CDCP1 as a biomarker of aggressive phenotype in triple negative breast cancer

Sasso, Marianna (2015). CDCP1 as a biomarker of aggressive phenotype in triple negative breast cancer. PhD thesis The Open University.

DOI: https://doi.org/10.21954/ou.ro.0000f872

Abstract

Triple negative breast cancer (TNBC) is an aggressive breast cancer subtype with a high risk of local and systemic recurrence in the first year post-surgery. TNBCs are defined by the lack of three known receptor markers (ER, PgR, HER2) and are thus unresponsive to endocrine and trastuzumab therapy, underscoring the need to identify a specific targetable molecule for therapy of TNBC besides chemotherapy to improve the clinical management of these highly aggressive tumours.

In this PhD program I addressed the search of biomarker(s) driving TNBC aggressiveness located on the cell membrane of the tumour cells; these class of molecules, easily accessible, can be potential therapeutic target(s) druggable with specific monoclonal antibody(ies). The novelty of adopted approach stand on the use of wound healing/luids (WHF) from breast carcinoma patients as a tool to stimulate tumour cells. WHFs are known to be extremely enriched in growth factors, cytokines and chemokines, and to be a representative model of the microenvironment in which residual and circulating tumour cells or micro-metastases seeded at the time of surgery reside. Reverse phase protein microarray (RPPM) of TNBC cell lines stimulated or not with WHFs identified significantly activated tyrosine-kinase receptors, i.e. HER3, IGF1-R and PDGFR already known to be involved in TNBC disease. The gene expression profile of the same treated or untreated cells identified the transmembrane non-catalytic receptor CUB domain-containing protein 1 (CDCP1) as the only significantly upmodulated gene among the plasma-membrane receptors analyzed. WHFs as well as single growth factors enriched in WHF, i.e. insulin, PDGF, heregulin, FGF and EGF, upmodulated the CDCP1 protein level on the cell surface of TNBC cells. As PDGF, insulin and heregulin are the ligands of receptors found activated in RPPM, all above described results support the existence of a network between WHF stimulation, activation of transmembrane receptors and CDCP1 upmodulation.

Reportedly, CDCP1 enhances metastatic potential through tyrosine phosphorylation-dependent interaction with the Src family, to confer resistance to anoikis in vitro. CDCP1 overexpression is associated with poor prognosis in lung, colon, pancreatic and renal cancers; nevertheless, prognostic data are still missing for breast cancer. By silencing CDCP1 in the most highly expressing TNBC lines MDA-MB-231 and BT549, I found a strong impairment of their migration and invasion ability as well as of the vasculogenic mimicry capacity, an aggressiveness feature proper of TNBC, suggesting a role for CDCP1 in TNBC dissemination. Immunohistochemistry analysis of CDCP1 in 115 human primary TNBC specimens revealed intense membrane staining in ~60% of cases. CDCP1 expression was found to be a risk factor that significantly reduces both c/isease-/ree survival (DFS) and distant c/isease-/ree survival (DDFS) of TNBC patients. Cox analysis revealed a significant interaction between CDCP1 and N-status in predicting DDFS, with significant association of CDCP1 with worse outcome only in N-positive tumours. To look for possible genetic alterations explaining CDCP1 over-expression FISH analysis for both CDCP1 and centromere of chromosome 3, where CDCP1 gene is located, in 75 TNBC cases were performed. The obtained results enabled to classify CDCP1 cases in four different genetic categories: 1. disomic; 2. deleted; 3. polysomic; and 4 amplified for CDCP1 gene. While no differences in term of prognosis were observed between disomic and deleted categories, cases with amplification showed an impressive high risk of relapse with respect to the cases with CDCP1 polysomy, suggesting that also in presence of unbalance of chromosome 3 the gain of CDCP1 has a important role in the disease aggressiveness of TNBC. Notably, FISH results significantly correlated with IHC positivity with 89,5% of FISH positive cases also positive for IHC versus 50% of FISH negative cases (p=0.0026). FISH positive /IHC positive tumours mainly showed strong membrane reactivity, while a weak CDCP1 labeling was mainly detected in FISH negative /IHC positive specimens. This data suggest once again that CDCP1 expression is important for TNBC aggressiveness. Therefore, we produced monoclonal antibodies against this new promising target of therapy for TNBC that are still under characterization.

Understanding how CDCP1 acts in driving TNBC progression and how its expression is regulated in tumour cells is crucial in defining the optimal use of CDCPl-specific antibodies in the clinical management of these highly aggressive tumours.