Comparative Effectiveness of Five Fecal Immunochemical Tests using Colonoscopy as the Gold Standard: Study Protocol

Abstract

Background

There are nearly 50,000 colorectal cancer (CRC) deaths in the United States each year. CRC is curable if detected in its early stages. Fecal immunochemical tests (FITs) can detect precursor lesions and many can be analyzed at the point-of-care (POC) in physician offices. However, there are few data to guide test selection. Broader use of FITs could make CRC screening more accessible, especially in resource-poor settings.

Methods

A total of 3600 racially and ethnically diverse individuals aged 50 to 85 years of age having either a screening or surveillance colonoscopy will be recruited. Each participant will complete five FITs on a single stool sample. Test characteristics for each FIT for advanced colorectal neoplasia (ACN) will be calculated using colonoscopy as the gold standard.

Results

We have complete data from a total of 2990 individuals. Thirty percent are Latino and 5.3% are black/African American. We will present full results once the study is completed.

Conclusions

Our focus in this study is how well FITs detect ACN, using colonoscopy as the gold standard. Four of the five FITs being used are POC tests. Although FITs have been shown to have acceptable performance, there is little data to guide which ones have the best test characteristics and colonoscopy is the main CRC screening test used in the United States. Use of FITs will allow broader segments of the population to access CRC screening because these tests require no preparation, are inexpensive, and can be collected in the privacy of one’s home. Increasing CRC screening uptake will reduce the burden of advanced adenomas and colorectal cancer.

Introduction

Colorectal cancer (CRC) is a leading cause of cancer death in the U.S., with more than 50,000 deaths each year.¹ Since CRC develops slowly over a number of years from precursor lesions called polyps,² it is both preventable with polypectomy and curable with surgery when detected in early stages.^3,4 Fecal occult blood testing can lead to detecting polyps or CRCs because they tend to bleed.^5–7 The efficacy of CRC screening to decrease CRC incidence and mortality has been established by randomized controlled trials (RCTs) of guaiac-based fecal occult blood tests (FOBT) and flexible sigmoidoscopy.⁸ There are no RCT data for colonoscopy, fecal immunochemical tests (FITs), fecal DNA, or CT colonography.⁸ Professional society guidelines strongly support screening, particularly with FITs and colonoscopy.^4,9–12

In the U.S., CRC screening rates of eligible adults are about 69%,¹³ below the 80% target set by the National Colorectal Cancer Round Table.¹⁴ Colonoscopies are used for 90% of CRC screening in the U.S.⁴ FITs, a type of FOBT, are an effective, safe, and inexpensive alternative to colonoscopy for CRC screening.^5,7,12,15,16 A strategy of annual FITs results in far fewer colonoscopies needed over a 25-year interval.^17,18 The use of colonoscopy as a primary screening strategy is very expensive and invasive; the approximately 19 million colonoscopies performed each year in the U.S. cost in excess of $57 billion, which is one reason the U.S. leads the world in healthcare costs.^19–21 The Canadian Task Force on Preventive Health Care explicitly recommends against screening colonoscopy, due to lack of high-quality evidence, greater potential for harms, and the concern that resources will be diverted from those with symptoms.²² FITs are used for CRC screening programs internationally, where individuals receiving positive results are referred for a colonoscopy.^6,23 In order for the U.S. to achieve the 80% CRC screening goal,¹⁴ and given the high cost and inconvenience of colonoscopy, other valid and reliable CRC screening options must be made available, especially in rural and under-resourced areas.

FITs are U.S. Food and Drug Administration (FDA)-cleared class II devices designed to detect intact human hemoglobin in stool samples.^5,6,24 A device can be FDA-cleared by applying to the FDA using a 510(k) application and showing “substantial equivalence” to a pre-existing legally marketed device, called a predicate. In 2020, there were 28 unique FDA-cleared FITs sold under 50 different labels available in the U.S.²⁵ An earlier search showed 26 unique FITs marketed under 65 names.²⁶ Data to guide FIT selection are extremely limited.⁵ It is critical to determine the FIT(s) with the best test characteristics in order to increase screening uptake by implementing successful FIT-based screening programs.^27–29 To address this knowledge gap, we will compare the test characteristics of 4 FITs to detect advanced colonic neoplasia (ACN). Four of the 5 POC FITs with the highest percentages of use in the U.S. in Proficiency Testing (PT) and an automated FIT will be tested using colonoscopy as the gold standard.³⁰

