Point of Care Faecal Immunochemical Testing for Colorectal Cancer

Introduction

The demand for early diagnosis of bowel cancer

In the United Kingdom, there are an estimated 41,700 new cases of bowel cancer diagnosed each year and it is the fourth most common cancer. Over 90% of patients that are diagnosed with the earliest stage of bowel cancer will survive at least five years. In 2016 NHS England made £200 million available to ensure we diagnose cancer earlier and faster.

Bowel cancer in symptomatic patients

Patients who present to their GPs that are anaemic or with symptoms of change in bowel habit, rectal bleeding, abdominal pain, bloating or weight loss are often referred to a colorectal clinic under the two-week rule (TWR) scheme. This is designed to enable patients to be seen within two weeks from primary care referral by a specialist clinician to enable quicker diagnosis of a cancer should it be present. Many of these patients will be referred by the colorectal clinician to have a colonoscopy or other form of endoscopy as part of their investigation and work up.

The National Institute for Health and Care Excellence (NICE) has set out guidelines for patients to be referred via the TWR, these are known as NG12 criteria. The number of patients referred for this service is increasing nationally and this in turn is creating significant burden on endoscopy services. In 2015, 1.7 million endoscopies cases were estimated for that year across England and this is expected to increase to over 2.4 million per year by 2020. Of the TWR colorectal referrals, only 3-4% will end up with a diagnosis of colorectal cancer (CRC) and only 27% of all colorectal cancers are picked up from these referrals nationally.

Once a patient is on the pathway, not only do they have to be seen within two weeks, but also a diagnosis must be made by 28 days. This leaves no realistic opportunity for any reasonable form of watch and wait approach for those that a clinician may have a low suspicion of CRC or other significant colorectal disease.

Colonoscopy is currently seen as the gold standard for investigating symptomatic patients to rule out bowel cancer. CT colonography is also widely used, yet both require bowel preparation to be taken the day before. This is inconvenient for patients and can cause dehydration and possible kidney injury to the frailer patients. In particular, colonoscopy is unpleasant and can be painful. There is also a small risk of perforation (up to 1 in 500), the consequences of which can be very serious. The cost of each diagnostic colonoscopy procedure to the NHS is approximately £372.9

The Royal Surrey County Hospital (RSCH) has been particularly impacted by the increased referrals of TWR patients to colorectal clinic. Recent analysis from our cancer database shows that in the last financial year, we received 1414 referrals, of whom 51 patients were found to have colorectal or anal cancer. The average for the years 2009-2015 was 709 referrals for 53 cancers per year. This shows that the number of referrals has doubled, yet we do not pick up any extra cancers. Of these referrals it was noted that 80.6% were primarily investigated by endoscopy, this has increased from 75% found in previous years. An initial and less invasive "rule out" test with near instant results could enable a decision to be made regarding triage and onward referral to endoscopy. Such a test with a high sensitivity and high negative predictive value will safely reduce referrals to a more manageable number without adverse risk.

2. Background to Faecal Immunochemical Testing (FIT)

Bowel Cancer and Screening

For some in the UK there is an opportunity for 'bowel scope screening' via flexible sigmoidoscopy as a one-off test from the age of 55 to 60 depending on local availability. Bowel cancer screening is uniformly available in the UK to all those aged 60-74. Traditionally, a guaiac-based Faecal Occult Blood test (gFOBT) has been carried out, which determines the presence of haemoglobin (Hb) in stool. A positive result warrants a colonoscopy.

Screening is changing across the UK with the enrolment of Faecal Immunochemical Testing (FIT). This is also a test from a stool sample, but offers a quantitative concentration result as opposed to a qualitative positive/negative result. Concentration of haemoglobin in stool has been showed to correlate with severity of colorectal neoplasia and other significant colorectal disease (SCD) such as high risk adenomas and colitis. Unlike the gFOBT, FIT is a more sensitive testing model and the quantitative cut off concentration can be set to determine the number sent for colonoscopy according to resource availability and likelihood of missing a cancer or SCD. It is only sensitive to human haemoglobin and therefore it is not affected by diet. FIT also only requires one sample compared to two sets of three that are required for gFOBT. It is therefore easier for the patient personally and produces better compliance.

