CXCL10 Tumor Expression in BRCA1 and Non-BRCA1-Associated Breast Cancer
AM Mulligan, I Raitman, D Pinnaduwage, FP O'Malley, IL Andrulis. St. Michael's Hospital, Toronto, ON, Canada; Samuel Lunenfeld Research Institute, Toronto, ON, Canada; Mount Sinai Hospital, Toronto, ON, Canada
Background: CXC chemokines are important for enhancing immunity, regulating angiogenesis and mediating tumor metastases. One such chemokine, CXCL10 (Interferon- inducible protein 10) has been shown to inhibit tumor progression by recruitment of mononuclear cells and antagonizing the effects of angiogenic factors in the tumor microenvironment as well as promoting cancer progression by upregulating factors related to invasion and metastases. BRCA1-associated breast cancers tend to have a strong host lymphocytic response. Here we explore tumor cell expression of CXCL10 in a large cohort of pathologically well characterized BRCA1 and non-BRCA1-associated breast cancers.
Design: On TMAs constructed from tumors from 58 BRCA1 carriers, 64 BRCA2 carriers and 242 control patients from the Ontario Familial Breast Cancer Registry immunohistochemical analysis of CXCL10 and its receptor, CXCR3, was performed and scored using the Allred method. Tumor cell expression of both markers was compared with morphologic parameters and biomarkers including Ki-67, HER2, hormone receptors (HR), and basal cytokeratins (CK). The associations were analyzed by Chi-square or Fisher's exact test.
Results: CXCL10 tumor cell expression was found to be associated with margin circumscription (p=0.018), a peritumoral lymphocytic infiltrate (p=0.03), high grade (p=0.02) and a high Ki-67 index (p=0.0001). An association with lymphovascular invasion or lymph node metastases was not identified. Expression did not correlate with HER2, HR or basal CK expression. Rather than down-regulating expression of its receptor CXCL10 expression positively correlated with CXCR3 expression in tumor cells. Furthermore, in BRCA1-associated tumors, basal-like tumors were more likely to be positive for CXCL10 expression when compared with non-basal BRCA1-associated tumors (78% versus 22%), though this result did not reach statistical significance, likely due to the small sample size.
Conclusions: Our finding that tumor cell expression of CXCL10 is associated with a peritumoral lymphocytic infiltrate in addition to increased expression of its receptor, CXCR3, on tumor cells suggests that it may act in both a paracrine manner, affecting the tumor microenvironment, as well as in an autocrine manner, acting on the tumor cells themselves. The effects of these circuits on breast cancer progression are complex and warrant further elucidation.
Tuesday, March 10, 2009 1:00 PM
Poster Session IV # 44, Tuesday Afternoon