2014 NCRI Cancer Conference

2 - 5 November 2014
The BT Convention Centre Liverpool UK

Prospective, multi-centre, case-control study to evaluate a novel Cytosponge™–TFF3 test for diagnosing Barrett’s oesophagus

Caryn Ross-Innes1, Irene Debiram-Beecham1, Maria O'Donovan2, Elaine Walker1, Sibu Varghese1, Pierre Lao-Sirieix1, Laurence Lovat3, Michael Griffin4, Krish Ragunath5, Rehan Haidry3, Sarmed Sami5, Philip Kaye5, Marco Novelli3, Babett Disep4, Richard Ostler6, Benoit Aigret6, Bernard North6, Peter Sasieni6, Rebecca Fitzgerald1, BEST2 Study Group7,
1MRC Cancer Unit, Hutchison/MRC Research Centre, Cambridge, UK,2Department Histopathology, Addenbrooke's Hospital, Cambridge, UK,3University College London Hospital, London, UK,4Royal Victoria Infirmary, Newcastle upon Tyne, UK,5Nottingham Queen’s Medical Centre, Nottingham, UK,6Cancer Prevention Trials Unit, London, UK,7Multi-centre BEST2 Study group, England, UK,


Barrett's oesophagus (BE) is a common condition which is often undiagnosed and predisposes to oesophageal adenocarcinoma.  A minimally-invasive cell sampling device, the CytospongeTM, coupled with an immunohistochemical marker, trefoil factor 3 (TFF3), has shown promise as a diagnostic tool.


A multicentre, prospective study was performed to evaluate the safety, acceptability and accuracy of the CytospongeTM-TFF3 test in patients with reflux and dyspepsia symptoms without BE (controls) and cases with BE (≥ 1cm circumferential BE or ≥3 cm tongues). The data were compared with endoscopy.


1,110 individuals took part comprising 463 controls (median age 56 years (interquartile range (IQR) 44-66), Male:Female ratio 1.0:1.3) and 647 cases (median age 66 years (IQR 58-73), Male:Female ratio 4.0:1.0).  1,042 (93.9%) patients successfully swallowed the CytospongeTM and no serious adverse events were attributed to the device. Using a visual analogue scale, the CytospongeTM was rated favourably compared with endoscopy (p=0.0003) and patients who were not sedated for endoscopy were more likely to rate the CytospongeTM higher than endoscopy (Mann-Whitney test, p<0.001).   The overall sensitivity of the test was 79.9% (95% confidence interval (CI) 76.4-83.0%) increasing to 87.2% (95%CI 83.0-90.6%) for BE segments with ≥3 cm of circumferential BE. There was no loss of sensitivity in patients with dysplasia. The specificity for diagnosing BE was 92.4% (95%CI 89.5-94.7%).


The CytospongeTM-TFF3 test is safe, acceptable and has very good accuracy for diagnosing BE. This test warrants consideration as an alternative to endoscopy for diagnosing BE with potential applicability to screening in primary care.


We would like to gratefully acknowledge Theo Giannopoulos and Susanne Booth at Castle Hill Hospital for their help in acquiring ovarian tissue samples, and Jane Smales for coordinating the clinical aspects of the study. This study is funded by a grant from NC3Rs (Registry File: G1100600).


References to go here