Use of FIT in Symptomatic Patients

FIT is a potential first line investigation for symptomatic patients as a "rule out" test, thus reducing the demand for colonoscopy. The advantage of the quantitative result is that the concentration (micrograms Hb/g faeces) has been shown to directly corelate with the presence of significant colorectal disease including high risk adenomas and colitis, with malignant neoplasia having the highest concentrations. In 2017, NICE published a diagnostic guideline (DG30), which recommend the use of FIT for the detection of FOB in the NG12 criteria and a value of >10micrograms Hb/g faeces is seen as positive and warrants further investigation. There are many forms of FIT analysis, but only four laboratory-based machines are endorsed in the criteria. This can act as guidance for patients under investigation in primary care with symptoms. However, once a patient is referred on the TWR pathway, the NG12 criteria state that patients be referred for further investigation without further recourse to FIT12. The majority of FIT studies relate to those amongst the screening population and so the evidence cannot be used in symptomatic patients. There are 10 small studies pertaining to FIT in the symptomatic regarding the DG30 guidance, these total approximately 8,000 patients. Other larger national studies are currently underway that aim to address this issue as well as encourage its use further in primary care.

It seems that FIT may be valuable for all referrals to the TWR colorectal pathway. However, the NG12 criteria also includes patients that may not have symptoms as it also incorporates patients that are shown to be anaemic with or without iron deficiency. Patients that do not have a recognised cause of anaemia come on to the pathway as the low serum Hb could be related to a tumour that is bleeding in a subclinical fashion, particularly if the tumour is on the right side of the colon. Extra caution for use of FIT should be given to these patients as blood in the stool from the right side of the colon is older and the haemoglobin may have been subjected to degradation and therefore not detectable using FIT.

Study Opportunity

Point of Care FIT

FIT has been developed as a point of care (POC) investigation by a number of manufacturers and test kits are readily available for use by purchase online or over the counter. Some are designed for home use and some for professional use in a clinical setting. Home FIT kits only offer a qualitative result, which avoids the delay and cost of a laboratory test, however generally have a higher cut off concentration than 10micrograms Hb/g faeces for positive or negative and are also open to user interpretation errors. The cost ranges from £2.50 to £34.99 (median £4.09), which is competitive with the average £9.42 for a laboratory-based FIT. The POC professional kits differ, with some offering qualitative results and some work to give a quantitative result via a portable machine or even mobile phone app.

Identified Requirement

An instant quantitative result in the clinical setting has the potential advantage over the laboratory test in terms of triaging and referring patients for ongoing investigations immediately. This is because the decision can be made in the one clinic setting rather than the extra step of waiting a few days for the laboratory-based result to be determined. Consequently, it is those machines that are designed for POC professional use that we are interested in taking forward into trial. However, there is no scientific study that has been done on these test kits to prove that they are accurate.

The POC FIT kits that are designed for professional use do not all detect the very low haemoglobin concentrations that the laboratory-based machines can pick up, but the manufacturers do report the ability to pick up positive results as low as 10microgram Hb/g. This would be useful for primary care practitioners as that is in keeping with the DG30 guidance. Many of the available POC FIT kits have a "Conformite Europeene" (CE) approval but have not had Food and Drug Administration (FDA) or other approval. The potential for harm is high if either an inaccurate reading falsely indicates a high faecal Hb resulting in an unnecessary colonoscopy, or more seriously a false negative result reassures the patient and clinician that they do not have CRC.

The Southern Bowel Cancer Screening Hub (that is affiliated with RSCH) uses all the laboratory-based machines that are recognised by NICE for FIT from the DG30 update and they are internationally validated. The Hub has identified the need to assess the professional use POC FIT kits against the standard set of baselines used by the laboratory machines. This will help determine key criteria such as accuracy, effectiveness and safety for their use in the decision-making process for onward investigation.

This protocol describes a study for a POC FIT machine that aligns with key criteria set out by the Hub for use in the TWR colorectal clinic.

Digital Rectal Examination Sampling

Although FIT only requires a single sample, we recognise that not all patients will comply with being able to perform this at home. It has been advocated that the digital rectal examination (DRE), which is part of routine assessment in the TWR clinic provides enough stool on the glove to wipe on to a sample stick. It would therefore be useful to assess which of the two sampling methods provide the best compliance and most accurate results.

Rationale for study

• The referral rate to the TWR clinic via NG12 guidelines is dramatically increasing.
• This has not only increased our patients that are needed to be seen in clinic, but also had a large effect on our endoscopy burden.
• If proven to be of value in secondary care clinic, then these results will be transferable to the primary care clinic setting.
• This study needs to be performed in secondary care to obtain the colonoscopy result for comparison.
• The NG12 criteria for these patients once referred into secondary care does not currently advocate occult blood testing as an investigative tool due to lack of evidence.
• POC testing over laboratory testing offers a one-stop investigation that can give the clinician an instant result to help make a decision in the clinic regarding need for endoscopic evaluation.
• A close correlation between POC FIT tests and endoscopy results for CRC will add to the body of evidence regarding FIT as a triage and referral tool in symptomatic patients for primary and secondary care.
• This will be safer and more convenient for patients in terms of reducing the need for endoscopy.
• The burden on NHS funding and resources will be significantly reduced if the demand for clinic time and colonoscopy is reduced.