Epstein-Barr virus as a marker of biological aggressiveness in breast cancer

A viral aetiology is one recently evocated theory behind the physiopathology of breast cancer (BC) (Glaser et al, 2004; de Villiers et al, 2005; zur Hausen, 2009). Even though, the mechanistic aspects of cancer induction by infectious agents sound multiples, that is, immunosuppressive, linked to animal–human transmission, direct or indirect oncogenic, there are epidemiological evidences of pathogens involvement in human cancer (zur Hausen, 2009). Among the putative viruses observed in BC tissue, the presence of the Epstein-Barr virus (EBV), a γ-herpes virus, has been reported in a number of studies (Bonnet et al, 1999; Fina et al, 2001; Glaser et al, 2004). The implication of EBV in carcinogenesis associated with other cancers, such as Burkitt's lymphoma, undifferentiated nasopharyngeal carcinoma, as well as Hodgkin's disease, has been well documented (zur Hausen, 1991). However, the presence and implication of EBV in BC remains controversial. The use of conventional technical approaches (in situ hybridisation, immunochemistry and standard PCR) for its detection may explain the conflicting results. Some groups have failed to detect EBV (Chang et al, 1992; Gaffey et al, 1993; Lespagnard et al, 1995; Chu et al, 1998; Glaser et al, 1998; Dadmanesh et al, 2001; Deshpande et al, 2002; Herrmann and Niedobitek, 2003; Perrigoue et al, 2005), whereas results from others show discrepancy and depended on the methodology used. For instance, although Murray et al (2003)could detect EBV nuclear antigen-1 by immunochemistry using 2B4-1 monoclonal antibody, they failed to detect the EBV genome by quantitative PCR. The reasons behind these apparently conflicting results remain to be clarified; however, technical limitations of the assays, dissimilarities in the archival materials and heterogeneity among cluster cells contaminated by the EBV genome may be same. Moreover, EBV positivity has been linked to the presence of latently infected lymphocytes in the tumours (Horiuchi et al, 1994; Brink et al, 2000) thus, questioning the role of EBV in BC (Chu et al, 2001). However, in accordance with other groups (Labrecque et al, 1995; Luqmani and Shousha, 1995; Bonnet et al, 1999; Chu et al, 2001; Huang et al, 2003; Preciado et al, 2005; Arbach et al, 2006; Perkins et al, 2006; Tsai et al, 2007), we have shown the presence of EBV genetic information in a subset of BC tissue with a specific localisation in the epithelial malignant cells (Fina et al, 2001). Currently, real-time PCR (RT–PCR is increasingly being used for both research and clinical applications. For BC in particular, the detection of HER2 gene amplification has been validated by comparison with conventional methods, such as FISH (Lamy et al, 2006). Analysis using RT–PCR might also help to clearly identify the presence of EBV in BC. However, the use of whole tissue can result in the risk of contamination and this risk has been corrected with the introduction of laser-assisted microdissection (Fina et al, 2001). In studies on formalin-fixed sections, micro- and macro-dissected breast tumours have been tested for the presence of multiple regions of the EBV genome with few actually uncovering the viral sequence (McCall et al, 2001; Thorne et al, 2005). Interestingly, by in situ hybridisation using a (35)S-labelled riboprobe for Epstein-Barr encoded RNA 1 and a laser capture microdissection on frozen samples, combined with quantitative PCR (Q-PCR), we showed that EBV localisation was restricted to certain tumour epithelial cell clusters (Fina et al, 2001). In accordance with our findings, Arbach et al (2006) observed that viral load is variable between tumours and is heterogeneously distributed among morphologically identical tumour cells, some clusters containing high genome numbers compared with others negative for EBV genome within the same specimen. In the present study, we hypothesised that EBV-infected BC cells might behave differently in comparison to those negative for EBV. In order to test this, we sought to (i) measure the frequency of EBV positivity using RT–PCR and (ii) to compare the biological phenotype of EBV-negative and EBV-positive tumours.