Randomized Discontinuation Trials With Binary Outcomes

  • Valerii V. FedorovEmail author
  • Tao Liu


Randomized discontinuation trial (RDT) has gained popularity across a number of therapeutic areas. Oncology is one of the most known. In the simplest case, at the initial open-label stage all patients are treated with the experimental treatment to identify a population of responders. This stage is followed by a randomized two-arm trial to compare two conditions, for example, treatment versus placebo. Potentially RDT increases the efficiency of trials relatively to traditional designs gaining information from a sensitized population if the open-label stage provides a reliable separation of responders from nonresponders. Often a sensitized population is called an enriched population, and respectively, the RDT is called “enrichment experiment.” We compare RDT with the traditional two-arm randomized clinical trials (RCT) for binary outcomes assuming that the population of interest consists of three groups: placebo responders, treatment-only responders, and nonresponders. Our results are derived in the “parameter estimation” setting and they are based on the comparison of estimator variances. We identify conditions under which RDT is either superior or inferior to RCT in terms of response rates, misclassification rates, and clinical ethics. Extension of our results to design optimization, hypothesis testing, and sample size calculation is rather straightforward.


Enrichment experiments Experimental design Randomized discontinuation trial Open-label first stage 

AMS Subject Classifications



Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Amery, W., and J. Dony. 1975. Clinical trial design avoiding undue placebo treatment. J. Clin. Pharmacol, October, 674–679.CrossRefGoogle Scholar
  2. Burzykowski, T., G. Molenberghs, and M. Buyse. 2005. The evaluation of surrogate endpoints. New York, NY: Springer.CrossRefGoogle Scholar
  3. Capra, W. B. 2004. Comparing the power of the discontinuation design to that of the classic randomized design on time-to-event endpoints. Controlled Clin. Trials, 25, 168–177.CrossRefGoogle Scholar
  4. Chiron, C., O. Dulac, and L. Gram. 1996. Vigabatrin withdrawal randomized study in children. Epilepsy Res., 25, 209–215.CrossRefGoogle Scholar
  5. Fedorov, V. V., and A. C. Atkinson. 1988. The optimum design of experiments in the presence of uncontrolled variability and prior information. In Optimal design and analysis of experiments, ed. Y. Dodge, V. V. Fedorov, and H. P. Wynn, 327–344. New York, NY: Elsevier Science.Google Scholar
  6. Fedorov, V. V., and P. Hackl. 1997. Model-oriented design of experiments. New York, NY: Springer-Verlag.CrossRefGoogle Scholar
  7. Fedorov, V. V., and T. Liu. 2005. Randomized discontinuation trials: Design and efficiency. GlaxoSmithKline Biomedical Data Science Technical Report, 2005-3. Available upon request ( Scholar
  8. Fedorov, V. V., and T. Liu. 2007. Enrichment design. In Wiley encyclopedia of clinical trials, 1–8. New York, NY: John Wiley & Sons.Google Scholar
  9. Fedorov, V. V., F. Mannino, and R. Zhang. 2008. Consequences of dichotomization. Pharm. Stat., 8, 50–61.CrossRefGoogle Scholar
  10. Freidlin, B., and R. Simon. 2005. Evaluation of randomized discontinuation design. J. Clin. Oncol., 23(22), 5094–5098.CrossRefGoogle Scholar
  11. Hallstrom, A. P., and L. Friedman. 1991. Randomizing responders. Controlled Clin. Trials, 12, 486–503.CrossRefGoogle Scholar
  12. Kopec, J., M. Abrahamowicz, and J. Esdaile. 1993. Randomized discontinuation trials: Utility and efficiency. J. Clin. Epidemiol., 46, 959–971.CrossRefGoogle Scholar
  13. Korn, E. L., S. G. Arbuck, J. M. Pulda, R. Simon, R. S. Kaplan, and M. C. Christian. 2001. Clinical trial designs for cytostatic agents: Are new approaches needed? J. Clin. Oncol., 19, 265–272.CrossRefGoogle Scholar
  14. Lehmann, E., and G. Casella. 1998. Theory of point estimation. New York, NY: Springer.zbMATHGoogle Scholar
  15. MacLehose, R. R., B. C. Reeves, I. M. Harvey, T. A. Sheldon, I. T. Russell, and A. M. Black. 2000. A systematic review of comparisons of effect sizes derived from randomized and non-randomized studies. Health Technol. Assess., 41–154.
  16. Pablos-Méndez, A., R. G. Barr, and S. Shea. 1998. Run-in periods in randomized trials. J. Am. Med. Assoc., 279, 222–225.CrossRefGoogle Scholar
  17. Rosner, G. L., W. Stadler, and M. J. Ratain. 2002. Randomized discontinuation design: Application to cytostatic antineoplastic agents. J. Clin. Oncol., 20, 4478–4484.CrossRefGoogle Scholar
  18. Simon, R., and A. Maitournam. 2004. Evaluating the efficiency of targeted designs for randomized clinical trials. Clin. Cancer Res., 10, 7659–6763.Google Scholar
  19. Senn, S. J. 2001. Individual therapy: New dawn or false dawn. Drug Information J., 35, 1479–1494.CrossRefGoogle Scholar
  20. Senn, S. J. 2004. Individual response to treatment: Is it a valid assumption? Br. Med. J., 329, 966–968.CrossRefGoogle Scholar
  21. Shao, J. 1999. Mathematical statistics. New York, NY: Springer.zbMATHGoogle Scholar
  22. Stadler, W. M., G., Rosner, E., Small, D., Hollis, B., Rini, S. D. Zaentz, and J. Mahoney. 2005. Successful implementation of the randomized discontinuation trial design: An application to the study of the putative antiangiogenic agent carboxyaminoimidazole in renal cell carcinoma—calgb 69901. J. Clin. Oncol., 23, 3726–3732.CrossRefGoogle Scholar
  23. Trippa, L., and G. L. Rosner. 2012. Bayesian enrichment strategies for randomized discontinuation trials. Biometrics, 68, 203–225.MathSciNetCrossRefGoogle Scholar

Copyright information

© Grace Scientific Publishing 2014

Authors and Affiliations

  1. 1.QuintilesMorrisvilleUSA
  2. 2.Department of Biostatistics, Center for Statistical SciencesBrown UniversityProvidenceUSA

Personalised recommendations