Abstract P1-03-04: Genomic Prediction and Genetic Validation of a Role for the E2F Transcription Factors in Breast Cancer

Abstract

Background: A key and dominant feature of human breast cancer is the morpholiogical, molecular and genomic heterogeneity evident in the disease. This presents a problem in both understanding the mechanisms of progression and in developing therapeutic options. Although mouse models of cancer have traditionally been used to simplify the study of human disease, we suggest that there are opportunities to also model the complexity and heterogeneity of human cancer. Recently we have used signatures of cell signaling pathways to investigate the pattern of gene expression in a Myc initiated mouse model of breast cancer. This revealed the presence of a number of distinct subgroups that correlated with the histological variation. An examination of the predicted activation of E2F transcription factors revealed that individual E2Fs were differentially regulated in the various subtypes of Myc driven breast cancers. Materials and Methods: To directly test the computational prediction that E2Fs had divergent roles in breast cancer, we interbred MMTV-Myc transgenic mice with E2F knockout mice. Mammary gland morphology, tumor latency, growth rate and histology were then examined. The resulting tumors were also examined through microarrays (Affymetrix 430A 2.0) for gene expression and activation of gene expression signatures using our previously characterized methods. Results: The ablation of E2F1 resulted in a decrease in Myc initiated tumor latency and an increase in growth rate of the resulting tumors. Conversely, E2F2 and E2F3 mutant backgrounds resulted in a delay of tumor formation and a marked decrease in incidence. The E2F3 mutant background also result in tumors that grew significantly more slowly compared to the wild type controls. Investigating the mechanism behind these effects revealed that E2F1 loss was associated with a decrease in apoptosis in the normally highly apoptotic Myc tumors. E2F2 and E2F3 effects were a result of reducing the Myc mediated effects to the mammary gland, reducing the pool of susceptible cells. In addition to examining the mechanism that resulted in latency alterations we examined the tumors from the various backgrounds to determine whether loss of the individual E2Fs had effects on the genomic heterogeneity through gene expression analysis. This revealed that the E2F2 null tumors had a reduction of the EMT tumor type, a finding we validated through a histological examination. Finally, to determine whether the role of the E2Fs in the mouse model would translate to human breast cancer we examined a panel of human tumors and noted that a low probability of E2F2 activation was associated with extended relapse free survival times. Discussion: Previous studies have illustrated that breast cancer is a heterogeneous disease. Here we have predicted and then genetically tested a variable role for the E2F transcription factors in a heterogeneous mouse model of breast cancer. This genetic test has illustrated the power of the genomic signatures in predicting involvement of signaling pathways in breast cancer. In addition, we demonstrated a key role for the E2Fs in both a mouse model of mammary tumorigenesis and in human breast cancer. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P1-03-04.