Articles
Risk stratification of Barrett's oesophagus using a non-endoscopic sampling method coupled with a biomarker panel: a cohort study

https://doi.org/10.1016/S2468-1253(16)30118-2Get rights and content

Summary

Background

Barrett's oesophagus predisposes to adenocarcinoma. However, most patients with Barrett's oesophagus will not progress and endoscopic surveillance is invasive, expensive, and fraught by issues of sampling bias and the subjective assessment of dysplasia. We investigated whether a non-endoscopic device, the Cytosponge, could be coupled with clinical and molecular biomarkers to identify a group of patients with low risk of progression suitable for non-endoscopic follow-up.

Methods

In this multicentre cohort study (BEST2), patients with Barrett's oesophagus underwent the Cytosponge test before their surveillance endoscopy. We collected clinical and demographic data and tested Cytosponge samples for a molecular biomarker panel including three protein biomarkers (P53, c-Myc, and Aurora kinase A), two methylation markers (MYOD1 and RUNX3), glandular atypia, and TP53 mutation status. We used a multivariable logistic regression model to compute the conditional probability of dysplasia status. We selected a simple model with high classification accuracy and applied it to an independent validation cohort. The BEST2 study is registered with ISRCTN, number 12730505.

Findings

The discovery cohort consisted of 468 patients with Barrett's oesophagus and intestinal metaplasia. Of these, 376 had no dysplasia and 92 had high-grade dysplasia or intramucosal adenocarcinoma. In the discovery cohort, a model with high classification accuracy consisted of glandular atypia, P53 abnormality, and Aurora kinase A positivity, and the interaction of age, waist-to-hip ratio, and length of the Barrett's oesophagus segment. 162 (35%) of 468 of patients fell into the low-risk category and the probability of being a true non-dysplastic patient was 100% (99% CI 96–100) and the probability of having high-grade dysplasia or intramucosal adenocarcinoma was 0% (0–4). 238 (51%) of participants were classified as of moderate risk; the probability of having high-grade dysplasia was 14% (9–21). 58 (12%) of participants were classified as high-risk; the probability of having non-dysplastic endoscopic biopsies was 13% (5–27), whereas the probability of having high-grade dysplasia or intramucosal adenocarcinoma was 87% (73–95). In the validation cohort (65 patients), 51 were non-dysplastic and 14 had high-grade dysplasia. In this cohort, 25 (38%) of 65 patients were classified as being low-risk, and the probability of being non-dysplastic was 96·0% (99% CI 73·80–99·99). The moderate-risk group comprised 27 non-dysplastic and eight high-grade dysplasia cases, whereas the high-risk group (8% of the cohort) had no non-dysplastic cases and five patients with high-grade dysplasia.

Interpretation

A combination of biomarker assays from a single Cytosponge sample can be used to determine a group of patients at low risk of progression, for whom endoscopy could be avoided. This strategy could help to avoid overdiagnosis and overtreatment in patients with Barrett's oesophagus.

Funding

Cancer Research UK.

Introduction

Patients with oesophageal adenocarcinoma have a median survival of 1 year, despite advances in treatment.1 The burden of oesophageal adenocarcinoma could be reduced by diagnosing more cases of the precursor lesion Barrett's oesophagus while identifying individuals at increased risk for cancer development before treating them endoscopically.2 However, this is a formidable task given that the incidence of gastro-oesophageal reflux is approximately five per 1000 person-years (15% incidence) in the UK and Europe.3 Furthermore, despite clinical guidelines for endoscopic referral, primary care practice varies and low endoscopy referral rates correlate with poor outcomes from oesophageal cancer.4 The scale of the problem and the psychological and fiscal costs of endoscopy necessitate a new strategy.

We have developed a non-endoscopic diagnostic test involving a cell-collection device that, coupled with the biomarker trefoil factor 3 (TFF3), diagnoses Barrett's oesophagus.5, 6 The device, called Cytosponge, comprises a medical-grade foam sphere on a string compressed within a gelatine capsule that is swallowed while holding onto the string. After 5 min, the capsule dissolves in the stomach, allowing the foam sphere to expand before being pulled from the stomach through the oesophagus to the mouth. Cells are collected along the entire oesophageal lining, minimising the sampling bias that is inevitable with endoscopic biopsies. The sample is transported to the laboratory in preservative at room temperature and processed to paraffin for TFF3 biomarker assessment. Data from previously published studies, amounting to more than 2000 patients, have shown that this approach is safe, has favourable acceptability compared with endoscopy, has a sensitivity of 79·5–87% (depending on Barrett's oesophagus segment length), and has a specificity of 92·4% for diagnosing Barrett's oesophagus.5, 6

