Monte carlo evaluation of several sequential optimization techniques when the response is time to an event
Clinical trials are frequently concerned with the comparison of an experimental treatment and a control. In this paper, we are concerned with the use of a clinical trial for estimating the optimum combination of a group of drugs, especially in connection with the combination chemotherapy of cancer. Three methods designed for this situation are compared by a Monte Carlo study. Two methods are variations of the method of steepest ascent while another is a modification of a simplex search procedure. It was found that an adaptive modification of the method of steepest ascent was most efficient.