The aim of this study was to get a better understanding of the molecular and cellular influences of aging on breast cancer biology and clinical behaviour.

The hypothesis of this study was that gene expression in breast cancer is age-specific and could explain age-related differences in breast cancer incidence and clinical behavior.

Cryobanked breast cancer specimens excised from newly diagnosed Caucasian females were obtained from the National Cancer Institute – Bari (NCI-Bari) (n=71), following review board approval. Tumor specimen selection criteria included sporadic incidence (no first-degree relatives with breast cancer), >50% invasive cancer cellularity, a frozen wet weight of at least 100 mg, ER-positivity (>10% nuclear immunohistochemical stain), and patient age at diagnosis either ? 45 years (younger cases) or ? 70 years (older cases). The 71 NCI-Bari cases were all of ductal histology with mixed nodal status and without any outcome annotation. The two age cohorts in this tumor set showed no significant imbalances in the stage, the grade, and the PR status or ERBB2 status. PR-positivity was defined as ? 10% nuclear immunohistochemical staining, and ERBB2-positivity was defined as gene amplification.

The total RNA (3–5 ?g per sample) was labeled and analyzed using Affymetrix (Santa Clara, CA, USA) HT-HG_U133A Early Access Arrays with 22.9 K probes representing ~13 K unique UniGenes. Analyses were performed by standard Affymetrix procedures within the Lawrence Berkeley National Laboratory and Life Science Divison's Molecular Profiling Laboratory. Probe set measurements were generated from quantified Affymetrix image files (.CEL files) using the RMA algorithm in Bioconductor R. Array data were deposited in the public Gene Expression Omnibus database (GSE8193).

Gene expression values were mean centered, with a low variation filter applied to exclude probe sets that did not have at least 10 observations exhibiting a twofold change from the mean. Filtered probes were annotated (GeneTraffic annotation file, March 2006) and those with unknown UniGene symbols were omitted, yielding a final significant probe set of 6,632 annotated probes representing 5,109 unique genes. Unsupervised hierarchical clustering of the mean centered significant probe set was performed using Cluster (Eisen et al), and was visualized with Java TreeView (Saldanha et al). Phenotypes (for example, age cohort, PR status and ERBB2 status) of the resulting clusters were compared by chi-square test.

Yau, Christina, Vita Fedele, Ritu Roydasgupta, Jane Fridlyand, Alan Hubbard, Joe W. Gray, Karen Chew, et al. “Aging Impacts Transcriptomes but Not Genomes of Hormone-Dependent Breast Cancers.” Breast Cancer Research: BCR 9, no. 5 (2007): R59. doi:10.1186/bcr1765.

Eisen, M. B., P. T. Spellman, P. O. Brown, and D. Botstein. “Cluster Analysis and Display of Genome-Wide Expression Patterns.” Proceedings of the National Academy of Sciences of the United States of America 95, no. 25 (December 8, 1998): 14863–68.

Saldanha, Alok J. “Java Treeview--Extensible Visualization of Microarray Data.” Bioinformatics (Oxford, England) 20, no. 17 (November 22, 2004): 3246–48. doi:10.1093/bioinformatics/bth349.

PROBLEM MATCHING WITH OUR DATASET