Background: The purpose of the present investigation is to determine whether centrosome amplifications are present in breast tumor cells, whether there are differences of centrosome amplification between benign breast lesions and breast carcinomas, and whether centrosomal analysis can be of value in the diagnosis and prognosis of breast carcinoma.
Methods: Using immunofluorescence analysis with an antibody against gamma-tubulin, we analyzed centrosome abnormalities in fine-needle aspirations of 100 breast lesions (25 cases with benign lesions and 75 cases with carcinomas).
Results: We found that centrosome amplifications, including numerical centrosome amplification and structural centrosome amplification, were present in most breast tumors. Cells with numerical centrosome amplification were found in 23 of 25 benign lesions, and in all 75 cases of breast carcinomas. Cells with structural centrosome amplification were found in three of 25 benign lesions, and in 69 of 75 breast carcinomas. The breast carcinomas showed a mean percentage of cells with numerical centrosome amplification of 4.86% and a mean percentage of cells with structural centrosome amplification of 3.98%. These percentages were significantly higher than those in benign lesions, with a numerical centrosome amplification of 2.77% and a structural centrosome amplification of 0.10%. Furthermore, the mean percentage of cells with structural centrosome amplification was significantly associated with HER2/neu overexpression (P < 0.05) and with negative estrogen receptor status (P < 0.05), and had a borderline association with negative progesterone receptor status (P = 0.056) in breast carcinomas.
Conclusion: Structural centrosome amplification may bear a close relationship with breast carcinoma and may be a potential biomarker for diagnosis and prognosis of breast carcinoma.