Statistical Properties of The Traditional Algorithm-based Designs for Phase I Cancer Clinical Trials
By Yong Lin and Weichung J. Shih
University of Medicine and Dentistry of New Jersey,
Robert Wood Johnson Medical School,
The Cancer Institute of New Jersey, New Brunswick, NJ 08901, USA
linyo@umdnj.edu & shihwj@umdnj.edu
ABSTRACT
Although there are several new designs for phase I cancer clinical trials including the continual reassessment method and accelerated titration design, the traditional algorithm-based designs, like the `3+3' design, are still widely used in practice because of their simplicity for clinical investigators to carry out the experiment. In this paper, we study some key statistical properties of the traditional algorithm-based designs in a general framework and derive the exact formulas for the corresponding statistical quantities. These quantities are important for the investigator to gain insights regarding the design of the trial, which are: (i) Probability of a dose being chosen as the MTD (maximum-tolerated dose); (ii) Expected number of patients treated at each dose level; (iii) Target toxicity level (i.e., the expected DLT, dose-limiting toxicity, incidences at the MTD); (iv) Expected DLT incidences at each dose level; (v) Expected overall DLT incidences in the trial. Real examples of clinical trials are given, and a computer program to do the calculation can be found here.
Download the paper in
Download the Splus computer program
Introduction to the Splus program
Last Updated: 10/5/00