Abstract
Faecal Immunochemical Testing (FIT) can identify the presence of faecal occult blood (FOB). It is a screening tool that identifies those at risk of having colorectal cancer (CRC). Initial roll out of FIT was for the asymptomatic screening population, but recently it has been advocated as a ‘rule out’ test for those with bowel symptoms. FIT is also known to identify those who may have other forms of serious bowel disease (SBD), this includes high risk adenomas and inflammatory bowel disease.
Traditionally, FIT is performed in laboratories with analysers that have a high throughput, but there are also point-of-care (POC) analysers that are on the market for professional use. These devices have little evidence with regards to their application and diagnostic accuracy.
This single centre observational study evaluated POC FIT in a symptomatic cohort considered at risk for CRC. Patients were recruited from those referred on the two-week wait (TWW) pathway at the Royal Surrey NHS Foundation Trust. An optimal POC analyser (the QuikRead go®) was identified, which can detect FOB, and quantitate it as a concentration like laboratory-based FIT analysers. Faecal haemoglobin (f-Hb) results are reported as μg haemoglobin/gram of faeces (μg/g). This enables different cut-offs for positivity to be set. The lower limit of detection for the device is <10 μg/g. The performance of the analyser was assessed by receiver operator characteristic (ROC) curves.
In total, 633 patients provided paired FIT samples for the POC QuikRead go® and a laboratory-based FIT system (FOB God Wide®). Of these, n=553 obtained definitive diagnostic outcomes (after lower gastrointestinal endoscopy or CT colonography), which enabled comparison to the performance of the prior FIT results.
There were 14 diagnoses of CRC and 52 with SBD. Using <10 μg/g as a cut-off, the sensitivity for the QuikRead go® analyser was 92.9% (95% CI 68.5%–98.7%) for CRC and 76.9% (95% CI 63.9%–86.3%) for SBD. A negative FIT was present in n=379 (68.5%) patients (negative predictive value for CRC = 99.7%). The area under the ROC curve for CRC detection was 0.92.
This is the largest study to assess a POC FIT analyser for symptomatic patients and the first to be conducted within a TWW clinic setting. In addition, data was gathered on the feasibility of faecal sampling via a digital rectal examination, patient opinion towards FIT and how FIT was rolled out for triage in March 2020 in response to the COVID-19 pandemic.