Quantification of TFF3 expression from a non-endoscopic device predicts clinically relevant Barrett's oesophagus by machine learning

被引:7
作者
Berman, Adam G. [1 ]
Tan, W. Keith [2 ,3 ]
O'Donovan, Maria [4 ,5 ]
Markowetz, Florian [1 ]
Fitzgerald, Rebecca C. [2 ,3 ]
机构
[1] Univ Cambridge, Canc Res UK Cambridge Inst, Cambridge, England
[2] Univ Cambridge, Early Canc Inst, Dept Oncol, Hutchison Bldg,Hills Rd, Cambridge, England
[3] Cambridge Univ NHS Fdn Trust Cambridge, Addenbrookes Hosp, Dept Gastroenterol, Cambridge, England
[4] Cambridge Univ NHS Fdn Trust, Addenbrookes Hosp, Dept Histopathol, Cambridge, England
[5] Cyted Ltd, Cambridge, England
关键词
Barrett's oesophagus; Trefoil-factor; 3; Intestinal metaplasia; Cytosponge; Non-endoscopic devices; Machine learning; INTESTINAL METAPLASIA; MANAGEMENT; DIAGNOSIS;
D O I
10.1016/j.ebiom.2022.104160
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
Background Intestinal metaplasia (IM) is pre-neoplastic with variable cancer risk. Cytosponge-TFF3 test can detect IM. We aimed to 1) assess whether quantitative TFF3 scores can distinguish clinically relevant Barrett's oesophagus (BO) (C >= 1 or M >= 3) from focal IM pathologies (C <1, M<3 or IM of gastro-oesophageal junction); 2) whether TFF3 counts can be automated to inform dinical practice. Methods Patients from the Barett's oEsophagus Screening Trial 2 (BEST2) case-control and BEST3 randomised trials were used. For aim 1, TFF3-positive glands were scored manually and correlated with clinical diagnosis. For aim 2, machine learning approach was used to obtain TFF3 count and logistic regression with cross-validation was trained on the BEST2 dataset (n = 529) and tested in the BEST3 dataset (n =158). Findings Patients with clinically relevant BO had higher mean TFF3 gland count compared to focal IM pathologies (mean difference 4.14; 95% confidence interval, CI 2.76-5.52, p < 0.001). The mean class-balanced validation accuracy was 0.84 (95% CI 0.77-0.90), and precision of 0.95 (95% CI 0.87-1.00) for detecting clinically relevant BO. Applying this model on BEST3 showed precision of 0.91 (95% CI 0.85-0.97) for focal IM pathologies with a class-balanced accuracy of 0.77 (95% CI 0.69-0.84). Using this model, 55% of patients (87/158) in BEST3 would fall below the threshold for clinically relevant BO and could avoid gastroscopy, while only missing 5.1% of patients (8/158). Interpretation Automated Cytosponge-TFF3 gland quantification may enable thresholds to be set to trigger confirmatory gastroscopy to minimize overdiagnosis of focal IM pathologies with very low cancer-associated risk. Copyright (C) 2022 The Authors. Published by Elsevier B.V.
引用
收藏
页数:10
相关论文
共 28 条
[1]   Survival for oesophageal, stomach and small intestine cancers in Europe 1999-2007: Results from EUROCARE-5 [J].
Anderson, L. A. ;
Tavilla, A. ;
Brenner, H. ;
Luttmann, S. ;
Navarro, C. ;
Gavin, A. T. ;
Holleczek, B. ;
Johnston, B. T. ;
Cook, M. B. ;
Bannon, F. ;
Sant, M. .
EUROPEAN JOURNAL OF CANCER, 2015, 51 (15) :2144-2157
[2]  
Berman AG, PATHML UNIFIED FRAME, V2021
[3]   Risk of Malignant Progression in Barrett's Esophagus Patients: Results from a Large Population-Based Study [J].
Bhat, Shivaram ;
Coleman, Helen G. ;
Yousef, Fouad ;
Johnston, Brian T. ;
McManus, Damian T. ;
Gavin, Anna T. ;
Murray, Liam J. .
JOURNAL OF THE NATIONAL CANCER INSTITUTE, 2011, 103 (13) :1049-1057
[4]   The incidence of oesophageal adenocarcinoma in non-dysplastic Barrett's oesophagus: a meta-analysis [J].
Desai, Tusar K. ;
Krishnan, Kumar ;
Samala, Niharika ;
Singh, Jashanpreet ;
Cluley, John ;
Perla, Subaiah ;
Howden, Colin W. .
GUT, 2012, 61 (07) :970-976
[5]  
Fitzgerald RC, 2020, LANCET, V396, P333, DOI 10.1016/S0140-6736(20)31099-0
[6]   British Society of Gastroenterology guidelines on the diagnosis and management of Barrett's oesophagus [J].
Fitzgerald, Rebecca C. ;
di Pietro, Massimiliano ;
Ragunath, Krish ;
Ang, Yeng ;
Kang, Jin-Yong ;
Watson, Peter ;
Trudgill, Nigel ;
Patel, Praful ;
Kaye, Philip V. ;
Sanders, Scott ;
O'Donovan, Maria ;
Bird-Lieberman, Elizabeth ;
Bhandari, Pradeep ;
Jankowski, Janusz A. ;
Attwood, Stephen ;
Parsons, Simon L. ;
Loft, Duncan ;
Lagergren, Jesper ;
Moayyedi, Paul ;
Lyratzopoulos, Georgios ;
de Caestecker, John .
GUT, 2014, 63 (01) :7-42
[7]   Relevance of the detection of intestinal metaplasia in non-dysplastic columnar-lined oesophagus [J].
Gatenby, Piers A. C. ;
Ramus, James R. ;
Caygill, Christine P. J. ;
Shepherd, Neil A. ;
Watson, Anthony .
SCANDINAVIAN JOURNAL OF GASTROENTEROLOGY, 2008, 43 (05) :524-530
[8]   Triage-driven diagnosis of Barrett's esophagus for early detection of esophageal adenocarcinoma using deep learning [J].
Gehrung, Marcel ;
Crispin-Ortuzar, Mireia ;
Berman, Adam G. ;
O'Donovan, Maria ;
Fitzgerald, Rebecca C. ;
Markowetz, Florian .
NATURE MEDICINE, 2021, 27 (05) :833-+
[9]   Detection of intestinal metaplasia in Barrett's esophagus: An observational comparator study suggests the need for a minimum of eight biopsies [J].
Harrison, Rebecca ;
Perry, Ian ;
Haddadin, William ;
McDonald, Stuart ;
Bryan, Richard ;
Abrams, Keith ;
Sampliner, Richard ;
Talley, Nicholas J. ;
Moayyedi, Paul ;
Jankowski, Janusz A. .
AMERICAN JOURNAL OF GASTROENTEROLOGY, 2007, 102 (06) :1154-1161
[10]   Normalization of intestinal metaplasia in the esophagus and esophagogastric junction: Incidence and clinical data [J].
Horwhat, J. David ;
Baroni, Darren ;
Maydonovitch, Corinne ;
Osgard, Eric ;
Ormseth, Eric ;
Rueda-Pedraza, Eugenia ;
Lee, Hyun J. ;
Hirota, William K. ;
Wong, Roy K. H. .
AMERICAN JOURNAL OF GASTROENTEROLOGY, 2007, 102 (03) :497-506