Why FIT?
The Faecal Immunochemical Test (FIT) is a cheap, non-invasive test that has been shown in peer-reviewed studies to exclude colorectal cancer in symptomatic patients. While multiple studies across Europe have demonstrated these results, further evidence is needed to show this test could be used within the UK. This study aims to establish the diagnostic accuracy of FIT in symptomatic patients referred with suspected colorectal cancer under NICE guidelines in London.
BACKGROUND
In 2014-15, 262,745 patients were referred with suspected colorectal cancer (CRC), of whom 93.5% of whom were seen within 14 days [1]. Patients are seen after a referral on the two week wait (2WW) pathway for suspected cancer. The referral criteria for patients with suspected CRC had an expected positive predictive value of 3-4% [2]. For every 100 patients referred with possible CRC for investigation, 97 would not have bowel cancer. While this yield may be regarded as low, it was not seen as unacceptable. Early CRC diagnosis is critical to improving survival. Despite improvements in pre-operative imaging, neoadjuvant and adjuvant therapy, surgical techniques and pathological quality control, ultimately disease stage is still the dominant factor in survival (Figure 1) of CRC.
Figure 1: Five-Year Relative Survival (%) by Stage, Adults Aged 15-99, Former Anglia Cancer Network. Data courtesy of CRUK [3]
Detection and subsequent treatment of cancers at an earlier stage is essential to improve a patient’s chance of cure. Twenty five percent of patients with bowel cancer are diagnosed during an emergency presentation [4], which will be at a late stage. Their survival at 3 years is half that of patients referred by 2WW (34% vs 68%), which is still not as good as patients diagnosed during bowel cancer screening (92%, Figure 2).
Figure 2: Three-Year Relative Survival (%) by route of diagnosis. Data courtesy of Public Health England [4].
Only 27% of patients with bowel cancer are identified by the 2WW [4]. Organisations such as Bowel Cancer UK therefore called for guidelines to be expanded to include patients who do not present with high risk symptoms [5]. The expanded NICE criteria include more non-specific gastrointestinal symptoms; abdominal pain or weight loss (<50 years of age), change in bowel habit (>60 years of age) and non-iron deficiency anaemia (>60 years of age). Patients who meet NICE criteria for suspected CRC are referred to a gastrointestinal specialist, where a colonoscopy is usually requested to investigate the patient.
Colonoscopy remains the gold standard test for detection of CRC and is the first line test in 68% of patients referred with suspected colorectal cancer [5]. If a suspected cancer is found, it will be biopsied for histological confirmation, and tattooed to enable intra-operative localisation. Smaller polyps may also be removed and other bowel conditions diagnosed.
However, colonoscopy is not a perfect test. It can cause considerable inconvenience and potential harm to patients. Patients will be awake for the majority of the previous night after ingesting bowel preparation. Elderly patients frequently struggle to mobilise to the bathroom and may become dehydrated. Patients will require a day off work if taking sedation. The procedure itself can be painful. The risk of bowel perforation during the procedure is rare at approximately 1 in 2000 [6], but serious. Furthermore, endoscopy units in the UK are struggling to respond to the additional demand from Bowel Scope and the bowel cancer screening programmes[7]. Endoscopy demand is expected to expand from 1.7 million GI endoscopy cases by a further 750,000 cases per year by 2020[7]. The Bowel Scope screening programme is expected to increase the number of flexible sigmoidoscopy procedures by 325,000 patients by 2017 [8]. Financing the expansion of endoscopy capacity is particularly challenging during this period of efficiency savings in the NHS.
The NHS therefore faces a daunting struggle to improve CRC survival in symptomatic patients. It has further broadened the referral criteria to improve the detection rate of symptomatic CRC. However, this has resulted in an increase in referrals, which has not been met by an increase in endoscopy capacity to investigate these patients. If endoscopy is unable to meet the demand created by the new referral criteria, performance metrics such as the 2WW or 62-day treatment targets will not be met. Endoscopy services will also struggle to process cases referred from the bowel cancer screening programme, where cancers can be detected at the earliest possible stage. All of this will ultimately lead to poor patient satisfaction and patient experience.
Other tools exist that can exclude bowel cancer. The Faecal Immunochemical test is the most promising example we have chosen to study.
The Faecal Immunochemical Test (FIT)
FIT detects the globin component of haemoglobin using antibodies. Immunoassays then measure the development of antibody-globin complexes. These can give binary or quantitative (FHb level) readings. While some degree of gastrointestinal bleeding is physiological, FHb concentration is related to the severity of disease, and highest in CRC [9, 10].
FIT has major advantages over the Faecal Occult Blood Test (FOBT), the older and current faecal blood test. FIT only requires one stool sample. The process can be automated, with end points that are objective, reproducible and can be easily quality-assured. FIT is specific for lower GI bleeding as globin from the upper GI tract is degraded by digestive enzymes. It is not subject to interference from dietary haem or peroxidases, nor medications. As FIT levels increase with serious bowel neoplasia, from adenoma to cancer, FHb cut-off levels can therefore be calibrated towards higher sensitivity (no false negatives) or higher positive predictive value (indicating a higher risk of CRC). Systematic reviews have shown that FIT is more accurate for the diagnosis of CRC than the currently used FOBT with a higher sensitivity [11]. This means that a negative test could reliably rule out CRC and a referral for colonoscopy would therefore not be required. In the same NICE analysis that recommended FOBT as a triage tool for symptomatic patients, FIT was estimated to be the most cost-effective test for detecting CRC, but was not included in the main analysis due to a lack of evidence.
