Some facts must be taken in consideration when interpreting the results of this study. Firstly, the actual number of analysed patients, both in primary and metastatic sites, is very small (13 patients). Secondly, in some of the cases the metastases were local or regional, while in other cases the metastases were distant. It should not be assumed that both types of metastases are biologically equivalent. Notwithstanding, the results are consistent with other (all small) published studies, and appear to indicate that possible discrepancies in HER-2 and topoisomerase II-a gene expression, between primary and metastatic tumours, are not a major cause of treatment failure in metastatic breast cancer. This conclusion, however, needs to be confirmed in larger studies. Accordingly, a series of 107 patients have been analysed, and the results found less than 10% of discordant cases between primary tumour and its distant metastases (Gancberg et al., unpublished data). Data from the present and similar articles also suggest that HER-2 amplification probably occurs very early in the genetic cascade, at least before dissemination occurs.