In his editorial ‘Postmenopausal HRT – a role for tamoxifen?‘ in the July/August issue,1 Dr Graham Jackson raises the important issue of using antioestrogens in postmenopausal HRT. The potential benefits of such an approach are impressive: bone preservation, atherosclerosis prevention and, as Dr Jackson discussed, a decrease in breast and even uterine cancer risk. The possibility of cancer reduction is especially attractive compared to conventional HRT, which is associated with an increase in breast cancer risk.2 Antioestrogens could be used as the sole therapy in women with increased risk for breast cancer, or after about 10 years of conventional HRT, to prevent the late onset increase in breast cancer mortality. Nevertheless, I wish to comment on two very important points. Firstly, tamoxifen is not the only available antioestrogen with selective oestrogen receptor modulatory activity for this indication; a related analogue, toremifene (Fareston), has been approved for marketing in Europe, Japan and the US. Toremifene is equally effective in treating breast cancer; it lowers serum LDL cholesterol and Lp(a) levels comparable with tamoxifen and preserves bone density.3-5 In contrast to tamoxifen, toremifene has been reported to increase serum HDL levels.4 Secondly, a huge population of women without any cancer history would be expected to be the target group of this new antioestrogen indication for a considerable period (10 years or more). Accordingly, the safety aspects of the therapy need to be taken into account. Dr Jackson already points out the potential increase in endometrial cancers with tamoxifen.6 In non-clinical studies, it has been shown to be a strong liver carcinogen in the rat.7 It is activated into highly reactive intermediates which alkylate DNA and lead to mutations (permanent DNA changes)8 and carcinogenesis. Some studies in humans show DNA adduct formation in endometrium and blood leucocytes,9, 10 differing from earlier negative results. In addition, a recent preliminary communication states that tamoxifen-induced endometrial cancers exhibit a high frequency of p53 mutations (91%) compared with sporadic endometrial cancers (19%),11 which supports a genotoxic mechanism in endometrial cancer induction by tamoxifen in women. Thus, I suggest it might be unwise to treat healthy women with a genotoxic carcinogen. Toremifene, which does not seem to be genotoxic or mutagenic in vivo to any significant extent,7, 8 would be preferable. The idea of using antioestrogens as HRT is attractive and warrants further evaluation, ultimately large-scale trials. At this point, however, the drug of choice would be the obviously more safe toremifene, which is now available.