Abstract Background: The expression of many genes is known to be regulated by ambient hormone levels. In a preliminary study of random fine-needle aspirate (rFNA) samples from benign breast tissue, we found several genes that were highly correlated with the serum levels of progesterone (P4) and estradiol (E2). We now present data to further validate these genes as markers of menstrual cycle phase (MCP) and menopausal status (MPS) in benign breast tissue which may allow retrospective classification of archived breast samples with respect to MCP and MPS at the time of sampling. Methods: 240 rFNA samples from healthy women with recorded hormonal data at the time of sampling were analyzed. We divided these subjects by menstrual cycle phase (MCP) and menopausal status (MPS): 41 early follicular: (low circulating E2 and P4); 48 mid-cycle (high E2 and low P4); 31 luteal (moderate E2 and high P4). 120 post-menopausal (low E2 and low P4). 100 ng of RNA from rFNA samples of the breast was reverse transcribed. Amplicons of interest were linearly amplified to 14 cycles for 35 genes related to hormone responsiveness. qPCR reactions were carried out using the TaqMan OpenArray (Applied Biosystems). For each gene of interest, expression levels were normalized to the average expression of GAPDH. Gene expression difference between groups were conducted using the Mann-Whitney Test. P-values from gene expression difference were adjusted via the Benjamini-Hochberg (1995) approach. Results: The mean value of TNFSF11 expression level was 13.19 fold higher in luteal phase subjects than in post-menopausal subjects (p = 0.0003) where there was also the biggest difference of serum P4 level between groups. The expression of DIO2 and MYBPC1 was also significantly higher in luteal phase group than in the post-menopausal group (p = 0.005, p = 0.02, respectively). These 3 genes also demonstrated a higher expression pattern in luteal phase than mid-cycle and follicular phase but analysis is still ongoing. All comparisons between these groups will be presented at the meeting. Conclusion: The expression levels of TNFSF11, DIO2 and MYBPC1 vary with MCP and MPS. These hormone-responsive genes are candidate MCP classifiers which could be applied to archived breast samples to assess whether biomarkers of breast cancer risk are stable across the menstrual cycle, since MCP and MPS variation is likely an important source of biological noise in studies of archived breast biopsy material. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-04-08.
Read full abstract