Chemotherapy for breast cancer is given on the basis of empirical information from clinical trials, an approach which falls to take into account the known heterogeneity of chemosensitivity between patients. This study aimed to demonstrate the degree of heterogeneity of chemosensitivity in breast cancers. In this study, we examined the heterogeneity of chemosensitivity in breast cancer specimens (n = 50) using an ex vivo ATP-tumor chemosensitivity assay (ATP-TCA). Assay evaluability was 92% in surgical biopsies or pleural aspirates. A variety of chemosensitivity agents were tested. We found that the most active single agent tested was paclitaxel, to which 65.9% of samples were sensitive. Combinations of agents also showed more strong sensitivity cases. The Adriamycin+5-FU demonstrated a strong sensitivity in 23 of 43 (52.3%) of samples. Adriamycin+paclitaxel was more effective, with strong sensitivity in 37 of 43 cases tested (86.0%). There was a marked heterogeneity of chemosensitivity in breast cancer. Chemosensitivity testing may provide a practical method of testing new regimens before clinical trials in breast cancer patients.