Email updates

Keep up to date with the latest news and content from BMC Cancer and BioMed Central.

Open Access Highly Accessed Research article

High STAT1 mRNA levels but not its tyrosine phosphorylation are associated with macrophage infiltration and bad prognosis in breast cancer

Piotr Tymoszuk1, Pornpimol Charoentong2, Hubert Hackl2, Rita Spilka3, Elisabeth Müller-Holzner4, Zlatko Trajanoski2, Peter Obrist3, Françoise Revillion5, Jean-Philippe Peyrat5, Heidi Fiegl4 and Wolfgang Doppler1*

Author Affiliations

1 Division of Medical Biochemistry, Biocenter, Innsbruck Medical University, Innrain 80-82, 6020 Innsbruck, Austria

2 Division of Bioinformatics, Biocenter, Innsbruck Medical University, Innsbruck, Austria

3 Laboratory of Pathology, Hospital St. Vinzenz, Zams, Austria

4 Department of Obstetrics and Gynecology, Innsbruck Medical University, Innsbruck, Austria

5 Centre Oscar Lambret, Lille Cedex, France

For all author emails, please log on.

BMC Cancer 2014, 14:257  doi:10.1186/1471-2407-14-257

Published: 12 April 2014

Abstract

Background

STAT1 has been attributed a function as tumor suppressor. However, in breast cancer data from microarray analysis indicated a predictive value of high mRNA expression levels of STAT1 and STAT1 target genes belonging to the interferon-related signature for a poor response to therapy. To clarify this issue we have determined STAT1 expression levels and activation by different methods, and investigated their association with tumor infiltration by immune cells. Additionally, we evaluated the interrelationship of these parameters and their significance for predicting disease outcome.

Methods

Expression of STAT1, its target genes SOCS1, IRF1, CXCL9, CXCL10, CXCL11, IFIT1, IFITM1, MX1 and genes characteristic for immune cell infiltration (CD68, CD163, PD-L1, PD-L2, PD-1, CD45, IFN-γ, FOXP3) was determined by RT-PCR in two independent cohorts comprising 132 breast cancer patients. For a subset of patients, protein levels of total as well as serine and tyrosine-phosphorylated STAT1 were ascertained by immunohistochemistry or immunoblotting and protein levels of CXCL10 by ELISA.

Results

mRNA expression levels of STAT1 and STAT1 target genes, as well as protein levels of total and serine-phosphorylated STAT1 correlated with each other in neoplastic tissue. However, there was no association between tumor levels of STAT1 mRNA and tyrosine-phosphorylated STAT1 and between CXCL10 serum levels and CXCL10 expression in the tumor. Tumors with increased STAT1 mRNA amounts exhibited elevated expression of genes characteristic for tumor-associated macrophages and immunosuppressive T lymphocytes. Survival analysis revealed an association of high STAT1 mRNA levels and bad prognosis in both cohorts. A similar prognostically relevant correlation with unfavorable outcome was evident for CXCL10, MX1, CD68, CD163, IFN-γ, and PD-L2 expression in at least one collective. By contrast, activation of STAT1 as assessed by the level of STAT1-Y701 phosphorylation was linked to positive outcome. In multivariate Cox regression, the predictive power of STAT1 mRNA expression was lost when including expression of CXCL10, MX1 and CD68 as confounders.

Conclusions

Our study confirms distinct prognostic relevance of STAT1 expression levels and STAT1 tyrosine phosphorylation in breast cancer patients and identifies an association of high STAT1 levels with elevated expression of STAT1 target genes and markers for infiltrating immune cells.