Purpose: The behavior of invasive carcinomas in human can be very varied with different individual responses to chemo- therapy. Individualization is crucial to the optimization of che- motherapy. Therefore, the prediction of a tumor? sensitivity to anticancer agents has been the subject of intensive inve- stigation. In order to investigate the pathobiology of breast cancer, it is necessary to maintain or recreate the characte- ristics of the three-dimensional architecture of the tissues in culture. In this study, we have evaluated the relationship bet- ween the Histoculture Drug Response Assay (HDRA) assess- ment and chemotherapy responses in breast cancer patients.
Method: Tumor specimens from 30 patients with breast cancer were evaluated using the HDRA. Tumor tissues were cultured on gelfoam sponge gel in 24-well plates, followed by treatment with a variety of chemotherapeutic agents. All treatments were conducted in triplicate. The sensitivity of a chemotherapy regimen was defined as a tumor inhibition rate (IR) in excess of 30%. Neoadjuvant or palliative chemothe- rapy for patients, using anthracycline or taxane, was conduct- ed on the basis of the established protocols. The responses to treatments were compared with the results of the HDRA.
Results: The mean IR for the combinations of doxorubicin and docetaxel and for FAC and AC were 48, 45, and 36%, respectively. The above partial rate of response to chemo- therapy was 81.1%. The sensitivity and specificity of the HD- RA assessment, with a 30% inhibition rate, were 81.5 and 66.7%, respectively. The positive and negative response prediction values were 91.7 and 44.4%, respectively. The responses to treatments and the results of the HDRA asse- ssment were not correlated with the expressions of the hor- monal receptor or c-erbB2.
Conclusions: In cases in which the inhibition rate is in excess of 30%, the HDRA assessment yielded a high positive res- ponse prediction value. The sensitivity to chemotherapy, as determined by the HDRA, appears to be a good guide for selection in breast cancer patients. Thus the results presented herein should be integrated into future research on the sub- ject.