POC FITs make CRC screening accessible to under resourced populations because they are inexpensive, require no preparation, can be collected in the home, and analyzed in a primary care office. POC FITs are read as positive or negative. There are automated FITs that are non-CLIA-waived test that can be reported quantitatively, but in the U.S. they are only reported qualitatively. All FITs use proprietary reagents and antibodies to portions of the hemoglobin molecule. The automated tests measure hemoglobin-antibody complexes by latex turbidimetry.³¹ This comparative effectiveness study, which we are calling BestFIT, will provide essential information about FITs with the best test characteristics for detection of ACN for future expanded use of FITs, which is critically important to achieving the goal of reducing morbidity and mortality from CRC.

Methods

Overview

This is a comparative effectiveness study of 5 FITs for detection of ACN, using colonoscopy as the gold standard. Participants will collect samples for each FIT from a single stool sample. All tests will be run centrally at the University of Iowa. The table shows the FITs in use for the study. They were chosen because of our review of the literature,²⁶ our previous study of CLIA-waived FITs,³² and their high prevalence in Proficiency Testing (PT) program’s data.³⁰ CLIA-certified pathology laboratories participate in PT programs to ensure that procedures for analyte testing give accurate results. Currently, there are 7 PT programs in the U.S. that send blinded samples to laboratories for FOBTs.³⁰

Table.

FITs to be used in the study

FIT	Type of FIT	Hemoglobin Detection Level	Stability & Storage of Sample
Hemoccult ICT (Beckman Coulter, Inc.)	CLIA-waived, dry-slide	300 μg Hb/g feces	14 days at room temperature
Hemosure iFOB (Hemosure,Inc.)	CLIA-waived, liquid-vial	50 ng Hb/mL buffer N/A (μg Hb/g feces)	14 days at room temperature
QuickVue iFOB Test (Quidel Corporation)	CLIA-waived, liquid-vial	50 ng Hb/mL buffer or 10 μg Hb/g feces	8 days at room temperature below 95°F
OC-Light S (Eiken – distributed by Polymedco, Inc.)	CLIA-waived, liquid-vial	50 ng Hb/mL buffer or 10 μg Hb/g feces	15 days at room temperature or 30 days at 36–46°F
OC-Auto FIT (Eiken – distributed by Polymedco, Inc.)	Non-CLIA waived; Automated, liquid-vial	100 ng Hb/mL buffer or 20 μg Hb/g feces	15 days at 77°F

CLIA-waived tests are low-complexity tests designed to be analyzed in physician offices, while the automated test requires a pathology lab certified in running non-CLIA-waived (higher complexity) tests. Because automated FITs would likely be preferred for population-based CRC screening, we had planned to study two automated tests.³³ However, once funding was received in 2017, a company that makes a quantitative FIT was not able to provide kits or the analyzer, so we added the OC-Light S FIT part way into the study. No company has had or will have any influence over the study design, data analyses, or the presentation of results. We did not include proprietary tests that could not be run in our laboratories. Companies have donated FITs and analysis materials, other than the OC-Light S, which we are paying for. The study and methods were approved by the Institutional Review Boards (IRBs) at each of the 3 respective participating institutions.

Study Sample

The study sample will be a diverse total sample of 3,600 individuals aged 50 to 85 years recruited from 3 participating sites: 1) University of Iowa (UI), 2) University of North Carolina at Chapel Hill (UNC), and 3) Texas Tech University Health Sciences Center El Paso (TTUHSCEP). All three centers serve and conduct research with significant numbers of rural patients, and the study sample will be racially and ethnically diverse. UNC expects to recruit 12% African American and TTUHSCEP expects to have 82% Latino. Overall, we expect to enroll 92.3% Caucasian, 4.9% black/African American, 1.3% Asian, and 1.5% other races, with 28% Latino. This recruitment plan was never intended to match proportions in the U.S. census, but rather what we could expect to recruit at 3 diverse academic centers with reasonable budget constraints. All study materials have been translated into Spanish, to help encourage participation of those who are more comfortable with Spanish.

