Journal of Statistical Theory and Practice

, Volume 9, Issue 2, pp 266–287 | Cite as

Pattern-Mixture-Type Estimation and Testing of Neuroblastoma Treatment Regimes

  • Xinyu Tang
  • Abdus S. Wahed


Sequentially randomized designs are commonly used in biomedical research, particularly in clinical trials, to assess and compare the effects of different treatment regimes. In such designs, eligible patients are first randomized to one of the initial therapies, and then patients with some intermediate response (e.g., without progressive diseases) are randomized to one of the maintenance therapies. The goal is to evaluate dynamic treatment regimes consisting of an initial therapy, the intermediate response, and a maintenance therapy. In this article, we demonstrate the use of the pattern-mixture model (commonly used for analyzing missing data) for estimating the effects of treatment regimes based on familiar survival analysis techniques such as Nelson-Aalen and parametric models. Moreover, we demonstrate how to use estimates from pattern-mixture models to test for the differences across treatment regimes in a weighted log-rank setting. We investigate the properties of the proposed estimators and test in a Monte Carlo simulation study. Finally, we demonstrate the methods using the long-term survival data from the high-risk neuroblastoma study.


Dynamic treatment regime High-risk neuroblastoma study Parametric model Pattern-mixture model proportional hazard model Sequentially randomized design 

AMS Subject Classification

62L05 62N02 62N03 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Feng, W., and A. S. Wahed. 2008. Supremum weighted log-rank test and sample size for comparing two-stage adaptive treatment strategies, Biometrika, 95, 695–707.MathSciNetCrossRefGoogle Scholar
  2. Fleming T. R., and D. P. Harrington. 1991. Counting processes and survival analysis. New York, NY: John Wiley & Sons.zbMATHGoogle Scholar
  3. Guo, X., and A. A. Tsiatis. 2005. A weighted risk estimator for survival distributions in two-stage randomization designs with censored survival data. Int. J. Biostat., 1, 1–15.MathSciNetCrossRefGoogle Scholar
  4. Hernan, M. A., E. Lanoy, D. Costagliola, and J. M. Robins. 2006. Comparison of dynamic treatment regimes via inverse probability weighting. Basic Clin. Pharmacol. Toxicol., 98, 237–242.CrossRefGoogle Scholar
  5. Klein, J. P., and M. L. Moeschberger. 2003. Survival analysis techniques for censored and truncated data. New York, NY: Springer.zbMATHGoogle Scholar
  6. Little, R. J. A. 1993. Pattern-mixture models for multivariate incomplete data. J. Am. Stat. Assoc., 88, 125–134.zbMATHGoogle Scholar
  7. Lokhnygina, Y., and J. D. Helterbrand. 2007. Cox regression methods for two-stage randomization designs. Biometrics, 63, 422–428.MathSciNetCrossRefGoogle Scholar
  8. Lunceford, J. K., M. Davidian, and A. A. Tsiatis. 2002. Estimation of survival distributions of treatment policies in two-stage randomization designs in clinical trials. Biometrics, 58, 48–57.MathSciNetCrossRefGoogle Scholar
  9. Matthay, K. K., J. G. Villablanca, R. C. Seeger, D. O. Stram, R. E. Harris, N. K. Ramsay, P. Swift, H. Shimada, C. T. Black, G. M. Brodeur, R. B. Gerbing, and C. P. Reynolds. 1999. Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation, and 13-cis-retinoic acid. N. Engl. J. Med., 341, 1165–1173.CrossRefGoogle Scholar
  10. Matthay, K. K., C. P. Reynolds, R. C. Seeger, H. Shimada, E. S. Adkins, D. Haas-Kogan, R. B. Gerbing, W. B. London, and J. G. Villablanca. 2009. Long-term results for children with high-risk neuroblastoma treated on a randomized trial of myeloablative therapy followed by 13-cis-retinoic acid: A Children’s Oncology Group study. J. Clin. Oncol., 27, 1007–1013.CrossRefGoogle Scholar
  11. Molenbergh, G., and M. G. Kenward. 2007. Missing data in clinical studies. Chichester, England: John Wiley & Sons Ltd.CrossRefGoogle Scholar
  12. Murphy, S. A. 2003. Optimal dynamic treatment regimes, J. R. Stat. Society Ser. B, 65, 331–355.MathSciNetCrossRefGoogle Scholar
  13. Murphy, S. A., and D. Bingham. 2009. Screening experiments for developing dynamic treatment regimes. J. Am. Stat. Assoc., 104, 391–409.MathSciNetCrossRefGoogle Scholar
  14. Murphy, S. A., M. J. van der Laan, and J. M. Robins. 2001. Marginal mean models for dynamic regimes. J. Am. Stat. Assoc., 96, 1410–1423.MathSciNetCrossRefGoogle Scholar
  15. Robins, J. M. 1986. A new approach to causal inference in mortality studies with sustained exposure periods-application to control of the healthy survivor effect. Math. Model., 7, 1393–1512.MathSciNetCrossRefGoogle Scholar
  16. Robins, J. M. 1987. Addendum to “A new approach to causal inference in mortality studies with sustained exposure periods-application to control of the healthy survivor effect” [Math Model 1986;7:1393-1512]. Computers Math. Appl., 14, 923–945.MathSciNetCrossRefGoogle Scholar
  17. Wahed, A. S., and A. A. Tsiatis. 2004. Optimal estimator for the survival distribution and related quantities for treatment policies in two-stage randomization designs in clinical trials. Biometrics, 60, 124–133.MathSciNetCrossRefGoogle Scholar
  18. Wahed, A. S., and A. A. Tsiatis. 2006. Semiparametric efficient estimation of survival distributions in two-stage randomization designs in clinical trials with censored data. Biometrika, 93, 163–177.MathSciNetCrossRefGoogle Scholar

Copyright information

© Grace Scientific Publishing 2015

Authors and Affiliations

  1. 1.College of Medicine, PediatricsUniversity of Arkansas for Medical SciencesLittle RockUSA
  2. 2.Graduate School of Public HealthUniversity of PittsburghPittsburghUSA

Personalised recommendations