Abstract 4598: Temporal changes of gene expression in breast cancer cells during adaptation to estrogen deprivation highlight the clinical potential for targeting kinase pathways

Abstract

Abstract Aromatase inhibitors (AI) have improved the treatment of estrogen receptor positive (ER+) early and advanced breast cancer (BC). However, the efficacy of AIs is limited by disease progression or relapse. No data is available assessing the temporal changes in gene expression during the development of acquired resistance to long-term estrogen deprivation (LTED). To model the adaptive changes associated with resistance to AI therapy, ER+ human BC cells MCF7 were cultured in the absence of estrogen (E). Genome-wide expression analysis using Illumina 48K arrays was performed to assess changes in gene expression during adaptation at 10 time-points up to 40 weeks (wks). Strikingly, the expression of a proliferation metagene showed resistance to E-deprivation occurred as early as 9 wks. Validation by global assessment of gene expression, revealed a stabilisation of the gene signatures after this time-point. Further analyses were restricted to a triangular pairwise comparison: wt MCF7 cells (modelling pre-treatment), 1 wk post E-deprivation (on AI treatment) and at 9 wks (clinical relapse). Comparison of wt MCF7 versus 1 wk showed 2158 genes were down-regulated and 1724 genes were up-regulated (p<0.001). Down-regulated genes were associated with metabolic and cell cycle pathways. The up-regulated genes were linked to interferon and JAK/STAT canonical pathways indicative of cell stress. Comparison of 1 wk versus 9 wks showed 636 down-regulated genes and 607 up-regulated genes (p<0.001). Down-regulated genes were associated with immune responsiveness. Up-regulated genes were involved in metabolism and cell cycle as expected. Comparison of wt MCF7 versus 9 wks negated the overriding effect of the proliferation signature and unmasked the underlying adaptive changes associated with acquired resistance. This analysis showed 1909 genes were down-regulated and 1082 genes were up-regulated (p<0.001). Strikingly, all major up-regulated canonical pathways were involved in cell signalling including PDGF, integrin, PI3K/AKT/p70S6 and IGF1. Validation of these in-vitro findings, using data from 81 patients pre- and post- 2 wks AI therapy prior to surgery, highlighted the clinical significance of down-regulation of the metabolic/cell cycle pathways, which associated with response to therapy. Furthermore preliminary assessment showed PDGFRβ expression increased after 2 wks of AI and low PDGFRβ at pre-treatment was associated with a better response. These temporal changes in gene expression suggest that resistance can manifest as early as 9 wks post E-deprivation. This underlines the clinical significance of diagnostic screenings in short intervals during the course of treatment. Notably alterations in kinase signalling pathways appear pivotal to the plasticity of breast tumors highlighting the potential for tailored targeted treatments. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 4598.