Poster Presentation 37th Lorne Cancer Conference 2025

Improving the diagnosis of Barrett’s oesophagus with dysplasia and oesophageal adenocarcinoma using transcriptome profiling (#263)

Thejaani Udumanne 1 , Yi Jin Liew 2 , Dana Pascovici 2 , Michelle Lee-Ng 1 , Sarah J Lord 1 3 , Jason Ross 2 , Reginald V Lord 1 4
  1. Gastroesophageal Cancer Research Program, St Vincent's Centre for Applied Medical Research, Sydney, NSW, Australia
  2. Health & Biosecurity, CSIRO, Westmead, NSW, Australia
  3. NHMRC Clinical Trials Centre, University of Sydney, Sydney, NSW, Australia
  4. Department of Surgery, University of Notre Dame School of Medicine, Sydney, NSW, Australia

Background: In Barrett’s oesophagus (BO), the normal squamous epithelium of oesophagus is replaced by an intestinal-like columnar epithelium containing goblet cells. While most patients have stable non-dysplastic BO (NDBO), a minority progress through the stages of low-grade dysplasia (LGD) and high-grade dysplasia (HGD) to oesophageal adenocarcinoma (OAC). The presence of dysplasia, identified by histopathology, is the main risk factor for development of OAC in BO patients and guides management (endoscopic surveillance or resection/ablation). Unfortunately, the current histopathological diagnosis of dysplasia is unreliable resulting in poor inter-observer agreement across trained pathologists1.

Aim: To identify a panel of differentially expressed transcripts in LGD, HGD, and OAC as potential protein antigen targets for developing a diagnostic immunohistochemistry (IHC) test to improve diagnostic accuracy for BO patients undergoing endoscopy.

Materials and methods: RNA-seq was performed on 86 fresh frozen oesophageal biopsy tissue samples (Normal=10, NDBO=30, LGD=21, HGD=11, OAC=14) collected from 71 patients in the PROBE-NET study. Data analysis was conducted using the Genomic Data Commons mRNA Analysis Pipeline, followed by transcript quantification using Salmon and differential expression analysis via DESeq2, which incorporated tissue purity scores as covariates (calculated using ESTIMATE). The findings were validated using our previous data consisting of 51 samples (Normal=17, NDBO=14, LGD=8, OAC=12) from 44 patients (Maag et al. 2017)2.

Results: By considering both the abundance and ratio of expression, this study identified a total of 277 protein coding genes that are significantly differentially expressed in dysplasia/OAC vs. NDBO (median TPM>5, |log2FC|>2). This group consists of 96 upregulated genes, mainly related to immune and inflammatory response, and 181 downregulated genes in cytoskeleton organisation and epithelial differentiation pathways. A panel of 8 upregulated and 1 downregulated proteins were identified as potential IHC targets for further testing using tissue microarrays (TMAs). Moreover, the analysis detected 44 membrane/secreted proteins, and 25 plasma proteins as significantly upregulated in dysplasia and OAC, which could serve as potential candidates for future evaluation in molecular imaging endoscopy studies and liquid biopsy proteomics tests.

Conclusions: Transcriptomic profiling and differential expression analysis of BO, dysplasia and OAC tissue identified several potential biomarkers for improving current diagnostic methods for Barrett’s dysplasia.

  1. Montgomery E, Bronner MP, Goldblum JR, et al. Reproducibility of the diagnosis of dysplasia in Barrett esophagus: a reaffirmation. Human Pathology. 2001;32(4):368-378.
  2. Maag JLV, Fisher OM, Levert-Mignon A, et al. Novel Aberrations Uncovered in Barrett's Esophagus and Esophageal Adenocarcinoma Using Whole Transcriptome Sequencing. Mol Cancer Res. 2017;15(11):1558-1569.