Dr René Roux

2005 Derek Butler Fellow

Title: A screen for effectors of MYB upregulation in Barrett’s Adenocarcinoma

Project Start Date: April 2005

Completion Date: April 2008


The MYB gene is an important regulator of cell growth and differentiation and has been extensively studied in blood cells where it is implicated in the development of several leukaemias. Recent work has shown that MYB is also inappropriately expressed in a number of solid tumours, including the gastro-intestinal tract.

The project aims:

  • To determine the expression and localization of MYB in the Barrett’s disease sequence (metaplasia-dysplasia-adenocarcinoma).
  • To investigate if MYB regulates the expression of genes thought to be important for the development of Barrett’s.
  • To establish if MYB expression affects the prognosis of patients with oesophageal

I have generated a novel antibody against MYB that is the first of its kind to be suitable for work on human tissue in a variety of applications. After validating the antibody to check its specificity, I applied the antibody to human tissue sections covering the entire spectrum of the Barrett’s disease process. I concluded that MYB is expressed early on in Barrett’s metaplasia and that its expression is associated with well differentiated sections of metaplasia and well differentiated tumours. Its expression is less striking in poorly differentiated tumours and high grade dysplasia sections.

A literature search revealed that CDX2, a gene thought to be important for the development of Barrett’s metaplasia, had a remarkably similar expression pattern to what I had observed with MYB. I confirmed the published literature by staining the same spectrum of tissue sections with CDX2. The expression pattern was indeed similar and by looking at MYB and CDX2 expression in a number of Barrett’s cell lines, I was able to again establish a similar pattern and level of expression.

In order to investigate if this was simply an interesting co-incidence or if a real relationship between the genes existed, I established that there were three sites on the CDX2 promoter that MYB could potentially bind to and thereby control the expression of CDX2. Using molecular techniques I have shown that MYB binds at least two of these sites in vitro and as a result I hypothesized that MYB is one of the genes involved in the regulation of CDX2. Using a system a colleague in my lab developed for reducing MYB expression in an oesophageal adenocarcinoma cell line, I have shown that knocking down MYB results in a corresponding reduction in CDX2 expression. I am currently in the process of determining which MYB binding sites are involved in regulating CDX2 expression.

I obtained 20 early Barrett’s metaplasia biopsy samples, with matched controls from

patients enrolled on the AspECT trial, the largest UK study on Barrett’s to date. This was made possible through collaboration with Prof JA Jankowski, the lead investigator of this trial. Utilizing molecular and staining techniques, I have shown that MYB is overexpressed earlier than CDX2 and thus MYB could prove to be a useful marker for detecting Barrett’s.

I went on to obtain ethics approval to examine a large series of adenocarcinoma tissue sections in whom the matched survival data could be retrieved. By staining the sections for MYB and performing statistical analysis (kindly done by Dr Sue Dutton, CRUK medical statistics in Oxford), I was able to show that MYB staining at the invasive front of tumours had a statistically significant adverse outcome.

My data conclusively shows that MYB plays a role in Barrett’s and related adenocarcinoma and its overexpression at different stages could reflect different functions. MYB is consistently detected earlier than CDX2 and this work has shown that MYB regulates this important gene in Barrett’s. MYB may prove to be a useful biomarker for Barrett’s metaplasia and may ultimately be a marker to predict outcome. This is born out by the data showing that MYB in adenocarcinomas is localized to the invasive edge of tumours, defining a completely novel role for MYB in invasion and/or metastasis in oesophageal adenocarcinoma

Ideally, the patients from the AspECT trial that overexpress MYB early on should be followed up to see if they progress to adenocarcinoma over time and if this was the case, MYB would be a major predictor of clinical outcome, thus allowing for better stratification of high risk patients. The prognostic series has defined an important and new role for MYB in invasion and further research to delineate the mechanisms involved is certainly warranted.