Molecular surveillance of Barrett’s esophagus and putative strategies for genetic therapy

Abstract

Conventional assessment of the nature of esophageal mucosa is at present by histological analysis of endoscopic biopsies. Unfortunately this requires multiple biopsies which may not be representative of oligoclonal lesions such as Barrett’s metaplasia. The use of cytology potentially overcomes these problems but is labour intensive and many subtle morphological features are lost in the processing. In this regard, application of molecular cell sciences can overcome these problems and, in addition, can elucidate genetic lesions not visible. Abnormalities of differentiation characterise earlier stages of Barrett’s tumorigenesis and we studied c-erbB3 a tyrosine kinase receptor known to be expressed differentially in oesophageal lesions. We studied protein expression by Western blot analysis and RNA expression by RT-PCR and Northern blot analysis. Western blot analysis confirmed that the protein product corresponded with the mature c-erbB3 protein of 160 kDa. In addition c-erbB3 expression is strongest in Barrett’s gastric-type metaplasia but in specimens with dysplasia and those with intestinal metaplasia c-erbB3 immunoreactivity is reduced. Furthermore the majority of adenocarcinomas had diminished or absent immunoreactivity, especially the poorly differentiated lesions. RT-PCR confirmed that expression of c-erbB3 mRNA in Barrett’s mucosa derived from biopsies is moderately strong and expression in dysplastic specimens and invasive neoplasia decreases steadily and in the most poorly differentiated lesions is absent. Application of RT-PCR analysis indicates that esophageal cytology from dysplastic mucosa expresses less c-erbB3 compared with non-dysplastic metaplasia. We applied the analysis of RT-PCR to oesophageal cytology. Brushings were taken in 1 cm sweeps from metaplastic epithelium during endoscopy and c-erbB3 was reduced or absent in cytology corresponding with histological features of intestinal metaplasia and dysplasia, respectively. Furthermore we have applied this biomarker of tissue differentiation to cytological preparations from upper alimentary epithelia in one of the first studies of its kind. In order to further analyse the biology of growth regulatory genes during esophageal carcinogenesis we briefly summarise some of the in vivo and in vitro clinical cell models available. Subsequently some of the putative genetic therapies are discussed and as such it may be possible to make initial steps towards gene transfer therapy in esophageal cancer during this decade.