Abstract P5-05-04: Relationship between plasma estradiol levels and estrogen-regulated gene expressions in premenopausal estrogen receptor-positive breast cancers

Abstract

Abstract Background: Estrogen receptor (ER) and progesterone receptor (PgR) expression status in breast cancer are important factors in determining the use of adjuvant therapy. Plasma estradiol (E2) and progesterone vary markedly through the menstrual cycle. Recently, the levels of PgR expression were reported to be correlated with E2 at the time of surgery in Caucasian premenopausal breast cancer patients. Materials and methods: In this study, plasma E2 levels at surgery were examined for correlations with expressions of estrogen-regulated genes (ERGs), including PgR, and proliferation genes in 147 Japanese premenopausal ER-positive breast cancers. Plasma E2 and progesterone, FSH, and LH were measured on the day of surgery by ECLIA and mRNA expressions were analyzed for 4 ERGs (PgR, GREB1, TFF1, PDZK1), 2 progesterone-regulated (RANKL, cyclin D1) and 2 proliferation genes (MKI67, BIRC5) by RT-PCR using the TaqMan system. ER and PgR protein expression levels were estimated by immunohistochemistry. Results: Our results revealed that expressions of PgR mRNA and protein in breast cancer tissues were significantly correlated with plasma E2 levels at surgery (p = 0.005, p = 0.0165, respectively). Other ERGs were also significantly correlated with plasma E2 levels. However, there were no correlations between mRNA expressions of proliferation genes, MKI67 and BIRC5, and plasma E2 levels. Furthermore, mRNA expressions of the progesterone-regulated genes, RANKL and cyclin D1, were significantly correlated with PgR mRNA expression (p<0.0001, p<0.0001, respectively). Discussion: Significant correlations were seen in the expression of ERGs in Japanese ER-positive breast cancers according to the plasma E2 levels, which vary markedly through the menstrual cycle. Our results supported the previous report in Caucasian patients by other group. This variability may affect the interpretation of gene expression profiles incorporating ERGs such as the 21-gene recurrence score. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P5-05-04.