Abstract 1610: Expression of genes of the uPA system from LCM-procured breast carcinoma cells and their relationships with clinical outcomes

Abstract

Abstract Introduction: The urokinase plasminogen activator (uPA), its receptor uPAR and serine protease inhibitors PAI-1 or PAI-2 play a key role in tissue membrane remodeling and invasion of basement membranes by induction of a fibrinolytic pathway. Earlier studies reported that uPA and PAI-1 protein levels assist in prediction of breast cancer response to chemotherapy. Our goal is to develop molecular signatures of candidate genes and identify novel relationships with these four protein biomarkers that demonstrate clinical utility for assessment of breast carcinoma outcomes. Methods: The retrospective study used de-identified tissue biopsies from primary breast cancers on which biomarker and clinical outcomes were stored in an IRB-approved Database. ELISA analyses of uPA, uPAR and PAI-1 performed using IMUBIND kits (American Diagnostica Inc.) used cutoff values previously reported. Estrogen (ER) and progestin receptor (PR) assays were performed either by EIA (Abbott Labs) or by radioligand binding (NEN/DuPont). Relative expression levels of 22,000 genes were determined by microarray using RNA extracted and amplified from Laser-Capture Microdissection (LCM) procured breast carcinoma cells. Univariable and multivariable Cox regression analyses, Kaplan-Meier plots, Violin and scatter grams were performed by R Studio version 3.4.1. External validation of gene subsets derived were performed with SurvExpress. Results: uPA and PAI-1 protein content of a carcinoma were predictive of overall survival (OS). Examination of biomarker gene expression by Violin plots revealed that either ER- or PR- breast cancers expressed elevated levels of UPA, UPAR and PAI2 compared to that of ER+ or PR+ carcinoma cells. Scatter grams suggested an inverse relationship between ER/PR protein levels and expression of UPA, UPAR and PAI2. Univariable Cox regression analyses indicated that PAI2 expression was associated with progression-free survival (PFS) and OS while UPA and PAI1 expression was only associated with OS with a p value < 0.3 (selected as the discovery limit). When carcinomas were sorted by biomarker levels, qPCR expression of RERG and NQO-1 were elevated in uPA- lesions while CD34 and EDG-1 were elevated in uPAR- cancers. However, ERBB4 expression was elevated in PAI-1+ carcinomas. Multivariate Cox regression, performed with backward conditional selection using microarray data with ER or PR status revealed clinically relevant genes for PFS and OS. SurvExpress, an online tool, validated gene subsets externally. Conclusions: Use of LCM-procured breast carcinoma cells in a nondestructive manner uncovered relationships between gene expression and either uPA, uPAR, PAI-1 or PAI-2 protein content of a lesion to reveal candidates for predicting clinical outcomes. Certain of these gene subsets appear related to patient prognosis when considered with ER/PR status of the carcinomas. Citation Format: Seth B. Sereff, Michael W. Daniels, James L. Wittliff. Expression of genes of the uPA system from LCM-procured breast carcinoma cells and their relationships with clinical outcomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1610.