An improved method for analyzing survival data from combination chemotherapy experiments

In this paper, we present a new method for analyzing survival data from combination chemotherapy experiments. The analysis consists of relating survival to the dosage level of each drug in the combination and using response surface techniques to determine the importance of drug interactions and to estimate optimal doses. A combination experiment using cyclophosphamide, mechlorethamine, and mitomycin C in early L1210 leukemia, advanced L1210 leukemia, and advanced P388 leukemia is used to illustrate the analyses. A therapeutic synergism has been shown. As a result of the various drug interactions, the predicted optimal dose of mitomycin C is found to be zero. This result was duplicated in each tumor system studied.