We chose the lower age threshold of 50 years because this was the starting age recommended by most screening guidelines in June 2016 when this proposal was sent to NIH.^34,35 The U.S. Multi-Society Task Force has issued a weak recommendation based on low quality evidence to start screening at age 45 in black/African American individuals,¹¹ and the American Cancer Society 2018 practice guidelines have a “qualified” recommendation for beginning screening at age 45 in all groups.¹⁰ However, this lower age limit has not made its way into mainstream practice and the USPSTF continues to recommend age 50 as the starting age for CRC screening.⁹ The upper age limit of 85 years was chosen because one-third of screening colonoscopies were performed in patients aged ≥ 75 years of age.³⁶ The United States Preventive Service Task Force guidelines recommends personalized decision making for CRC screening for those age 76 to 85 years, taking into account comorbid conditions and past screening.⁹

Recruitment of individuals with a planned colonoscopy is efficient and cost-effective, since the colonoscopies were already scheduled and paid for by other sources. Having the colonoscopy and pathology information will allow calculation of important test characteristics, such as sensitivity and specificity (and not just predictive value positive, as would be the case if only those positive on a FIT received a colonoscopy). A family history of CRC will not result in exclusion of a participant.

Recruitment and Informed Consent

The UI and UNC sites identify individuals aged 50 to 85 years who have been scheduled for a screening or a surveillance colonoscopy using Epic. At these two sites, all patients in the report undergo a subsequent medical record review. Evaluators at the two sites are Research Assistants (RAs) with Master’s degrees. We published a manuscript concerning our Epic algorithm and shared the final algorithm with the University of North Carolina site (UNC).³⁷ At the third site (Texas Tech Health Sciences Center, El Paso), patients scheduled for colonoscopy are seen for face-to-face instruction on how to do the preparation for colonoscopy and will be approached by an RA about the study at that time.

The colonoscopy will have been ordered by the potential participant’s physician. The study invitation mailing includes a cover letter, two copies of the Informed Consent (IC) document (one to be retained by the subject), and a two-page Health Questionnaire to be able to control for possible confounders. Besides age, exclusion criteria related to bowel issues include a personal history of colorectal cancer, ulcerative colitis or Crohn’s disease, rectal bleeding in the previous two months, iron deficiency anemia, history of surgical resection of the colon, personal or family history of a hereditary colorectal cancer syndrome, colonoscopy within the previous 34 months, or the colonoscopy being recommended for diagnostic purposes. We exclude patients with a less than 34-month interval since prior colonoscopy for two reasons. First, a 3-year (or shorter) surveillance interval is typically recommended for patients with advanced adenomas or multiple adenomatous polyps on prior exam. Such patients would be considered higher risk, and a population in which FIT has less relevance compared to colonoscopy. Second, patients who had a colonoscopy in the preceding 34-months without a history of an advanced adenoma (e.g., patients who were prescribed an overly aggressive surveillance interval) are likely at lower risk of harboring polyps than average risk patients.

In addition, we exclude prisoners, nursing home residents, and those with paralysis, dementia, or serious mental health illnesses due to likely difficulties with complying with FIT collection. Following the invitation mailing, using a participant tracking database, RAs at each site telephone those who were mailed an invitation up to three times to invite them to participate and answer all questions about the study. Potential participants can call a site-specific toll-free number for further information if they have questions after receiving the mailing. Individuals deciding to participate return one signed IC and their Health Questionnaire to the recruiting site. As the Health Questionnaires and ICs are returned to each recruiting site, RAs note these in their tracking database, store signed ICs at their site, and electronically upload Health Questionnaire forms to UI securely. The signed IC allows investigators to access a participant’s colonoscopy and pathology reports.

Return of the IC triggers mailing of 5 FITs labeled with unique identification numbers linked to the participant’s study ID, fecal collection materials, detailed directions on how to complete the FITs, and a postage-paid return mailer. FIT samples are collected by participants from a single stool sample before starting their colonoscopy preparation using a plastic hat specimen container to keep the stool sample from contamination with toilet water. All stool samples are mailed directly back to the UI in a postage-paid box for exempt human specimens in accordance with U.S. Postal Services policy.

Patients scheduled for a screening or surveillance colonoscopy at TTUHSCEP are approached by a study research assistant when they attend their pre-colonoscopy assessment visit at one of three sites which do face-to-face colonoscopy directions. This typically occurs one to two weeks before the scheduled procedure. The center’s staff or nurse informs the patient about the study directly or by program flyer at registration. If the patient expresses an interest, study personnel approach the patient and take them to a private area where the study is explained, eligibility is checked and if they are eligible and agree, they complete the IC form, the Health Questionnaire, and receive the FIT testing supplies.

All clinical study documents uploaded to UI are stripped of identifiers, other than the study ID. Our recruitment target is 1200 individuals at each site over a 3.5 year period for a total of 3600 participants.

