Statistical Science

Approximate Dynamic Programming and Its Applications to the Design of Phase I Cancer Trials

Jay Bartroff and Tze Leung Lai

Full-text: Open access


Optimal design of a Phase I cancer trial can be formulated as a stochastic optimization problem. By making use of recent advances in approximate dynamic programming to tackle the problem, we develop an approximation of the Bayesian optimal design. The resulting design is a convex combination of a “treatment” design, such as Babb et al.’s (1998) escalation with overdose control, and a “learning” design, such as Haines et al.’s (2003) c-optimal design, thus directly addressing the treatment versus experimentation dilemma inherent in Phase I trials and providing a simple and intuitive design for clinical use. Computational details are given and the proposed design is compared to existing designs in a simulation study. The design can also be readily modified to include a first stage that cautiously escalates doses similarly to traditional nonparametric step-up/down schemes, while validating the Bayesian parametric model for the efficient model-based design in the second stage.

Article information

Statist. Sci., Volume 25, Number 2 (2010), 245-257.

First available in Project Euclid: 19 November 2010

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Dynamic programming maximum tolerated dose Monte Carlo rollout stochastic optimization


Bartroff, Jay; Lai, Tze Leung. Approximate Dynamic Programming and Its Applications to the Design of Phase I Cancer Trials. Statist. Sci. 25 (2010), no. 2, 245--257. doi:10.1214/10-STS317.

Export citation


  • Abdelbasit, K. M. and Plackett, R. L. (1983). Experimental design for binary data. J. Amer. Statist. Assoc. 78 90–98.
  • Babb, J., Rogatko, A. and Zacks, S. (1998). Cancer phase I clinical trials: Efficient dose escalation with overdose control. Statist. Med. 17 1103–1120.
  • Bayard, D. S. (1991). A forward method for optimal stochastic nonlinear and adaptive control. IEEE Trans. Automat. Control 36 1046–1053.
  • Bertsekas, D. P. (2007). Dynamic Programming and Optimal Control 2, 3rd ed. Athena Scientific, Belmont, MA.
  • Cheung, Y. K. (2005). Coherence principles in dose-finding studies. Biometrika 92 863–873.
  • Dette, H., Melas, V. B. and Pepelyshev, A. (2004). Optimal designs for a class of nonlinear regression models. Ann. Statist. 32 2142–2167.
  • Eisenhauer, E. A., O’Dwyer, P. J., Christian, M. and Humphrey, J. S. (2000). Phase I clinical trial design in cancer drug development. J. Clin. Oncol. 18 684.
  • Fedorov, V. V. (1972). Theory of Optimal Experiments. Academic Press, New York.
  • Haines, L. M., Perevozskaya, I. and Rosenberger, W. F. (2003). Bayesian optimal design for phase I clinical trials. Biometrics 59 591–600.
  • Han, J., Lai, T. L. and Spivakovsky, V. (2006). Approximate policy optimization and adaptive control in regression models. Computat. Econom. 27 433–452.
  • Hastie, T., Tibshirani, R. and Friedman, J. H. (2001). The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer, New York.
  • Korn, E. L., Midthune, D., Chen, T. T., Rubinstein, L. V., Christian, M. C. and Simon, R. M. (1994). A comparison of two phase I trial designs. Statist. Med. 13 1799–1806.
  • Lai, T. L. and Robbins, H. (1979). Adaptive design and stochastic approximation. Ann. Statist. 7 1196–1221.
  • O’Quigley, J., Pepe, M. and Fisher, L. (1990). Continual reassessment method: A practical design for phase I clinical trials in cancer. Biometrics 46 33–48.
  • O’Quigley, J. and Shen, L. Z. (1996). Continual reassesment method: A likelihood approach. Biometrics 52 163–174.
  • Sacks, J. (1958). Asymptotic distribution of stochastic approximation procedures. Ann. Math. Statist. 29 373–405.
  • Storer, B. (1989). Design and analysis of phase I clinical trials. Biometrics 45 925–937.
  • Sutton, R. S. and Barto, A. G. (1998). Reinforcement Learning: An Introduction. MIT Press, Cambridge, MA.
  • Tighiouart, M., Rogatko, A. and Babb, J. S. (2005). Flexible Bayesian methods for cancer phase I clinical trials: Dose escalation with overdose control. Stat. Med. 24 2183–2196.
  • Von Hoff, D. and Turner, J. (1991). Response rates, duration of response, and dose response effects in phase I studies of antineoplastics. Investigational New Drugs 9 115–122.
  • Wu, C. F. J. (1985). Efficient sequential designs with binary data. J. Amer. Statist. Assoc. 80 974–984.