(n = 443), all those positive by qualitative endpoint
`were colonoscoped; b) Colonoscopic examination for
`high risk settings (n = 202). Predetermined diagnostic
`categories were allocated independent of FOBT result.
`InSure was significantly more sensitive than Hemoc-
`cult Sensa, detecting 16/18 cancers compared to 9/18.
`It also detected significantly more adenomas 27/51
`versus 18/51. False-positive rates were similar at 7.8%
`and 7.0% respectively. Expressed in another way,
`InSure resulted in 21 more colonoscopies being done
`than did Hemoccult SENSA but it detected 7 more
`with cancers and 9 more with adenomas.
`
`AVAILABILITY OF FIT
`On the international scene, FIT have been well accepted
`in terms of reimbursement and/or government-funded
`national programs in countries such as Australia and
`Japan. Other countries are planning to follow.
`In the USA, the Centers for Medicare and Medic-
`aid Services (CMS) have expanded the range of
`screening options covered under the Congressionally-
`mandated Medicare colorectal cancer screening bene-
`fit, to include annual screening using FIT. This paves
`the way for appropriate reimbursement.
`
`(continued on page 56)
`
`Geneoscopy Exhibit 1044, Page 6
`
`

`

`FIT vs. FOBT
`COLORECTAL CANCER, SERIES #3
`
`(continued from page 54)
`
`CONCLUSIONS
`FOBT continue to have an important place in screen-
`ing for colorectal cancer as they serve to identify peo-
`ple who are more likely to have neoplasia and so direct
`them to colonoscopy. There are two main FOBT tech-
`nologies: guaiac-based and immunochemical-based.
`They are quite different from each other in their bio-
`logical, behavioural, clinical and technological charac-
`teristics. The criteria for the ideal FOBT are best met
`by FIT. With FIT, the whole sampling process is sim-
`plified for the individual, especially if the brush-sam-
`pling technology is used. Clinical performance is also
`better with FIT as they have a better sensitivity:speci-
`ficity ratio. In this setting, people are most willing to
`undertake FOBT of the brush-sampling FIT type. FIT
`should replace gFOBT as the simple and inexpensive
`approach to population-based screening for CRC. ■
`
`References
`1. Allison JE. Screening for colorectal cancer 2003: Is there still a
`role for the FOBT? Tech Gastrointest Endosc, 2003;5:127-133.
`2. Watson JMG, Jungner G. Principles and Practice of Screening for
`Disease. WHO Public Health Paper 34, 1968.
`3. Young GP, Macrae FA, St John DJB. Clinical methods for early
`detection: Basis, use and evaluation. In: Young GP, Rozen P,
`Levin B Eds. Prevention and Early Detection of Colorectal Can -
`cer, London, Saunders, 1996: 241-270.
`4. Young GP, Rozen P, Levin B. How should we screen for early
`colorectal neoplasia. In: Rozen P, Young GP, Levin B, Spann SJ.
`Eds. Colorectal Cancer in Clinical Practice, London, Martin
`Dunitz, 2002:77-99.
`5. Young GP, St John DJB. Faecal occult blood tests, choice, usage
`and clinical applications. Clin Biochem Rev, 1992; 13:161-167.
`6. Feinberg EJ, Steinberg WM, Banks BL, Henry JP. How long to
`abstain from eating red meat before fecal occult blood tests. Ann
`Intern Med, 1990; 113:403-404.
`7. Hardcastle JD, Chamberlain JO, Robinson MHE, Moss SM,
`Amar SS, Balfour TW, et al. Randomised controlled trial of fae-
`cal-occult-blood screening for colorectal cancer. Lancet, 1996;
`348:1472-1477.
`8. Kronborg O, Fenger C, Olsen J, Jorgenson OD, Sondergaard O.
`Randomised study of screening for colorectal cancer with faecal-
`occult-blood test. Lancet, 1996; 348:1467-1471.
`9. Young GP, St John DJB, Winawer SJ, Rozen P. Choice of fecal
`occult blood tests for colorectal cancer screening: recommenda-
`tions based on performance characteristics in population studies.
`Am J Gastroenterol, 2002;97:2499-2507.
