Abstract A103: Risk of breast cancer with plasma prolactin concentrations by phosphorylated STAT3 and STAT5 status

A103

Prolactin, a lactogenic hormone, has been associated with risk of breast cancer in premenopausal and postmenopausal women. Prolactin may act by influencing transcription factors in the Signal Transducer and Activator of Transcription (STAT) pathway. In particular, STAT3 and STAT5 appear to be important in breast carcinogenesis; prolactin is known to activate STAT5 via phosphorylation. To determine whether prolactin may act through STAT signaling, we examined whether prediagnostic prolactin levels were differentially associated with risk of breast tumors based on their expression of phosphorylated STAT3 (pSTAT3) and STAT5 (pSTAT5). Tissue microarrays (TMAs) were constructed from 3,093 breast cancers from women enrolled in the Nurses Health Study. Of these, 443 cases had matching plasma prolactin levels drawn at diagnosis; one or two controls were matched to each case by age, menopausal status/hormone use, and characteristics of the blood draw. Immunohistochemical studies with pSTAT3 and pSTAT5 were performed on the TMAs. Tissue cores were scored from 0-3 based on intensity and quantity of nuclei staining. Relationships between expression of pSTAT3 or pSTAT5 and clinicopathologic features were examined by Chi-squared tests. Polytomous logistic regression was used to evaluate the relationship between prolactin and risk of breast tumors by staining expression, adjusting for matching factors and breast cancer risk factors. Interpretable tissue cores were available for 309 subjects. Of these, 195 (63%) and 114 tumors (37%) showed negative/weak (0-1) and moderate/strong staining (2-3) respectively for pSTAT3. 178 (58%) and 131 (42%) tumors showed negative/weak and moderate/strong staining respectively for pSTAT5. Absence of expression of pSTAT3 was significantly associated with high grade (p=0.04) and invasiveness (p=0.04). Though not statistically significant, pSTAT3- status also was associated with lymph node involvement and larger tumor size. pSTAT5 was not associated with other tumor characteristics. Prolactin levels were similarly associated with pSTAT3+ (RR, top vs. bottom quartile=1.7, 95% CI=1.1-2.8) and pSTAT3- tumors (comparable RR=1.8, 95%CI=1.0-3.3). However, high versus low prolactin levels were more strongly associated with pSTAT5+ tumors (RR=2.3, 95%CI=1.3-4.2) compared to pSTAT5- tumors (RR=1.5, 95%CI=0.9-2.4), although the p-heterogeneity was not statistically significant (p=0.28). Our findings suggest that prolactin may influence breast cancer risk via activation of STAT5. On-going analyses are exploring the combined role of pSTAT3 and pSTAT5 presence as well as other hormones thought to modulate or reflect STAT activity including IGFs, CRP, and adiponectin.

Citation Information: Cancer Prev Res 2008;1(7 Suppl):A103.