Proposal for the Application of Xenografts in Screening for New Anticancer Agents and in Selecting Tumor Types for Phase II Clinical Trials

Abstract

The selection of drugs for clinical evaluation in can cer chemotherapy is usually based on the activity in several antitumor test systems. If a drug has a promising activity profile, can be produced in a reliable dosage form and does not show prohibitive toxicity in animals, it is a candidate for clinical evaluation. This is different from most other areas of drug development where a clinical candidate must perform well also in preclinical toxicology studies, thus having a large therapeutic index. In the past decades antitumor testing has mainly been performed in murine tumors in vivo and in vitro. A systematic screen to discover and characterize new lead compounds has been applied by the American National Cancer Institute (NCI) (1). Other institutes performing preclinical antitumor testing used the same or similar models and methods as the NCI. Although the NCI screening system was revised several times, the primary selection of new lead compounds depended heavily on the use of the murine leukemia models L1210 and P388, as the primary screen (2, 3). In 1975 several murine solid tumors and 3 human tumor xenografts (tumor lines) in nude mice were included in the tumor panel, i.e., secondary screen. These tumor models were expected to select drugs active in human solid tumors and predict clinical activity in the major human solid tumors, i.e., lung, breast and colorectal cancer (4, 5, 6).