Review of experience with interferon and drug sensitivity testing of ovarian carcinoma in semisolid agar culture.

Studies were performed with a semisolid agar culture system to determine the in vitro sensitivity of human ovarian carcinoma to human leukocyte interferon (IFN-alpha) and standard chemotherapeutic agents (adriamycin, cis-platinum, hexamethylmelamine, melphalan, and velban). Growth in culture occurred in 67% of samples derived from ascitic fluid and 71% of these exhibited reduction in tumor colony number by greater than or equal to 25% in response to 300 units interferon/ml. The responsiveness of the ascitic fluid samples to interferon ran parallel to the overall responsiveness to the other agents tested. Sensitivity to interferon was not related to the histology or grade of the tumor or to the stage of the disease. As compared with cell suspensions prepared from ascitic fluid samples, solid tumor samples had markedly lower viability, 39% vs 89%, and had more tumor cells, 81% vs 28%. Also, colonies derived from solid tumor samples were less sensitive to interferon and more sensitive to cis-platinum and adriamycin than were ascitic fluid-derived colonies. Three of four ascitic fluid samples showed a reduction in tumor colony number of greater than or equal to 25%, whereas none of the solid tumor samples obtained from the same donors were affected by interferon to that degree. Retrospective analysis of drug testing (exclusive of interferon) for in vitro: in vivo correlations revealed that in 58% of 12 evaluable situations, when a single drug (or at least one of a group of drugs) gave a positive response in vitro, stabilization or regression of the tumor occurred in vivo after treatment with the drug. These studies prove the utility of the semisolid agar culture system for assessing the antiproliferative effects of interferon against ovarian carcinoma, and will be of utility in the future for assessing whether various types of interferon have the same degree, range, and mechanism of action of antitumor effect.