Introduction Identifying Barrett’s oesophagus (BE) patients at risk of progressing to oesophageal adenocarcinoma (EA) remains a challenge. Diagnosis of low grade dysplasia is limited by considerable intra and inter-observer variability and inflammation in biopsy samples can lead to a diagnosis of indefinite for dysplasia. Immunostaining using molecular biomarkers would therefore be useful as a diagnostic adjunct in the assessment of dysplasia.
The Aim of this study was to identify and validate a molecular biomarker that can objectively determine dysplasia status and thereby determine cancer risk in BE.
Methods Biomarkers of interest were identified through mining of a microarray gene expression dataset from 59 oesophageal samples with strict consensus diagnosis by expert pathologists [21 BE with no dysplasia (NDBE), 10 BE with low grade dysplasia (LGD), 13 BE with high grade dysplasia (HGD) and 8 EA]. Subsequent validation was performed on a BE tissue microarray (TMA) (60 NDBE, 19 LGD and 29 HGD), and EA TMA (n = 278).
Results Seventy eight genes were differentially expressed between NDBE and HGD. c-MYC was selected as a top target as the expression levels progressively increased with dysplasia stage and cancer with a fold change of 2.46 between NDBE and HGD. mRNA expression levels were significantly higher in HGD and EA compared to NDBE (p < 0.0001). Immunostaining for c-MYC was positive in 25 out of 29 cases of HGD with a sensitivity 86% of and a specificity of 97% (Table 1). 10 out of 19 LGD stained positive for MYC. 254 out of 278 (91%) EA cases showed significant c-MYC protein expression. Among the 19 LGD samples, 9 had HGD in the BE segment and 7 of these were positive for c-MYC. Strong c-MYC staining was also observed in areas regarded as indefinite for dysplasia which were adjacent to areas of HGD.
Table 1. Immunohistochemistry for c-MYC in oesophageal samples. NDBE: Non-dysplastic Barrett’s oesophagus, LGD: Low-grade dysplasia, HGD: High grade dysplasia.
Conclusion Immunohistochemistry for c-MYC is a promising molecular biomarker in Barrett’s oesophagus. It could also enable better risk stratification in patients with samples in the LGD and indefinite for dysplasia categories.
Disclosure of Interest None Declared.