Detection capability of quantitative faecal immunochemical tests for haemoglobin (FIT) and reporting of low faecal haemoglobin concentrations

Abstract

Faecal immunochemical tests for haemoglobin (FIT) are widely used in asymptomatic population screening for colorectal (bowel) cancer. FIT are also used to assist with the assessment of patients presenting with lower abdominal symptoms. Quantitative FIT allow the generation of numerical estimates of faecal haemoglobin (f-Hb) concentrations. There is now great interest in "low" f-Hb concentrations in these clinical settings: in consequence, knowledge of the detection capability is very important for f-Hb concentration examinations. There are a number of current problems associated with the reporting of low f-Hb concentrations and wide misunderstanding of the metrological aspects of examinations of f-Hb at low concentrations. These would be solved if the detectability characteristics of f-Hb concentration examinations, namely, the limit of blank (LoB), limit of detection (LoD) and limit of quantitation (LoQ), were generated, validated and used in reporting systems exactly as recommended in the EP17-A2 guideline of the Clinical Laboratory Standards Institute. LoB and LoD are statistical concepts, but the LoQ depends on definition of analytical performance specifications (APS). In this Opinion Paper proposals for interim APS are made, based on the current state of the art achieved with examinations of faecal samples. It is proposed that LoQ is determined at an examination imprecision of CV≤10% using faecal samples naturally positive for Hb rather than faeces spiked with haemolysate. Detailed proposals for reporting f-Hb data at low concentrations are also made.

Faecal immunochemical test is superior to symptoms in predicting pathology in patients with suspected colorectal cancer symptoms referred on a 2WW pathway: a diagnostic accuracy study

“…FIT detects the globin component of haemoglobin (Hb) by immunoassay and can reliably measure the faecal Hb concentration (f-Hb) to the nearest microgram of Hb per gram of faeces (µg/g). 5 Since 2010, over 25 diagnostic accuracy studies have reported data on the use of FIT in symptomatic patients utilising a range of cut-offs. [6][7][8] In 2014, a study of 787 symptomatic patients from Spain suggested that FIT is more accurate for the detection of CRC than NICE 2005 criteria (CG27) although NICE have since expanded its referral criteria to include lower risk symptoms (NG12).…”

Faecal immunochemical tests for haemoglobin (FIT) in the assessment of patients with lower abdominal symptoms: current controversies

“…and a series of proposals made including the following:40  f-Hb concentrations should not be reported to more significant figures than whole integers,  f-Hb concentrations less than the limit of detection (LoD) of the FIT analytical system should be termed "not detected" or "undetectable",  for academic use: f-Hb greater than the LoD could advantageously be documented,  for routine clinical use: numerical f-Hb should be reported only when greater than the limit of quantitation (LoQ): f-Hb less than the LoQ (x) should be reported as < x μg Hb/g faeces and  if a more sophisticated reporting system is required, one option is to report:o f-Hb < LoD = f-Hb not detected o LoD < f-Hb < LoQ = f-Hb detected o f-Hb ≥ LoQ = report the found f-Hb Adherence to these proposals would result in harmonisation of the reporting of f-Hb data, which should facilitate understanding and transferability of information across geography and time.A further controversy is whether the high sensitivity makes FIT a good rule-in test for CRC or the high negative predictive value (NPV) shown in many studies on use of FIT in assessment of the symptomatic10,11 demonstrates that FIT provides a good rule out investigation for SBD. In reality, f-Hb in patients presenting in primary carer with lower abdominal symptoms could be considered as a continuous variable.…”

Faecal immunochemical testing (FIT): Sources of analytical variation based on three years of routine testing in the context of DG30

“…Current proposals for FIT testing at low concentrations are dependent on accurate LOD and LOQ estimates (7). In this evaluation we have derived these detection characteristics for a commonly used FIT method.…”

“…There is on-going debate as to how FIT results should be reported from those that consider the result as "detected"/"not detected" through to quantitative reporting of numeric values (4,7). Given the high imprecision of analysis when sampling is taken into consideration we would caution against quantitative reporting of faecal Hb due to the high uncertainty of measurement.…”

“…The threshold recommended for use with symptomatic patients in DG30 is 10 µg/g. FIT methods may not be optimised, or fully characterised, for detection at the lower concentrations now being considered important for application in DG30 (7).…”