Research in context

Evidence before this study

We searched PubMed from database inception to Sept 1, 2011, with the MeSH terms: “biomarkers”, “cancer progression”, and “Barrett's oesophagus” before the start of the BEST2 trial, in order to review the status of the scientific literature. This important area has attracted a lot of attention, and aside from the histopathological assessment of dysplasia, the most promising biomarkers include P53 status, copy number alterations, and methylation panels. All of these studies have relied on endoscopic sampling, which is invasive and resource intensive. Furthermore, most effort has focused on identifying patients at high risk for cancer. However, since most Barrett's oesophagus cases will not progress to cancer, strategies to risk-stratify patients and avoid overdiagnosis are also very important. Furthermore, given the heterogeneity in the molecular genetic patient profiles of those progressing to cancer, identification of very low-risk patients might be a more achievable biomarker strategy. A non-endoscopic cell collection device (Cytosponge) is effective at diagnosing Barrett's oesophagus when coupled with the biomarker, TFF3. The aim of this study was to determine whether additional biomarkers could identify patients at very low risk of cancer, in whom further monitoring could be performed via the Cytosponge. Endoscopy and treatment could then focus on the patients at highest risk.

Added value of this study

We tested a clinical and molecular biomarker panel on Cytosponge samples (seven variables) and used multivariable logistic regression to identify a minimum panel that could risk-stratify patients. The optimum panel comprises age, length of Barrett's oesophagus segment, waist-to-hip ratio (or body-mass index), and three biomarkers scored in a binary method to indicate P53 status, Aurora kinase A expression, and glandular atypia, and divides patients into low-risk, moderate-risk and high-risk groups. This risk score was validated in an independent patient cohort and shown to confidently identify approximately 30% of patients with a low risk for malignant progression. The high-risk and moderate-risk patients would be triaged for endoscopy, which could be prioritised accordingly. To the best of our knowledge, this is the first multidimensional panel applied to non-endoscopic oesophageal cell samples (ie, one involving DNA mutation, protein expression, and clinical variables) to determine the risk profile for a patient with Barrett's oesophagus.

Implications of all the available evidence

At present, the monitoring of patients with diagnosed Barrett's oesophagus relies on endoscopy, which is invasive for patients and expensive for the health-care system. The use of a Cytosponge-biomarker test has the potential to more objectively risk-stratify patients and identify a low-risk group for monitoring within the primary care setting that could be spared endoscopy.

For TFF3-positive patients, it is essential that additional biomarkers are tested to assess the presence of genetic or molecular abnormalities indicative of dysplasia, because positive tests in patients with very benign disease at low risk of progression are common in early cancer detection.7 Overdiagnosis and overtreatment have adverse consequences for both patients and health-care providers.8 Our primary goal is thus to identify patients with Barrett's oesophagus with a very-low-risk profile, so that these individuals can be reassured without having to undergo endoscopy.

We have shown that TP53 mutations qualify for a risk stratification biomarker and are detectable using the Cytosponge.9 We have also shown that the Cytosponge sample is representative of the multiple clones within a diverse genetic content.10 However, for a risk-stratification tool to be clinically applicable, it is important not to miss patients with high-grade dysplasia and intramucosal adenocarcinoma. Although TP53 mutation has a high specificity, it will not identify all patients with high-grade dysplasia or early cancer because its prevalence is 70–80%.9, 11 To increase sensitivity for high-risk patients with Barrett's oesophagus, we propose including additional molecular biomarkers combined with clinical factors.

Hence, we aimed to identify a clinically applicable risk-stratification biomarker panel that could be used on the Cytosponge samples; to apply this panel to a large cohort of patients with Barrett's oesophagus and a TFF3-positive Cytosponge sample to confidently identify a low-risk group of at least 30% of cases who could be spared endoscopy; and to test the Cytosponge risk-stratification model on an independent cohort of patients.

