The purpose of this study was to assess the benefit from adjuvant systemic tamoxifen therapy in breast cancer risk groups identified by the previously established prognostic 76-gene signature.

The hypothesis of this study is that the 76-gene signature can predict survival and benefit from adjuvant endocrine therapy in LNN ER-positive patients who received adjuvant systemic therapy with tamoxifen alone.

This dataset consists of a cohort of 136 breast cancer patients treated with adjuvant tamoxifen. Frozen tumor specimens were selected from the tumor banks in three of the four European institutions that provided the samples of patients without systemic therapy, i.e., Institute of Oncology, Ljubljana, Slovenia (36 samples; 1997–1999), National Cancer Institute, Bari, Italy (28 samples; 1990–1998), and Technische Universitaet Muenchen, Germany (9 samples; 1992–1999), and from one US institution, Cleveland Clinic Foundation (63 samples; 1981–2000). The distant metastasis-free survival (DMFS) among patients from different institutions was not different (log rank P = 0.63). All samples were drawn from the consecutive series of patients available in the institutions within the respective periods based on pre-defined inclusion criteria: histological diagnosis of LNN breast cancer, ER-positive primary tumor, adjuvant tamoxifen treatment, informed consent and/or approval by a local Ethical Committee, at least 5 years of follow-up except for patients who developed distant relapse within 5 years, no signs of disease within 1 month after primary surgery, and frozen tissue should be available. Exclusion criteria were: neoadjuvant or adjuvant chemotherapy, history of other primary cancer (except for basal cell carcinoma), less than 50% tumor area on hematoxylin/eosin (HE) section at time of extraction of RNA, or poor RNA quality.

Total RNA was extracted from tumor tissues (median 90 mg; range 40–120 mg) at the Erasmus MC, Rotterdam, as described by Wang et al and at Cleveland Clinic Foundation, Cleveland, and was sent to Veridex LLC, San Diego, for further analysis. The median RNA yield was 82 ?g (range 19–240 ?g). RNA quality was checked by use of the Agilent BioAnalyzer, and samples were profiled only if they had clear distinct 18S and 28S peaks with no minor peaks present, the area under the 18S and 28S peaks was more than 15% of the total RNA area, and if the 28S/18S ratio was between 1.2 and 2.0. Biotinylated targets were prepared using published methods (Affymetrix, CA) and hybridized to Affymetrix U133A GeneChips (Affymetrix, CA). Chips with average intensity of less than 40 or background signal of more than 100 were excluded. Each probe set was considered as a separate gene. Expression values for each gene were calculated using Affymetrix GeneChip analysis software MAS 5.0. For scaling, probe sets were scaled to a target intensity of 600 and scale mask filters were not selected.

The microarray data were nonlinearly normalized, using a cubic B-spline smoother, to ensure the same signal intensity distribution as the original discovery dataset used to establish the 76-gene signature (Wang et al).

Zhang, Yi, Anieta M. Sieuwerts, Michelle McGreevy, Graham Casey, Tanja Cufer, Angelo Paradiso, Nadia Harbeck, et al. “The 76-Gene Signature Defines High-Risk Patients That Benefit from Adjuvant Tamoxifen Therapy.” Breast Cancer Research and Treatment 116, no. 2 (July 2009): 303–9. doi:10.1007/s10549-008-0183-2.

Wang, Yixin, Jan G. M. Klijn, Yi Zhang, Anieta M. Sieuwerts, Maxime P. Look, Fei Yang, Dmitri Talantov, et al. “Gene-Expression Profiles to Predict Distant Metastasis of Lymph-Node-Negative Primary Breast Cancer.” Lancet (London, England) 365, no. 9460 (February 19, 2005): 671–79. doi:10.1016/S0140-6736(05)17947-1.