Validating quantitative real-time PCR analyses on formalin-fixed paraffin-embedded tissues, to identify the molecular characteristics of breast cancers that lose oestrogen receptor during the acquisition of tamoxifen resistance.

...

Technological advances have allowed the extraction and analysis of mRNA in FFPE tissues by Q-RT-PCR.

Technological advances have allowed the extraction and analysis of mRNA in FFPE tissues by Q-RT-PCR.The Biochemistry Department has created a collection of c.3,000 frozen breast carcinomas resected at RMH with >12 year median follow-up but there are many questions that we would like to address for which appropriate tissues are insufficiently represented or for which randomised data are needed.

Over the last few years technological advances have allowed the extraction and analysis of mRNA in FFPE tissues by Q-RT-PCR. This is an ideal method due to the highly shortened lengths of RNA following degradation: primers and probe are designed to span and detect a gene-specific region of between 50 and 150bp.

We have initiated the development and validation of our own Q-RT-PCR methodology for application to the very large stores of FFPE breast cancer biopsies held in the RMH archives (c.25,000 cases over the last 25 years) and in similar tissues held by the Dowsett lab from clinical trials such as ATAC and HERA.

We will initially apply the validated technology in studies of the loss of ER expression during tamoxifen resistance.

A highly consistent finding from expression microarray analysis of breast carcinomas is that ERα status is a dominant determinant of the segregation of the tumours on unsupervised hierarchical clustering. The biological corollary is that ER status appears to be the most important molecular determinant of the overall phenotype of breast cancer. Although most tumours that recur during tamoxifen resistance retain significant amounts of ER, a clinically important proportion recurs as ER- tumours.

A key question in understanding the biology of the acquisition of tamoxifen resistance, which also has substantial therapeutic consequences, is: do those tumours that become ER- lose the transcriptional phenotype of ER+ tumours and acquire that of ER- tumours (largely consisting of basal or HER2+ tumours) or do they lose the expression of only those genes most closely aligned with ER signalling?

We aim to answer this question by the assessment by Q-RT-PCR of 3 sets of selected genes to represent:

• the ER signalling pathway (eg TFF1 and CCND1 [cyclin D1], respectively)


• features of the luminal phenotype not directly associated with ER signalling (eg GATA3)


• features of basal tumours (eg. PgR-, HER2-, CK5/6+).