Back

Transcriptomic Changes and Biomarkers in Barrett's Metaplasia, Dysplasia and Cancer

Udumanne, T. P.; Liew, Y. J.; Pascovici, D.; Yang, T.; Lee-Ng, K. K. M.; Gracie, G.; Kumarasinghe, P.; McLeod, D.; Brown, I.; Bourke, M. J.; Lord, S. J.; Ross, J.; Lord, R. V.

2026-08-25 cancer biology
10.64898/2026.08.25.746910 bioRxiv
Show abstract

Esophageal adenocarcinoma (EAC) has a poor five-year survival rate and one of the fastest-rising incidences of any cancer. The presence of dysplasia in Barrett's esophagus (BE) is the main risk factor for EAC development and guides clinical management. Unfortunately, the current histopathological diagnosis of dysplasia is unreliable, with poor inter-observer agreement, highlighting the need for novel biomarkers that can improve diagnostic accuracy. Here, we performed transcriptome profiling across the full spectrum of BE-related neoplasia in 85 samples to delineate gene expression alterations in progressively worse disease stages and identify biomarkers that could complement histopathology to improve the detection of dysplasia and EAC in endoscopic biopsy specimens. Differential gene expression and pathway analyses revealed that the most extensive transcriptional changes occurred during the transition from normal squamous (NSq) to non-dysplastic BE (NDBE), consistent with metaplastic transformation. Compared to NDBE, dysplasia was characterized by enhanced cellular growth and proliferation; upregulation of immune processes and oncogenic signaling pathways were present in EAC. Using machine learning approaches, we identified a novel five-gene panel suitable for a potential RNAseq-based diagnostic test (SLC11A1, IL36A, LUCAT1, MIR215, RNU6-954P) and performed an initial validation of this signature in an additional 51 samples. We also identified several potential novel immunohistochemical markers that may warrant further evaluation, including TREM1, CXCL5, OSM, and motilin. In summary, by delineating transcriptional changes across the full disease spectrum, this study identifies several candidate biomarkers for improving current diagnostic methods for Barrett's dysplasia and EAC.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.