Health Questionnaire

Each participant completes a two-page, 26-item Health Questionnaire that was adapted from our previous studies.^38–40 Questions were added for height and weight (for BMI calculation), general health, tobacco use, and use of medications which might be expected to cause bleeding in the GI tract. Participants complete this when they mail in their IC to their participating site or complete this face-to-face at the TTUHSCEP. It is designed to be completed within 5 minutes.

Collection and Analysis of FITs

If a signed IC and Health Questionnaire are received, then participants collect 5 FIT samples from one stool sample using a specimen hat and return them using a postage-paid return box. We have prepared detailed directions for FIT collection with clear pictures for each of the selected FIT brands (available once final report has been published). We believe that order of collection will not matter, since the stool is collected in a hat and individuals take their samples for FITs from that.

Received FITs are analyzed daily every weekday on the day of receipt. FITs are not stored. The RA does the analysis without knowledge of the colonoscopy findings. If there are transit intervals greater than specified by individual manufacturers, those results will be excluded from analysis. For the participants thus far, the average days from collection to analysis is 4.1 days with a S.D. of 3.1 days.

Each participant receives $25 for their successful participation (return of IC and completed Health Questionnaire, return of 5 FITs, and colonoscopy completion).

Review and Abstraction of Colonoscopy and Pathology Reports

Colonoscopy and pathology reports are linked with corresponding FIT results and abstracted in a standard manner by UI site investigators using the Colonoscopy/Pathology Review form (available on request) adapted from our previous studies.^32,40 This form allows standardized collection of information including detailed collection of information regarding the colonoscopy and pathology information, for each biopsy.

Participants are having either a screening or surveillance colonoscopy. Investigators considered including only individuals undergoing screening colonoscopy, but this would have substantially reduced the eligible subject recruitment pool due to the large number of individuals with a history of an adenomatous polyp. Our data suggest that there is little difference in the rate of advanced adenomatous polyps between those having a screening versus surveillance colonoscopy (6.7% versus 8.1%, p=0.39).³² In our earlier study, our overall rate of advanced adenomas (7.1%) was almost identical to that of Imperiale, et al. (7.6%).^32,41

At each of the three sites, colonoscopies are performed by board-certified gastroenterologists. If fellows are performing the procedure, the gastrointestinal staff are always present in the room the entire time.

Pathology samples are identified by site of colon biopsied: cecum, ascending colon, hepatic flexure, transverse colon, splenic flexure, descending colon, sigmoid, rectosigmoid junction, rectum, random colon biopsy, or other. Lesions are classified as either proximal (cecum, ascending colon, hepatic flexure, or transverse colon) or distal (splenic flexure, descending colon, sigmoid, rectosigmoid junction, or rectum). Samples are also classified by type and size of polyp. Histology readings are recorded according to the most advanced pathology in the sample: no diagnostic abnormality, lymphoid aggregate, polypoid colonic mucosa; hyperplastic adenoma; tubular adenoma, sessile serrated adenoma, traditional serrated adenoma; tubulovillous, or villous adenoma; high grade dysplasia; adenocarcinoma. We will exclude exams with poor or inadequate prep, and incomplete exams (except when exam is incomplete due to obstructing cancer).

There is no planned centralized review of adenocarcinomas or advanced adenomas in the current study. All pathology samples are being read by board-certified Pathologists at academic centers, with expertise in GI pathology. Thus, the chances for an inaccurate reading are minimized and this is the clinical standard of care. CRCs will be classified according to American Joint Committee on Cancer stage.⁴²

Case Definitions for a Positive Colonoscopy

The case definition for a positive colonoscopy is any advanced colorectal neoplasia (ACN) defined as: any advanced pre-cancerous lesion (adenomatous or sessile serrated polyps 10 mm or greater in size), villous or tubulovillous, or traditional serrated adenoma of any size, any lesion with high grade dysplasia, or any stage adenocarcinoma.⁴³ An individual is considered to have a negative colonoscopy if they do not meet the case definition for a positive colonoscopy.⁴⁴ Colonoscopy results are summarized for each individual according to the most advanced lesion identified.

Statistical Analyses

Demographic and clinical data will be summarized using standard descriptive statistics. The main outcome (dependent) variables are the FIT screening results (positive or negative test result) and the colonoscopy case definition results, which combine to give us sensitivity, specificity, and false negative and positive rates. By incorporating prevalence data into Bayes’ Theorem, we can also examine positive and negative predictive value rates. We plan to compare FIT characteristics by Latino vs. non-Latino, by age groups of 60 and older vs. < 60 years, and men vs. women.