Despite these advantages, FIT is again not a perfect test. The sensitivity of FIT indicates that it will detect physiological or non-malignant bleeding. Therefore FIT sensitivity varies by gender, age, ethnicity [12] and possibly social deprivation[13]. It also varies by collection and storage techniques, the number of samples tested, or the analyser and process employed. Consequently, the correct choice of cut-off level to warrant further investigation must be validated by standardised data collection from each population. FHb may be undetectable in cases of high risk adenoma or inflammatory bowel disease (IBD), and levels in CRC vary. While a higher FHb cut-off level will result in a high positive predictive value of the test for CRC, CRCs or high risk adenomas may be missed. Choosing a lower cut-off level will result in a better CRC and adenoma detection rate, but also more false-positive results which will require a greater colonoscopy capacity to investigate but still this will be significantly less than the demand of the current approach.
No stool sample test is pleasant. FIT is more acceptable with significantly higher uptake amongst screening patients in the UK[14]; patients need to dip a sample stick once into their stool, while FOBT requires patients to smear samples of stool onto cardboard from three separate bowel motions. FIT is more expensive than FOBT (estimated costs of £9.42 vs £4.86 respectively), but certainly cheaper than an unnecessary colonoscopy £421 [15]. Mowat et al showed in Scotland that a negative FIT result in symptomatic patients could reduce the number of referrals by over 40%, freeing up endoscopy capacity in a overburdened NHS [16].
NICE + FIT = NICE FIT?
FIT offers the NHS to move into 21st century medicine, saving costs and providing better care. Patients presenting with bowel symptoms to their GP practice can be given a FIT testing kit, and have a result within days. Patients with undetectable or very low FHb levels can be reassured that no further investigation is needed for bowel cancer with close to 100% certainty. This will reduce the cost of specialist time from screening referrals and outpatient appointments, and free up endoscopy resources to offer rapid access to patients with higher FHb levels, as well as bowel cancer screening services in line with NHS England’s goal for early cancer diagnosis in its “Five Year Forward View”.
There is an accumulating body of evidence [16-19] that FIT could rule out CRC in symptomatic patients. However, FHb reference values cannot be transferred across programmes from other countries. Although FIT has consistently shown superior diagnostic accuracy to FOBT in studies from across the world, it cannot be rolled out across the NHS until reference values for age, sex and ethnicity have been derived for our own population by carrying out large-scale diagnostic studies. London in particular, has one of the most ethnically diverse populations in the world. Once the diagnostic accuracy of FIT has been verified for our own population, NICE will hopefully recommend its use in symptomatic patients.
REFERENCES
1. England, N., Waiting Times for Suspected and Diagnosed Cancer Patients. 2015-16 Annual Report. 2016.
2. NICE, Suspected cancer: recognition and referral. 2015.
3. UK, C.R. 2017 01/2017]; Available from: http://www.cancerresearchuk.org/health-professional/cancer-statistics/statistics-by-cancer-type/bowel-cancer/survival#heading-Three.
4. NCIN, Routes to diagnosis 2015 update: colorectal cancer, P.H. England, Editor. 2015.
5. UK, B.C., Diagnosing Bowel Cancer Early. Right Test, Right Time. . 2014.
6. Rutter, M.D., et al., Risk factors for adverse events related to polypectomy in the English Bowel Cancer Screening Programme. Endoscopy, 2014. 46(2): p. 90-7.
7. UK, C.R., Scoping the Future. An evaluation of endoscopy capacity across the NHS in England., S.U.a.N.M.a.L.C.S.U. Health Services Management Centre at the University of Birmingham, Editor. 2015.
8. PH-HSB-SH, S.E.-. Improving Outcomes: A Strategy for Cancer – Third Annual Report, D.o. Health, Editor. 2013.
9. McDonald, P.J., et al., Faecal haemoglobin concentrations by gender and age: implications for population-based screening for colorectal cancer. Clinical Chemistry and Laboratory Medicine, 2012. 50(5): p. 935-940.
10. Fraser, C.G., Screening for colorectal neoplasia with faecal tests. The lancet oncology, 2011. 12(6): p. 516-517.
11. Launois, R., et al., Systematic review and bivariate/HSROC random-effect meta-analysis of immunochemical and guaiac-based fecal occult blood tests for colorectal cancer screening. European journal of gastroenterology & hepatology, 2014. 26(9): p. 978-989.
12. Fraser, C.G., et al., Faecal haemoglobin concentrations vary with sex and age, but data are not transferable across geography for colorectal cancer screening. Clinical Chemistry and Laboratory Medicine (CCLM), 2014. 52(8): p. 1211-1216.
13. Digby, J., et al., Deprivation and faecal haemoglobin: implications for bowel cancer screening. Journal of medical screening, 2014: p. 0969141314535388.
14. Moss, S., et al., Increased uptake and improved outcomes of bowel cancer screening with a faecal immunochemical test: results from a pilot study within the national screening programme in England. Gut, 2016.
15. Cancer, N.C.C.f., Suspected cancer: recognition and referral. . 2015.
16. Mowat, C., et al., Faecal haemoglobin and faecal calprotectin as indicators of bowel disease in patients presenting to primary care with bowel symptoms. Gut, 2015: p. gutjnl-2015-309579.
17. McDonald, P., et al., Low faecal haemoglobin concentration potentially rules out significant colorectal disease. Colorectal Disease, 2013. 15(3): p. e151-e159.
18. Cubiella, J., et al., Development and external validation of a faecal immunochemical test-based prediction model for colorectal cancer detection in symptomatic patients. BMC medicine, 2016. 14(1): p. 128.
19. Auge, J.M., et al., Clinical utility of one versus two faecal immunochemical test samples in the detection of advanced colorectal neoplasia in symptomatic patients. Clinical Chemistry and Laboratory Medicine (CCLM), 2016. 54(1): p. 125-132.