Expression of Estrogen-Related Gene Markers in Breast Cancer Tissue Predicts Aromatase Inhibitor Responsiveness

Irene Moy,Seema A Khan,Serdar E Bulun,Alfred W Rademaker,Scott Reierstad,Zhihong Lin,Shannon M Hawkins

doi:10.1371/journal.pone.0077543

Abstract

Aromatase inhibitors (AIs) are the most effective class of drugs in the endocrine treatment of breast cancer, with an approximate 50% treatment response rate. Our objective was to determine whether intratumoral expression levels of estrogen-related genes are predictive of AI responsiveness in postmenopausal women with breast cancer. Primary breast carcinomas were obtained from 112 women who received AI therapy after failing adjuvant tamoxifen therapy and developing recurrent breast cancer. Tumor ERα and PR protein expression were analyzed by immunohistochemistry (IHC). Messenger RNA (mRNA) levels of 5 estrogen-related genes–AKR1C3, aromatase, ERα, and 2 estradiol/ERα target genes, BRCA1 and PR–were measured by real-time PCR. Tumor protein and mRNA levels were compared with breast cancer progression rates to determine predictive accuracy. Responsiveness to AI therapy–defined as the combined complete response, partial response, and stable disease rates for at least 6 months–was 51%; rates were 56% in ERα-IHC-positive and 14% in ERα-IHC-negative tumors. Levels of ERα, PR, or BRCA1 mRNA were independently predictive for responsiveness to AI. In cross-validated analyses, a combined measurement of tumor ERα and PR mRNA levels yielded a more superior specificity (36%) and identical sensitivity (96%) to the current clinical practice (ERα/PR-IHC). In patients with ERα/PR-IHC-negative tumors, analysis of mRNA expression revealed either non-significant trends or statistically significant positive predictive values for AI responsiveness. In conclusion, expression levels of estrogen-related mRNAs are predictive for AI responsiveness in postmenopausal women with breast cancer, and mRNA expression analysis may improve patient selection.

Highlights

Breast cancer is a hormone-dependent disease that relies on the mitogenic effects of estrogen to drive tumorigenesis and tumor growth [1,2]
Patients with estrogen receptor-a (ERa) or progesterone receptor (PR) proteinpositive tumors are treated with an Aromatase inhibitors (AIs)
In addition to ERa and PR Messenger RNA (mRNA), we explored the prognostic value of aromatase, AKR1C3, and BRCA1 mRNA expression levels for predicting AI responsiveness

Summary

Introduction

Breast cancer is a hormone-dependent disease that relies on the mitogenic effects of estrogen to drive tumorigenesis and tumor growth [1,2]. Endocrine therapy is indicated in patients who possess ERa and PR positive tumors. The estrogen pathway and synthesis has been targeted through receptor blockade, reduction in circulating levels of estrogen, or by suppression of synthesis in tissues of women diagnosed with breast cancer [1,2,3,4,5]. Aromatase inhibitors (AIs), which selectively inhibit aromatase activity in tissues responsible for estrogen production, have been used for the hormonal treatment of breast cancers. Given the fact that the overwhelming majority of breast cancer is hormone receptor-positive, most women are placed on AI treatment. With objective response rates of slightly more than 50% to AI therapy, there is a need for improved predictive methods to better identify patients who will benefit from it and sparing those patients who may not [7]

Objectives

Methods

Results

Conclusion