3971 - Unravelling the biological characteristics of MammaPrint extreme risk subgroups

Background

MammaPrint® (MP) is a 70-gene based prognostic assay that stratifies early-stage breast cancer patients into low and high-risk of relapse. Recently, further stratification of the 70-gene risk results identified extreme low and high-risk subgroups with specific clinical outcomes (Delahaye et al. 2017) and treatment response characteristics (Wolf et al. 2017). However, the biological profiles of these extreme MP subgroups are not fully investigated. In this study, we aim to gain more insight into their biological significance using differentially expression genes (DEGs) analysis.

Methods

We selected 400 samples from the whole MP range and defined 4 subgroups (Ultra high [UH], High risk [HR], Low risk [LR], Ultra low [UL]), for which FFPE microarray full-transcriptome data were available at Agendia. DEGs analysis was performed with limma and subsequent pathway analysis with Enrichr and GOrilla.

Results

Two separate comparative analyses were carried out to unravel biological processes associated with extreme risk subgroups: UL vs. LR and UH vs. HR. We found 101 DEGs (logFC > =0.485 & FDR <0.05) between UL and LR subgroups and 1714 DEGs between UH and HR subgroups. Based on the pathway analysis, our results showed that the UL subgroup was more homogeneous, with enrichment in pathways reflecting low proliferative and metastatic features. This is in line with the favorable long-term outcome characteristic of the UL group. Conversely UH exhibited higher heterogeneity, with the enrichment of more diverse pathways including immune response, cell cycle and proliferation, that could be associated with genomic instability. This would support the recent finding of UH samples being more sensitive to veliparib/carboplatin combination therapy compared to HR samples (Wolf et al. 2017)). Furthermore, clustering approach demonstrated UH and other subgroups as two distinct clusters.

Conclusions

Our preliminary findings give additional insights into the biological processes associated with extreme MP groups, which might open new avenues for therapeutic intervention in breast cancer.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

Agendia Inc.

Funding

Agendia Inc.

Disclosure

R. Bhaskaran: Full / Part-time employment: Agendia Inc. C. Griffioen: Full / Part-time employment: Agendia Inc. D. Wehkamp: Full / Part-time employment: Agendia Inc. L. Mittempergher: Full / Part-time employment: Agendia Inc. A.M. Glas: Full / Part-time employment: Agendia Inc.