`10. Mandel JS, Bond JH, Church TR, Snover DC, Bradley GM,
`Schuman LM, et al. Reducing mortality from colorectal cancer
`by screening for fecal occult blood. N Engl J Med, 1993;
`328:1365-1371.
`11. Mandel JS, Church TR, Bond JH, Ederer F, Geisser MS, Mongin
`SJ, et al. The effect of fecal occult-blood screening on the inci-
`dence of colorectal cancer. N Engl J Med, 2000; 343;1603-1607.
`12. Winawer SJ, Fletcher RH, Miller L, Godlee F, Stolar MH, Mul-
`row CD, et al. Colorectal cancer screening: clinical guidelines
`and rationale. Gastroenterology, 1997; 112:594-642.
`
`56
`
`PRACTICAL GASTROENTEROLOGY • JUNE 2004
`
`13. Sinatra M, St John DJB, Young GP. Interference of plant perox-
`idases with guaiac-based fecal occult blood tests is avoidable.
`Clin Chem, 1999, 45:123-126.
`14. Levin B, Hess K, Johnson C. Screening for colorectal cancer: a
`comparison of 3 fecal occult blood tests. Arch Intern Med, 1997;
`157:970-976.
`15. Rozen, P, Knaani J, Samuel Z: Eliminating the need for dietary
`restrictions when using a sensitive guaiac fecal occult blood test.
`Dig Dis Sci, 1999; 44:756-760.
`16. Pignone M, Campbell MK, Phillips C. Meta-analysis of dietary
`restriction during fecal occult blood testing. Eff Clin Pract,
`2001;4:150-156.
`17. St John DJB, Young GP, Alexeyeff MA, Deacon MC, Cuthbert-
`son AM, Macrae FA, et al. Evaluation of new occult blood tests
`for detection of colorectal neoplasia. Gastroenterology, 1993;
`104:1661-1668.
`18. Allison JE, Tekawa IS, Ransom LJ, Adrian AL. A comparison of
`fecal-occult blood tests for colorectal-cancer screening. N Engl J
`Med, 1996; 334:155-159.
`19. Cole SR, Young GP. Participation in faecal occult blood test-
`based screening for colorectal cancer is reduced by dietary
`restriction. Med J Aust, 2001;175:195-198.
`20. Castiglione G, Grazzini G, Miccinesi G, Rubeca T, Sani C, Turco
`P, Zappa M. Basic variables at different positivity thresholds of a
`quantitative immunochemical test for faecal occult blood. J Med
`Screen, 2002;9:99-103.
`21. Zappa M, Castiglione G, Paci E, Grazzini G, Rubeca T, Turco P,
`Crocetti E, Ciatto S. Measuring interval cancers in population-
`based screening using different assays of fecal occult blood test-
`ing: the District of Florence experience. Int J Cancer,
`2001;92:151-154.
`22. Young GP, St John DJB, Cole SR, Bielecki BE, Pizzey C, Sina-
`tra MA, Cadd B, Morcom J. A prescreening evaluation of a
`brush-based faecal immunochemical test for haemoglobin. J Med
`Screen, 2003; 10: 123-128.
`23. Cole SR, Young GP, Esterman A, Cadd B, Morcom J. A ran-
`domized trial of the impact of new fecal hemoglobin test tech-
`nologies on population participation in screening for colorectal
`cancer. J Med Screen, 2003; 10: 117-122.
`24. Cole S, Smith A, Bampton P, Sandford J, Morcom J, Young GP.
`Screening for Colorectal Cancer: Direct Comparison of a Brush-
`Sampling Fecal Immunochemical Test for Hemoglobin with
`Hemoccult. Gastroenterology, 2003;124:A80.
`
`PRACTICAL
`GASTROENTEROLOGY
`
`Practical Gastroenterology reprints are valuable,
`authoritative, and informative. Special rates are
`available for quantities of 100 or more.
`For further details on rates or to place an order:
`Practical Gastroenterology
`Shugar Publishing
`9 9 B Main Street, Westhampton Beach, NY 11978
`Phone: 631-288-4404 Fax: 631-288-4435
`
`Geneoscopy Exhibit 1044, Page 7
`
`