Open Access Open Badges Research article

Low E2F1 transcript levels are a strong determinant of favorable breast cancer outcome

Vincent Vuaroqueaux1*, Patrick Urban1, Martin Labuhn2, Mauro Delorenzi3, Pratyaksha Wirapati45, Christopher C Benz6, Renata Flury7, Holger Dieterich8, Frédérique Spyratos9, Urs Eppenberger1 and Serenella Eppenberger-Castori1

Author Affiliations

1 Stiftung Tumorbank Basel, Lörracherstrasse 50, 4125 Riehen, Switzerland

2 OncoScore AG, Lörracherstrasse 50, 4125 Riehen, Switzerland

3 Swiss Institute for Experimental Cancer Research and Swiss Institute of Bioinformatics, Boveresses 155, Office D208, CH-1006 Epalinges, Switzerland

4 Chemin des Boveresses 155, CH-1066 Epalinges, Switzerland

5 Room 2021, Genopode, UNIL Sorge, CH-1015 Lausanne, Switzerland

6 Therapeutics Buck Institute for Age Research 8001 Redwood Blvd., Novato, CA 94945, USA

7 Fachärztin FMH für Pathologie, Chefärztin Pathologie, Brauerstrasse 15,8401 Winterthur, Switzerland

8 Women's Hospital and Breast Cancer Center, Therese-Herzog-Weg 2, 79618 Rheinfelden, Germany

9 Laboratoire d'Oncogénétique/INSERM U735, Centre René Huguenin, 35 rue Dailly, 92210 St-Cloud, France

For all author emails, please log on.

Breast Cancer Research 2007, 9:R33  doi:10.1186/bcr1681

Published: 29 May 2007



We investigated whether mRNA levels of E2F1, a key transcription factor involved in proliferation, differentiation and apoptosis, could be used as a surrogate marker for the determination of breast cancer outcome.


E2F1 and other proliferation markers were measured by quantitative RT-PCR in 317 primary breast cancer patients from the Stiftung Tumorbank Basel. Correlations to one another as well as to the estrogen receptor and ERBB2 status and clinical outcome were investigated. Results were validated and further compared with expression-based prognostic profiles using The Netherlands Cancer Institute microarray data set reported by Fan and colleagues.


E2F1 mRNA expression levels correlated strongly with the expression of other proliferation markers, and low values were mainly found in estrogen receptor-positive and ERBB2-negative phenotypes. Patients with low E2F1-expressing tumors were associated with favorable outcome (hazard ratio = 4.3 (95% confidence interval = 1.8–9.9), P = 0.001). These results were consistent in univariate and multivariate Cox analyses, and were successfully validated in The Netherlands Cancer Institute data set. Furthermore, E2F1 expression levels correlated well with the 70-gene signature displaying the ability of selecting a common subset of patients at good prognosis. Breast cancer patients' outcome was comparably predictable by E2F1 levels, by the 70-gene signature, by the intrinsic subtype gene classification, by the wound response signature and by the recurrence score.


Assessment of E2F1 at the mRNA level in primary breast cancer is a strong determinant of breast cancer patient outcome. E2F1 expression identified patients at low risk of metastasis irrespective of the estrogen receptor and ERBB2 status, and demonstrated similar prognostic performance to different gene expression-based predictors.