Abstract for presentation at 11th International Congress of Human Genetics

KRT8/18 expression differentiates distinct molecular subtypes of grade 3 invasive ductal carcinomas

  • Logan Walker, Christchurch School of Medicine and Health Sciences, New Zealand
  • Dr Gavin Harris, Canterbury Health Laboratories, New Zealand
  • Dr Andrew Holloway, Peter MacCallum Cancer Centre, Australia
  • Dr Grant McKenzie, Christchurch School of Medicine and Health Sciences, New Zealand
  • A/Prof Bridget Robinson, Christchurch School of Medicine and Health Sciences, New Zealand
  • A/Prof Christine Morris, Christchurch School of Medicine and Health Sciences, New Zealand

    • Current classification systems identify breast carcinomas based on patterns of growth and cellular morphology. The majority of tumours, classified invasive ductal carcinoma of no special type (IDC-NST), are regarded as a single diagnostic entity that is subdivided further only by grading. Prognostic information available from traditional histopathology is limited in IDC-NST tumours. However, recent research suggests that IDC-NST tumours showing basal features are more aggressive than those with luminal features, and there is high clinical relevance in differentiating these tumour subtypes. The aim of this study was to ascertain whether critical assessment of tumour histology could predict breast tumours of basal or luminal-cell phenotype, and to identify molecular characteristics that differentiate the observed morphological variations. Along with detailed pathological assessments, immunohistochemistry was used to evaluate the expression of ER, PR, ERBB2, the basal markers TP73L, KRT5/6 and KRT14, and the luminal-specific cytokeratins KRT8/18 and KRT19. Genomic copy-number changes were assessed using comparative genomic hybridisation (CGH), and mRNA expression profiles determined using cDNA microarrays and QRT-PCR. When all pathological variables were considered, tumours fell into two main groups determined by heterogeneous or uniformly positive expression of KRT8/18. Furthermore, hierarchical clustering of mRNA expression data segregated the tumours into two clusters reflecting the KRT8/18unif and KRT8/18hetero staining phenotype. The 38 genes differentially expressed between these two classes included ERBB2, KRT8, and six other genes previously associated with ERBB2+ or luminal phenotypes. CGH revealed recurrent patterns of chromosomal aberrations involving 10q, 16q and 20q that associated with the KRT8/18unif or KRT8/18hetero tumour classes. These findings support indications that morphologically and molecularly distinguishable tumour subclasses exist within IDC-NST.

    Conference Organiser - ICMS Pty Ltd