Objective — to estimate the heterogeneity of HER2/neu gene amplification in HER2/neu-positive breast cancer (BC). Material and methods. Fluorescence in situ hybridization (FISH) assay was used to estimate HER2/neu gene amplification and HER2/CEP17 ratios in BC samples with an immunohistochemical evaluation of HER2/neu2+ expression. The results were interpreted according to the 2018 ASCO/CAP guidelines. BC samples with HER2/neu gene amplification (n = 25) was evaluated for variability in HER2/neu amplification and HER2/CEP17 ratios in 20 tumor cells counted using the FISH assay. Results. Significant intratumoral variability was found in the HER2/neu gene copy number and HER2/CEP17 ratios. HER2/neu-negative cells (5-15%) were present in 28% of the examined samples found to be HER2/neu positive. The HER2/neu gene copy number and HER2/CEP17 ratios for these tumors were statistically significantly lower than those in the group in which all the counted cells were characterized by HER2/neu amplification: 6.25 (95% CI 4.3—12.45; p=0.0166) and 2.37 (95% CI 2.06—3.43; p=0.0076), respectively. The threshold value of HER2/CEP17, at which cells without amplification were detected in HER2/neu-positive tumors, was 2.5. Conclusion. HER2/neu gene amplification in BC is extremely variable both within a single tumor and between the tumors of the same biological subtype. Amplification heterogeneity is statistically significantly more common in HER2/neu-positive BC with a HER2/CEP17 ratio <2.5 and may affect the outcome of the disease and also be important in the choice of treatment policy.