49P - Molecular classification of basal-like breast cancer subtypes based on predictive survival markers

Date 07 May 2015
Event IMPAKT 2015
Session Welcome reception and Poster Walk
Topics Breast Cancer
Translational Research
Presenter Heloisa Milioli
Citation Annals of Oncology (2015) 26 (suppl_3): 15-24. 10.1093/annonc/mdv117
Authors H.H. Milioli1, I. Tishchenko1, C. Riveros2, R. Berretta2, P. Moscato1
  • 1Centre For Bioinformatics, Biomarker Discovery And Information-based Medicine, The University of Newcastle, 2305 - Newcastle/AU
  • 2School Of Electrical Engineering And Computer Science, The University of Newcastle, Newcastle/AU



Background: Basal-like breast cancers (BLBCs) are a group of tumours that lack the hormone receptors oestrogen (ER) and progesterone (PR), and the human epidermal growth factor receptor-2 (HER2), as defined by immunohistochemistry. Although the terms triple-negative and basal-like have been used interchangeably, the basal-like set is reportedly a more homogenous disease compared to triple-negatives. In this study, we investigate the gene expression data of 240 BLBCs focusing on survival markers to identify molecular subtypes.

Methods: BLBC samples were pooled from the METABRIC transcriptomic dataset. The subtype identification approach is based on hierarchical clustering using a weighted Pearson correlation distance. For building the correlation matrix, probes were first selected based on their differential gene expression values in BLBCs versus control samples, and weighted according to their individual survival prediction power.

Results: Our method based on 183 weighted features revealed three subtypes of basal-like tumours with distinct Kaplan-Meier survival curves (p value of 6.3 x 10-5). Patients of group B1 show a better prognosis with survival rate of 8.6 years; group B2 had a median of 6.2 years; and group B3, with the worse prognosis, only 3.3 years. Overall, the median age was 52.1 years old for all samples and different between the three classes identified: B1 (52.3), B2 (50.6) and B3 (48.7). Patients with age < 50 years showed significantly more unfavourable prognosis compared to patients with > 50 years.

Conclusions: BLBCs are of particular interest due to their high frequency in younger women, aggressive behaviour, poor prognosis and limited therapy response. These arguments justify the effort in elucidating the molecular subgroups and their intrinsic gene profile to better describe the distinct classes. Additionally, the predictive markers enable the alignment of patients' survival and tailored therapy in order to improve clinical decision-making.

Disclosure: All authors have declared no conflicts of interest.