Résumé : BACKGROUND: The selection of therapies for breast cancer is today based on prognostic features (chemotherapy, radiotherapy), hormone receptor status (hormonal therapy) and HER-2 status (trastuzumab therapy). HER-2, p53 and BCL-2 are tumour-related proteins that have the potential to further improve individualisation of patient management, by predicting response to chemotherapy, hormonal therapy and radiotherapy. MATERIALS AND METHODS: This paper reviews the rationale for the use of these proteins as predictive factors, as well as the published literature addressing the use of each one to predict response to hormonal therapy, chemotherapy and radiotherapy. RESULTS: HER-2, p53 and BCL-2 remain inadequately assessed as predictive factors in breast cancer. HER-2 evaluation is required for the selection of patients for trastuzumab (Herceptin) therapy, as trials of this therapy have been limited to HER-2 overexpressors. HER-2 overexpression may be predictive of resistance to hormonal therapy. Anthracyclines are effective therapy for breast cancer regardless of HER-2 status, but patients whose tumours overexpress HER-2 appear to receive the greatest relative benefit from this therapy. Studies of HER-2 as a predictor of response to CMF and to radiotherapy are inconclusive at this time. No data yet exist to support the use of p53 or BCL-2 as predictive factors in the therapy of breast cancer. CONCLUSIONS: At this point in time, there is inadequate evidence to support the use of HER-2, p53 or BCL-2 to guide the selection of hormonal therapy, chemotherapy or radiotherapy for breast cancer.