Based on the resulting score, tumor samples are assigned to three different categories, either well-differentiated (grade 1/G1), moderately differentiated (grade 2/G2) or poorly differentiated (grade 3/G3) [14]. For patients whose tumors were characterized as G1 or G3, prognostic information is univocal, with a good prognosis

for G1 and a poor prognosis for G3 patients. However, a considerable percentage of patients are classified as G2 and in this instance a histologic grading provides no helpful information for treatment decisions. In recent years, reverse phase BMS-354825 mw protein arrays (RPPA) have emerged as a powerful high-throughput approach for targeted proteomics [15]. As a major advantage, RPPA allows to assess target protein expression quantitatively in large sample sets while requiring only a very low MLN0128 concentration amount of biological sample [16] making this platform attractive for the analysis of clinical materials and biomarker discovery [[17], [18] and [19]]. The objective of our study was to identify a robust protein signature using RPPA as a technical platform for targeted proteomics to assess the risk of cancer recurrence for breast cancer patients whose tumors had been diagnosed

with histologic G2. Quantitative protein expression data were generated for 128 breast cancer relevant target proteins analyzing a set of 109 hormone receptor-positive tumors. A novel bioinformatics workflow combining three different classification

algorithms was used to analyze RPPA data of histologic G1 and G3 tumors serving as surrogates of low and high risk breast cancer, respectively. The RPPA-derived signature was first subjected to an independent evaluation employing Western blot and mRNA profiling essentially confirming findings made by RPPA. Finally, the biomarker marker profile was translated into a risk classification score named R2LC suitable to predict the recurrence risk in single samples and validated using an independent test set comprising hormone-receptor positive tumors. Tumor specimens (discovery set, n = 109) from patients diagnosed with primary invasive 4��8C breast carcinoma were collected at the time of surgery between 2008 and 2010 at the Department of Gynecology and Obstetrics/National Center for Tumor Diseases, Heidelberg. None of the patients had received neoadjuvant therapy. Institutional Review Board approval was received as ethics vote no. S039/2008 and informed consent was obtained from all the patients. Tumor specimens were processed within 20 min after surgery. Samples were stored snap frozen at −80 °C until further use. Tumor specimens of the test set (n = 145) were obtained from the Tissue Bank of the National Center for Tumor Diseases (Heidelberg).