INTRODUCTION: The prognostic significance of HER-2/neu in breast cancer is a matter of controversy. We have performed a study in 101 node-negative breast cancer patients with long-term follow-up not treated in the adjuvant setting, and analysed the prognostic significance of immunohistochemistry (IHC) and fluorescence in situ hybridisation (FISH), both separately and in combination, in comparison with traditional prognostic factors. METHODS: Overexpression was classified semiquantitatively according to a score (0 to 3+) (HER-2_SCO). FISH was used to analyse HER2/neu amplification (HER-2_AMP). Patients classified 2+ by IHC were examined with FISH for amplification (HER-2_ALG). Patients with 3+ overexpression as well as amplification of HER-2/neu were positive for the combined variable HER2_COM. These variables were compared with tumour size, histological grade and hormone receptor status. RESULTS: HER-2_SCO was 3+ in 20% of all tumours. HER-2_ALG was positive in 22% and amplification (HER-2_AMP) was found in 17% of all tumours. Eleven percent of the tumours showed simultaneous 3+ overexpression and amplification. Only histological grade (relative risk [RR] 3.22, 95% confidence interval [CI] 1.73-5.99, P = 0.0002) and HER-2_AMP (RR 2.47, 95% CI 1.12-5.48, P = 0.026) were significant for disease-free survival in multivariate analysis. For overall survival, both histological grade (RR 3.89, 95% CI 1.77-8.55, P = 0.0007) and HER-2_AMP (RR 3.08, 95% CI 1.24-7.66, P = 0.016) retained their independent significance. CONCLUSION: The prognostic significance of HER-2/neu in node-negative breast cancer depends on the method of testing: only the amplification of HER-2/neu is an independent prognostic factor for the long-term prognosis of untreated node-negative breast cancer