Comparative studies on in vitro human tumor clonogenic assay (HTCA) and in vivo nude mouse-isotope assay (NM-IA).

Comparative studies between the in vitro human tumor clonogenic assay (HTCA) and nude mouse-isotope assay (NM-IA), in which the final evaluation was made with 3H-thymidine incorporations of tumor cells transplanted into nude mice, were performed simultaneously on 60 fresh human tumors. Tissues used included 27 gastric cancers, 10 breast cancers, 7 colorectal cancers, 4 gallbladder cancers, 4 sarcomas, 3 lymphomas, and 5 other tumors. Mitomycin C (MMC), 5-fluorouracil (5-FU), cyclophosphamide (CPM) and adriamycin (ADM) were tested. The overall evaluable rate was 66.7 per cent in HTCA and 83.3 per cent in NM-IA, respectively. When the per cent survival in HTCA and the per cent inhibition in NM-IA less than 50 per cent was defined as drug sensitive, the drug sensitive rates of MMC, 5-FU, CPM, and ADM were 23.1, 16.7, 11.8, and 27.8 per cent in HTCA and 17.6, 18.4, 22.4, and 25.5 per cent in NM-IA, respectively. Although statistically significant correlations between the results of HTCA and those of NM-IA were obtained for MMC and ADM, no correlation was observed for 5-FU and CPM. The overall predictive accuracy rate of clinical response was 83.3 per cent (true positive rate 50 per cent and true negative rate 92.9 per cent) in HTCA and 76.0 per cent (true positive rate 37.5 per cent and true negative rate 93.8 per cent) in NM-IA, respectively.