Search

Summary Background Screening for Barrett's oesophagus relies on endoscopy, which is invasive and few who undergo the procedure are found to have the condition. We aimed to use machine learning techniques to develop and externally validate a simple risk prediction panel to screen individuals for Barrett's oesophagus. Methods In this prospective study, machine learning risk prediction in Barrett's oesophagus (MARK-BE), we used data from two case-control studies, BEST2 and BOOST, to compile training and validation datasets. From the BEST2 study, we analysed questionnaires from 1299 patients, of whom 880 (67·7%) had Barrett's oesophagus, including 40 with invasive oesophageal adenocarcinoma, and 419 (32·3%) were controls. We randomly split (6:4) the cohort using a computer algorithm into a training dataset of 776 patients and a testing dataset of 523 patients. We compiled an external validation cohort from the BOOST study, which included 398 patients, comprising 198 patients with Barrett's oesophagus (23 with oesophageal adenocarcinoma) and 200 controls. We identified independently important diagnostic features of Barrett's oesophagus using the machine learning techniques information gain and correlation-based feature selection. We assessed multiple classification tools to create a multivariable risk prediction model. Internal validation of the model using the BEST2 testing dataset was followed by external validation using the BOOST external validation dataset. From these data we created a prediction panel to identify at-risk individuals. Findings The BEST2 study included 40 diagnostic features. Of these, 19 added information gain but after correlation-based feature selection only eight showed independent diagnostic value including age, sex, cigarette smoking, waist circumference, frequency of stomach pain, duration of heartburn and acidic taste, and taking antireflux medication, of which all were associated with increased risk of Barrett's oesophagus, except frequency of stomach pain, with was inversely associated in a case-control population. Logistic regression offered the highest prediction quality with an area under the receiver-operator curve (AUC) of 0·87 (95% CI 0·84–0·90; sensitivity set at 90%; specificity of 68%). In the testing dataset, AUC was 0·86 (0·83–0·89; sensitivity set at 90%; specificity of 65%). In the external validation dataset, the AUC was 0·81 (0·74–0·84; sensitivity set at 90%; specificity of 58%). Interpretation Our diagnostic model offers valid predictions of diagnosis of Barrett's oesophagus in patients with symptomatic gastro-oesophageal reflux disease, assisting in identifying who should go forward to invasive confirmatory testing. Our predictive panel suggests that overweight men who have been taking antireflux medication for a long time might merit particular consideration for further testing. Our risk prediction panel is quick and simple to administer but will need further calibration and validation in a prospective study in primary care. Funding Charles Wolfson Charitable Trust and Guts UK.

Background and Aims The standard whole-colon tests used to investigate patients with symptoms of colorectal cancer are barium enema and colonoscopy. Colonoscopy is the reference test but is technically difficult, resource intensive, and associated with adverse events, especially in the elderly. Barium enema is safer but has reduced sensitivity for cancer. CT colonography ("virtual colonoscopy") is a newer alternative that may combine high sensitivity for cancer with safety and patient acceptability. The SIGGAR trial aims to determine the diagnostic efficacy, acceptability, and economic costs associated with this new technology. Methods The SIGGAR trial is a multi-centre randomised comparison of CT colonography versus standard investigation (barium enema or colonoscopy), the latter determined by individual clinician preference. Diagnostic efficacy for colorectal cancer and colonic polyps measuring 1 cm or larger will be determined, as will the physical and psychological morbidity associated with each diagnostic test, the latter via questionnaires developed from qualitative interviews. The economic costs of making or excluding a diagnosis will be determined for each diagnostic test and information from the trial and other data from the literature will be used to populate models framed to summarise the health effects and costs of alternative approaches to detection of significant colonic neoplasia in patients of different ages, prior risks and preferences. This analysis will focus particularly on the frequency, clinical relevance, costs, and psychological and physical morbidity associated with detection of extracolonic lesions by CT colonography. Results Recruitment commenced in March 2004 and at the time of writing (July 2007) 5025 patients have been randomised. A lower than expected prevalence of end-points in the barium enema sub-trial has caused an increase in sample size. In addition to the study protocol, we describe our approach to recruitment, notably the benefits of extensive piloting, the use of a sham-randomisation procedure, which was employed to determine whether centres interested in participating were likely to be effective in practice, and the provision of funding for dedicated sessions for a research nurse at each centre to devote specifically to this trial. Trial registration Current Controlled Trials ISRCTN95152621