The Epstein–Barr virus (EBV) has been reported to be associated with a variety of different tumours; for some of these malignancies, including breast cancer, the association remains controversial. This might be due in part to differences in the methodologies used to detect EBV in tissue samples. One approach favoured by many groups is to use immunohistochemistry to detect an EBV protein, EBNA1, which is essential for the maintenance of viral latency in infected cells and therefore should be a good marker for the presence of the virus. However, in this issue of theJournal of Pathology, Hennard and colleagues report that one of the antibodies frequently employed to detect EBNA1 in tissue samples cross-reacts with the MAGE4 protein, a cancer testis antigen expressed in many cancer types. Their observation suggests that reports documenting an EBV association on the basis of reactivity with this antibody must be considered unreliable. It also re-opens the debate about whether breast cancer is an EBV-associated disease. Copyright © 2006 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.