Neoplasma Vol.59, No.4, p.424-432, 2012

The aim of our study was to examine an association between gene expression assessed using a 23-gene microarray and receptor status of breast cancer samples categorized as ER positive, HER2 positive and triple negative subtypes. The ER positive cohort was subsequently divided into Luminal A, Luminal B HER2 negative and Luminal B HER2 positive subtypes. Core- needle biopsies were collected from 78 female patients with inoperable locally advanced breast cancer or resectable tumors suitable for downstaging, before any treatment. Expressions of 23 genes were determined by means of TagMan Low Density Arrays. Analysis of variance was used to select genes with discriminatory potential between receptor subtypes. We introduced a correction for false discovery rates (presented as q values) due to testing multiple hypothesis. Pairwise post-hoc comparisons of receptor subtypes were performed using Tukey ‘s HSD test.

Five genes out of a 23-gene microarray differed significantly in relation to breast cancer receptor-based subtypes. Among these five genes, we identified: BCL2 (p=0.0002, q=0.0009), MKI67 (p=0.0037, q=0.0064), IGF1R (p=0.0040, q=0.0064), FOXC1 (p=0.0113, q=0.0135) and IRF1 (p=0.0435, q=0.0416) as ones showing ER positive, HER2 positive and triple negative -subtype specific expression profiles.

When incorporating Luminal A, Luminal B HER2 negative, Luminal B HER2 positive subtypes into analysis, four genes: BCL2 (p=0.0006, q=0.0034), MKI67 (p=0.0078, q=0.0198), FOXC1 (p=0.0102, q=0.0198) and IGF1R (p=0.0174, q=0.0254) were selected. Elevated levels of IGF1R and BCL2 were significantly linked with Luminal A subtype.

Triple negative breast cancer subtype was associated with higher expression of IRF1, FOXC1 and MKI67. In HER2 positive cohort lower expression of all five analyzed genes was noted.