Background: Evaluation of endocrine responsiveness of estrogen receptor (ER)-positive and human epidermal growth factor receptor 2 (HER2)-negative breast cancers is considered important for determining the effectiveness of endocrine therapy. Although the PI3K and MAPK signaling pathways have been reported as possibly associated with endocrine resistance, details of their involvement in human breast cancers remain unclear. Especially in relation to signal transduction inhibitors, identification of activating pathways can be considered essential. Since ER+/HER2- breast cancers are affected by differences in estrogen milieu between pre- and postmenopausals, a clear understanding of the biological characteristics of such cancers in terms of menopausal status is crucial. To this end, we determined the differences in activating pathways between luminal A and luminal B breast cancers while taking menopausal status into consideration.

Patients and Methods: Surgically removed ER+/HER2- breast cancers from 96 patients (35 premenopausals and 61 postmenopausals) treated at Hyogo College of Medicine or the Tokushima Breast Care Clinic were used for this study. Expression levels of ER (1D5), PR (PgR636), Ki67 (MIB-1), p53 (DO-7), pS6 (downstream molecule of PI3K/Akt/mTOR pathway, D57.2.2E) and pMAPK (D13.14.4E) were determined immunohistochemically in terms of the percentage of positive cancer cells and we categorized >10% as positive. Luminal A was characterized as Ki67<15% and luminal B as Ki67 >15%.

Results: The PR expression levels of luminal B were significantly lower than those of luminal A (P = 0.008) for premenopausals, but this difference was marginal for postmenopausals (P = 0.07). Luminal B cancers showed a significantly higher percentage of p53-positive tumors than did luminal A cancers (31.4% vs. 5.1%; P = 0.002). Similarly, luminal B cancers showed a significantly higher percentage of pS6-positive cancers than did luminal A cancers (56.9% vs. 20.5%; P = 0.0003). On the other hand, pMAPK positivity was not significantly different for luminal A and luminal B cancers as for (30.2% vs. 42.3%; P = 0.22). These tendencies were similar irrespective of menopausal status. The Ki67 expression levels of pS6-positive cancers (23.8+16.0%, mean+standard deviation) were significantly (P = 0.0008) higher than those of pS6-negative cancers (14.9%+13.6%), while there was also a significant difference (P = 0.02) in Ki67 expression levels between p53-positive (23.4+10.9%) and p53-negative cancers (17.9+16.2%). Multivariate analysis showed that pS6 (odds ratio: 5.71, 95% confidence interval: 2.02-16.19; P = 0.001) and p53 (8.46, 1.70-42.11, P = 0.009) expressions were independently associated with luminal B subtype.

Discussion: On the basis of the findings reported here, we conclude that pS6 and p53 expression levels are associated with characteristics of luminal B breast cancers. Since these associations were observed irrespective of menopausal status, the biological difference of luminal subtypes seems to be little affected by estrogen signaling. Our findings also indicate that targeting the PI3K/Akt/mTOR pathway may be a useful strategy for the treatment of ER+/HER2- luminal B breast cancers.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P5-06-02.