We completed a pilot study sending 5 FITs with our detailed, written and pictorial instructions to 30 individuals scheduled for a colonoscopy, and received 15 back (50%). We planned our mailings for the study based on this 50% response rate.

The sensitivity and specificity for each of the five qualitative FIT results will be estimated, along with 95% confidence intervals, based on the normal approximation to binomial probabilities or using exact permutation test-based confidence intervals (e.g., available in SAS Proc Freq). The sensitivity and specificity will each be compared across the five FITs using generalized linear mixed models that extend logistic regression to accommodate random effects (e.g., SAS Proc Glimmix or Proc Nlmixed). In these models, the random effects will be the participants, while the fixed effects with be the FIT, as well as demographics such as age, sex, and ethnicity. With these models, we can also test if there are differences across sites, look for interactions among covariates, and accommodate missing data resulting from participants who do not submit valid specimens from all five FITs.

For the automated FIT, the quantitative results will be used to construct ROC curves for each of the case definitions. These ROC curves will be fit via Proc Logistic and then compared using the SAS ROC macro that is based on the work by DeLong et al.⁴⁵

We chose our sample size to give reasonable precision in estimating sensitivity and specificity for screen-relevant CRC lesions, defined as advanced colorectal neoplasia (ACN). With a sample size of 3600 and an estimated prevalence of approximately 7%, we expect to have n=252 patients with screening relevant neoplasia and n=3348 without. For calculating 95% confidence intervals using the normal approximation, the margin of error is 1.96* $\sqrt{p(1-p)/n}$, where p is the test characteristic of interest (e.g., sensitivity or specificity). For p of 0.20 (or 0.80), this margin of error would be 4.9% for sensitivity and 1.4% for specificity. The highest margin of error occurs at p=0.5, with 6.2% for sensitivity and 1.7% for specificity.

We plan to recruit 28% Latino individuals. Assuming similar prevalence of screening relevant neoplasia, we would have n=71 Latino patients with screening relevant neoplasia and n=937 without. For the highest margin of error, when p=0.5, we will have a margin of error of 11.6% for sensitivity and 3.2 for specificity.

Results to Date

As of February 18, 2021, we have a total of 4027 informed consent documents returned; 1620 from UI, 1091 from UNC, and 1316 from TTUHSCEP. We have complete data (returned FITs and completed colonoscopy) from a total of 2990 individuals; 1257 from UI, 792 from UNC, and 941 from TTUHSCEP. A total of 30.3% are Latino and 5.3% are black/African American, and 1.6% are Asian. The average age for 2990 subjects is 62.1 years (SD 7.5; median 61.7), 62.8% are female and 37.2% are male. We will present full results once the study is completed.

Discussion

Our focus in this study is on point-of-care FIT testing with 4 of the 5 tests being studied being tests that can be analyzed at the point-of-care in the family medicine setting. Although these tests have been shown to have acceptable performance, the main screening test used in the U.S. is colonoscopy. Use of FITs will allow broader segments of the population to access colorectal cancer screening because these tests require no preparation, are inexpensive, and can be done in the privacy of one’s home. FITs have limitations including: 1) if positive, the need for adherence and access to diagnostic colonoscopy and 2) the need for adherence to repeat testing on an annual/biennial basis and the systems to support such repeat testing. If FITs are to be included in the CRC screening armamentarium, then programmatic screening must be the goal. This is the preferred method in many countries where medical resources are more closely monitored as compared with the U.S. Increasing screening uptake will reduce the burden of advanced adenomas and colorectal cancer.^5–7 Most screening in the U.S. is being done by colonoscopy, which is expensive, has an onerous bowel prep, requires a full day off from work and a driver, and is associated with complications.^4,19–21 Routinely offering screening colonoscopy for average-risk patients is impractical in under-resourced areas and federally qualified health centers.⁴⁶ FIT is an excellent alternative, it is acceptable, inexpensive, and RCT have shown the efficacy of stool-based screening (and not colonoscopy).⁸ However, if FIT testing, particularly point-of-care testing, is being used to increase screening rates, it is critical to know that these tests have the performance characteristics to suggest that they will be effective in reducing the burden of CRC—not just preventing unnecessary colonoscopies. A study in a Veterans Administration population suggests that FIT may be useful in targeting individuals for colonoscopy but did not collect data to ascertain test characteristics of the automated FIT used.⁴⁷