Section snippets

Study design and participants

The prospective, multicentre Barrett's oesophagus screening trial 2 (BEST2) case-control cohort examined risk stratification in the Barrett's oesophagus group as a secondary objective.6 Patients in the validation cohort were selected from newly registered BEST2 patients and from the CASE1 study (appendix 1, p 4).21 For this study, all Barrett's oesophagus patients with intestinal metaplasia and a TFF3-positive Cytosponge test were included. The earliest date of enrolment for participants was

Results

We assessed TFF3-positive Cytosponge samples from 468 patients with Barrett's oesophagus and intestinal metaplasia. Of these, 376 had no dysplasia and 92 had high-grade dysplasia or intramucosal adenocarcinoma (figure 1). Patients with high-grade dysplasia or intramucosal adenocarcinoma were older (Mann-Whitney test, p<0·0001), had longer Barrett's oesophagus segments (p<0·0001), and had a higher waist-to-hip ratio than those with no dysplasia (p=0·008; table 1). All seven biomarkers were

Discussion

This study shows it is possible to do a combination of biomarker assays from a single Cytosponge sample. Combining the biomarkers (P53 abnormality, glandular atypia, and AurKA staining) with clinical variables (age, length of Barrett's oesophagus, and obesity) and using logistic regression enables us to risk-stratify patients into three risk groups. Our objective was to determine with high confidence a group of patients at low risk of progression for whom endoscopy could be avoided. The

References (30)

  • P van Luijt et al.

    The distribution of ductal carcinoma in situ (DCIS) grade in 4232 women and its impact on overdiagnosis in breast cancer screening

    Breast Cancer Res

    (2016)
  • R Moynihan et al.

    Preventing overdiagnosis: how to stop harming the healthy

    BMJ

    (2012)
  • J Weaver et al.

    Ordering of mutations in preinvasive disease stages of esophageal carcinogenesis

    Nat Genet

    (2014)
  • C Ross-Innes et al.

    Whole-genome sequencing provides new insights into the clonal architecture of Barrett's esophagus and esophageal adenocarcinoma

    Nat Genet

    (2015)
  • A Dulak et al.

    Exome and whole-genome sequencing of esophageal adenocarcinoma identifies recurrent driver events and mutational complexity

    Nat Genet

    (2013)
  • Cited by (81)

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

      2022, eBioMedicine
      Citation Excerpt :

      We used the BEST2 trial cohort (n = 529) to derive our automated prediction model since this study had a range of segment lengths in patients who were undergoing BO surveillance. We took the raw output from the deep learning-based Cytosponge-TFF3 IM-prediction model14, and extracted the positive tile count from within the WSI.22 Interpretation of TFF3 using this model had been previously performed in a binary fashion – TFF3 positive or negative.

    • Noninvasive tests for eosinophilic esophagitis: Ready for use?

      2022, Annals of Allergy, Asthma and Immunology
      Citation Excerpt :

      Despite these limitations, this novel tool will likely be important for both clinical care and research in this field, and in some hospitals, the EST is already being used as a billable procedure during clinical care. The Cytosponge is a novel, minimally invasive device that was originally designed as a screening tool to evaluate for Barrett's esophagus14-17 and esophageal cancer.18,19 This device consists of a 30-mm spherical polyurethane sponge that is compressed into a capsule and attached to a string (Fig 2).

    • Use of a Cytosponge biomarker panel to prioritise endoscopic Barrett's oesophagus surveillance: a cross-sectional study followed by a real-world prospective pilot

      2022, The Lancet Oncology
      Citation Excerpt :

      As a logical extension of this finding, additional biomarkers could be applied to Cytosponge to identify dysplasia and carcinoma. We have previously shown that a panel encompassing p53 immunohistochemistry and TP53 mutation, aurora kinase A (AUKA; a surrogate for aneuploidy), and clinical factors (age and length of Barrett's oesophagus) was able to predict patients with high-grade dysplasia or intramucosal cancer with good accuracy (area under the receiver operating characteristic curve [AUROC] 0·87; 99% CI 0·73–0·95).16 The clinical applicability of this Cytosponge risk stratification panel is limited because it requires TP53 gene sequencing from formalin-fixed, paraffin-embedded samples, and the AUKA immunohistochemistry is complex to score and the antibody is no longer commercially available.16

    • Utility and Cost-Effectiveness of a Nonendoscopic Approach to Barrett's Esophagus Surveillance After Endoscopic Therapy

      2022, Clinical Gastroenterology and Hepatology
      Citation Excerpt :

      Therefore, the sensitivity of the assay at 74% in this study, while lower than that in treatment-naïve patients, is promising to detect BE in postablation patients. The specificity is also lower than previously reported in BE screening studies using Cytosponge,8,16,17 and may reflect focal cardia IM, or minute islands of CE identified on the sponge but not on biopsies due to sampling error; however, we are unable to demonstrate this with the current study. The cost-utility analysis showed that a Cytosponge-only surveillance strategy was preferred over strategies incorporating endoscopy, dominating the endoscopy only as well as the alternating endoscopy-Cytosponge strategies.

    View all citing articles on Scopus
    *

    Contributed equally to this work

    Members of the BEST2 study group are listed in appendix 1

    View full text