Clinical applications of the histoculture drug response assay.

There is a need for a clinically useful drug-response assay for cancer patients to individualize their chemotherapy. Collagen sponge-gel-supported histoculture has been shown to maintain tissue architecture and function in vitro and has been utilized to develop the histoculture drug-response assay (HDRA) for individualizing chemotherapy. In order to evaluate the HDRA with the 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide end point for clinical use, chemosensitivity to mitomycin C, doxorubicin, 5-fluorouracil, and cisplatin of 107 advanced gastric and 109 advanced colorectal cancers was determined in vitro in a correlative clinical trial. Two hundred eight (96.3%) of 216 of the patient specimens were evaluable in the HDRA. Thirty-eight patients with remaining measurable lesions after surgery were evaluable for comparison of the effects of chemotherapy in the HDRA with clinical outcome. Their overall response in the HDRA to all four drugs correlated to published historical data. Twenty-nine patients were treated with drugs shown to be ineffective in the HDRA, and all 29 cases showed clinical chemoresistance. In nine patients treated with drugs shown to be effective in the HDRA, six showed clinical chemoresponse and three showed arrest of disease progression. The correlation rate of the assay to clinical drug-sensitivity response was thus calculated to be 92.1% (35/38), with 100% (29/29) true-negative and 66.7% (6/9) true-positive rates, 100% (6/6) sensitivity, and 90.6% (29/32) specificity. Thirty-two patients with stage III and IV gastric cancer without remaining measurable tumor lesions after surgery were treated with mitomycin C and a fluoropyrimidine adjuvantly. The survival rate of 10 patients whose tumors were sensitive to either mitomycin C and/or 5-fluorouracil in the assay was significantly (P < 0.005) better than that of 22 patients whose tumors were shown to be insensitive to both drugs. Twenty-nine patients with stage III and IV colorectal cancer without remaining measurable tumor lesions after surgery were treated with fluoropyrimidines adjuvantly. The recurrence-free survival rate of 7 patients whose tumors were sensitive to 5-fluorouracil in the assay was significantly (P < 0.05) better than that of 22 patients whose tumors were insensitive. Thus the HDRA with the 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide end point should be of clinical value to choose optimal chemotherapy for response as well as for survival.