Purpose: Endocrine therapies, such as tamoxifen, are commonly given to most patients with estrogen receptor (ERa)^positive breast carcinoma but are not indicated for persons with ERa-negative cancer. The factors responsible for response to tamoxifen in 5% to 10% of patients with ERa-negative tumors are not clear. The aim of the present study was to elucidate the biology and prognostic role of the second ER, ERh, in patients treated with adjuvant tamoxifen. Experimental Design: We investigated ERh by immunohistochemistry in 353 stage II primary breast tumors from patients treated with 2 years adjuvant tamoxifen, and generated gene expression profiles for a representative subset of 88 tumors. Results: ERh was associated with increased survival (distant disease-free survival, P = 0.01; overall survival, P = 0.22), and in particular within ERa-negative patients (P = 0.003; P = 0.04), but not in the ERa-positive subgroup (P = 0.49; P = 0.88). Lack of ERh conferred early relapse (hazard ratio, 14; 95% confidence interval, 1.8-106; P = 0.01) within the ERanegative subgroup even after adjustment for other markers. ERa was an independent marker only within the ERh-negative tumors (hazard ratio, 0.44; 95% confidence interval, 0.21-0.89; P = 0.02).

Breast ducts contain two types of epithelial cells, inner luminal cells and outer basal/myoepithelial cells. These cells can be distinguished by their immunophenotype. Cytokeratins (CKs) 8 and 18 are expressed in the luminal layer, whereas CK5/14 and the transcription factor p63 characterize the basal epithelial layer. We studied a population-based cohort of 288 sporadic ductal invasive cancers and found 9% positive for CK5/14 and 4% positive for p63. Using a highly sensitive polymer-based immunohistochemical staining, all sporadic tumors were positive for the luminal CK8/18, including those positive for CK5/14. Pairs of primary tumors and metastases (n ¼ 38) were always concordant for CK5/14 expression. The majority of the CK5/14-positive cases were of histologic grade III (P ¼ 0.0007) and steroid hormone receptor negative (Po0.0001). CK5/14 expression was inversely associated with HER-2 oncogene amplification, but only in the subgroup of estrogen receptornegative tumors (P ¼ 0.007). In a separate set of 42 hereditary breast cancers, the majority (78%) of the BRCA1-associated tumors, but only one of 15 BRCA2-associated tumors was positive for CK5/14. In contrast to sporadic CK5/14-positive tumors, BRCA1-associated tumors displayed less intense CK8/18 staining, including some truly CK5/14-positive CK8/18-negative cases. These results suggest that CK5/14-positive sporadic breast cancers arise from glandularly committed progenitor cells rather than true CK8/18-negative basal cells.

Lesions of ERBB2, PTEN, and PIK3CA activate the phosphatidylinositol 3-kinase (PI3K) pathway during cancer development by increasing levels of phosphatidylinositol-3,4,5-triphosphate (PIP 3 ). 3-Phosphoinositide-dependent kinase 1 (PDK1) is the first node of the PI3K signal output and is required for activation of AKT. PIP 3 recruits PDK1 and AKT to the cell membrane through interactions with their pleckstrin homology domains, allowing PDK1 to activate AKT by phosphorylating it at residue threonine-308. We show that total PDK1 protein and mRNA were overexpressed in a majority of human breast cancers and that 21% of tumors had five or more copies of the gene encoding PDK1, PDPK1. We found that increased PDPK1 copy number was associated with upstream pathway lesions (ERBB2 amplification, PTEN loss, or PIK3CA mutation), as well as patient survival. Examination of an independent set of breast cancers and tumor cell lines derived from multiple forms of human cancers also found increased PDK1 protein levels associated with such upstream pathway lesions. In human mammary cells, PDK1 enhanced the ability of upstream lesions to signal to AKT, stimulate cell growth and migration, and rendered cells more resistant to PDK1 and PI3K inhibition. After orthotopic transplantation, PDK1 overexpression was not oncogenic but dramatically enhanced the ability of ERBB2 to form tumors. Our studies argue that PDK1 overexpression and increased PDPK1 copy number are common occurrences in cancer that potentiate the oncogenic effect of upstream lesions on the PI3K pathway. Therefore, we conclude that alteration of PDK1 is a critical component of oncogenic PI3K signaling in breast cancer. [Cancer Res 2009;69(15):6299-306]