Of the 50 different FIT labels available in the U.S., 16 of these comprise nearly 90% of the FOBT market.⁴⁸ Very few of these FITs have peer-reviewed published results of their performance characteristics in detecting CRC or advanced adenomas,^5,32,41 since they only must show “substantial equivalence” to previously cleared tests. The “substantial equivalence” determination is made using samples spiked with human blood, not fecal samples from participants in a study who undergo colonoscopy. In addition, data on test performance in average-risk individuals are not in FDA files and there is no post-marketing surveillance by the FDA. Specifically, there have been only two studies on U.S. populations using colonoscopy as the gold standard, assessing test characteristics of FITs currently cleared in the U.S.^32,41 Our study of four CLIA-waived tests found relatively low sensitivity across devices for advanced adenomas and cancer.³² Two of those devices are no longer on the market and one still on the market had a sensitivity of only 4% for advanced adenomas.³² Imperiale, et al. evaluated a proprietary automated FIT, which would likely be preferable for population-based screening, and found more favorable results after results were combined with DNA markers.⁴¹

The accuracy of screening tests is extremely important, but tests that are accurate most of the time can still have substantial numbers of false positive (FP) and false negative (FN) results. Predictive values are functions of test sensitivity and specificity and the disease prevalence in the study population. FP FIT results lead to unnecessary follow-up colonoscopies and FN FITs mean advanced polyps or CRC are being missed. Because the FDA has not required companies to have data for FIT clearance in people with real colonic lesions, as identified by colonoscopy and accompanying pathology, these test characteristics are not known. Thus, this study addresses a critical gap in the literature.

Our review of FIT literature found that test characteristics varied widely within and across brands. In general, the automated tests had better performance than the CLIA-waived FITs.²⁶ Identical devices have the same 510(k) number, but often are sold by several different companies under various labels, making it impossible for physicians and laboratory personnel making screening program purchasing decisions to make an informed decision.

Studies done on FITs in other countries often used FITs not available in the U.S., high-risk populations,^28,29,49,50 individuals new to screening,^51,52 or frozen stool samples (not recommended by the manufacturer) which can lead to erroneously higher hemoglobin levels, in addition to many studies on largely overlapping study populations.^51,53 Researchers in Hong Kong published 3 papers on test characteristics of Hemosure, a qualitative FIT, using many of the same individuals new to screening in their largely overlapping study populations.^54–56 Thus, these results are not applicable in the U.S., and do not comprise as many different populations as appears at first glance. It is critical to determine the FIT(s) with the best test characteristics in order to implement successful FIT-based screening programs in the U.S.

In the U.S., the automated FIT is not cleared to read quantitatively, unlike other countries which have automated FITs that provide a quantitative reading.^{28,50,52,57–59} In these countries, it has been found that more advanced colonic lesions are associated with higher amounts of blood in the stool and some countries adjust these cutoffs so that the number of individuals recommended for colonoscopy does not overwhelm available resources.

Strengths of this study include assessing how well each of 5 different FITs detect advanced colorectal neoplasia with single-sample testing. While it would be more powerful to test FITs in a longitudinal fashion, such a study would be extremely expensive. Weaknesses include the fact participants are selected using both a screening and surveillance population, which would be expected to possibly have different rates of adenomas. Given the heavy use of colonoscopy at these centers, it would have been very challenging to select a pure screening population. Our previous study did not show a difference in ACN in a screening vs. surveillance population. However, other studies have shown a higher adenoma detection rate with surveillance as compared with screening colonoscopies.⁶⁰

This BestFIT study is a 5-year prospective comparative effectiveness trial of five FITs for detection of screen-relevant CRC lesions with results expected in March 2022. This study will assess important test characteristics of 5 FIT products, 4 of which are point-of-care products which can be analyzed in family medicine offices. If these have acceptable test characteristics, then more individuals can be screened for colorectal cancer, since colonoscopy is a very expensive modality out of the reach of those with little or no insurance. We will recruit a diverse sample of 3600 individuals, aged 50 to 85 years from each of three sites using colonoscopy as the gold standard. Investigators at study sites were chosen for their expertise in CRC screening and ability to successfully recruit participants and conduct study procedures.^{30,32,38,40,61–73} Participants will use 5 FIT collection devices on a single stool, FITs will be analyzed centrally, and FIT characteristics determined using colonoscopy as the gold standard.