Several models of pancreatic adenocarcinoma are now available for experimental evaluation of newer chemotherapeutic agents. The present study represents an attempt to develop a rapid in vitro screening technique that would allow prediction of cytotoxic activity (or lack thereof) as reliably as the clonogenic or colony formation assay. To this end, seven drugs (cisplatin, dactinomycin, doxorubicin, 5-fluorouracil, menogarol, mitoxantrone, and streptozocin) were tested against two pancreatic adenocarcinoma cell lines using a standard colony formation assay and a 24-hr microcytotoxicity assay. The cell lines tested were PANC-1, of human poorly differentiated pancreatic adenocarcinoma origin, and WD PaCa, of hamster well-differentiated pancreatic adenocarcinoma origin. The dose-survival curves and resulting determinations of drug dose (µg/ml/1-hr exposure) at which there is a 50% inhibition of survival as compared to controls were compared for the two cell lines by each assay system. Lack of correlation of the two assays and considerable interdrug and inter-cell line variation were found. In addition, the microcytotoxicity assay was felt to underestimate the in vitro drug sensitivity of PANC-1 to three drugs (dactinomycin, doxorubicin, and mitoxantrone) and of WD PaCa to two drugs (5-fluorouracil and mitoxantrone). Despite the possible utility of the microcytotoxicity assay with other experimental models, the colony formation assay technique appears to provide the most reliable in vitro assessment of antineoplastic activity for pancreatic adenocarcinoma cell lines and should continue to be the standard to which other assay systems are compared.
Supported by Grant CA34170, awarded by the National Cancer Institute, Department of Health and Human Services. Presented in part at the Joint Meetings of the American Pancreatic Association and National Pancreatic Cancer Project, Chicago, November 1982.