Optimization of individualized dynamic treatment regimes for recurrent diseases

Patients with cancer or other recurrent diseases may undergo a long process of initial treatment, disease recurrences, and salvage treatments. It is important to optimize the multi-stage treatment sequence in this process to maximally prolong patients' survival. Comparing disease-free survival for each treatment stage over penalizes disease recurrences but under penalizes treatment-related mortalities. Moreover, treatment regimes used in practice are dynamic; that is, the choice of next treatment depends on a patient's responses to previous therapies. In this article, using accelerated failure time models, we develop a method to optimize such dynamic treatment regimes. This method utilizes all the longitudinal data collected during the multi-stage process of disease recurrences and treatments, and identifies the optimal dynamic treatment regime for each individual patient by maximizing his or her expected overall survival. We illustrate the application of this method using data from a study of acute myeloid leukemia, for which the optimal treatment strategies for different patient subgroups are identified.

[1]  Inbal Nahum-Shani,et al.  Q-learning: a data analysis method for constructing adaptive interventions. , 2012, Psychological methods.

[2]  Bin Nan,et al.  Doubly Penalized Buckley–James Method for Survival Data with High‐Dimensional Covariates , 2008, Biometrics.

[3]  Abdus S Wahed,et al.  Evaluating joint effects of induction–salvage treatment regimes on overall survival in acute leukaemia , 2013, Journal of the Royal Statistical Society. Series C, Applied statistics.

[4]  J. Robins.  A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect , 1986 .

[5]  Debashis Ghosh,et al.  MARGINAL REGRESSION MODELS FOR RECURRENT AND TERMINAL EVENTS , 2002 .

[6]  Eric B. Laber,et al.  A Robust Method for Estimating Optimal Treatment Regimes , 2012, Biometrics.

[7]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data: Kalbfleisch/The Statistical , 2002 .

[8]  Limin Peng,et al.  Rank Estimation of Accelerated Lifetime Models With Dependent Censoring , 2006 .

[9]  T Cai,et al.  Regularized Estimation for the Accelerated Failure Time Model , 2009, Biometrics.

[10]  Jian Huang,et al.  Regularized Estimation in the Accelerated Failure Time Model with High‐Dimensional Covariates , 2006, Biometrics.

[11]  J. Robins,et al.  Recovery of Information and Adjustment for Dependent Censoring Using Surrogate Markers , 1992 .

[12]  Brent A. Johnson,et al.  Penalized Estimating Functions and Variable Selection in Semiparametric Regression Models , 2008, Journal of the American Statistical Association.

[13]  Peter F Thall,et al.  Evaluation of Viable Dynamic Treatment Regimes in a Sequentially Randomized Trial of Advanced Prostate Cancer , 2012, Journal of the American Statistical Association.

[14]  M. J. van der Laan,et al.  The International Journal of Biostatistics Targeted Maximum Likelihood Estimation for Dynamic Treatment Regimes in Sequentially Randomized Controlled Trials , 2012 .

[15]  H. Zou,et al.  Regularization and variable selection via the elastic net , 2005 .

[16]  P F Thall,et al.  Randomized phase II study of fludarabine + cytosine arabinoside + idarubicin +/- all-trans retinoic acid +/- granulocyte colony-stimulating factor in poor prognosis newly diagnosed acute myeloid leukemia and myelodysplastic syndrome. , 1999, Blood.

[17]  Robin Henderson,et al.  Estimation of optimal dynamic anticoagulation regimes from observational data: a regret-based approach. , 2006, Statistics in medicine.

[18]  Sihai Dave Zhao,et al.  The Dantzig Selector for Censored Linear Regression Models. , 2014, Statistica Sinica.

[19]  A. Wahed,et al.  Comparison of treatment regimes with adjustment for auxiliary variables , 2011 .

[20]  A. Philip Dawid,et al.  Identifying the consequences of dynamic treatment strategies: A decision-theoretic overview , 2010, ArXiv.

[21]  S. Murphy,et al.  An experimental design for the development of adaptive treatment strategies , 2005, Statistics in medicine.

[22]  Susan Murphy,et al.  Inference for non-regular parameters in optimal dynamic treatment regimes , 2010, Statistical methods in medical research.

[23]  M. Kosorok,et al.  Reinforcement learning design for cancer clinical trials , 2009, Statistics in medicine.

[24]  P. Pisters,et al.  Estimation of the Causal Effects on Survival of Two‐Stage Nonrandomized Treatment Sequences for Recurrent Diseases , 2006, Biometrics.

[25]  Erica E M Moodie,et al.  Demystifying Optimal Dynamic Treatment Regimes , 2007, Biometrics.

[26]  Debashis Ghosh,et al.  Semiparametric analysis of recurrent events data in the presence of dependent censoring. , 2003, Biometrics.

[27]  Jianwen Cai,et al.  Some Graphical Displays and Marginal Regression Analyses for Recurrent Failure Times and Time Dependent Covariates , 1993 .

[28]  S. Murphy,et al.  Optimal dynamic treatment regimes , 2003 .

[29]  Philip W. Lavori,et al.  A design for testing clinical strategies: biased adaptive within‐subject randomization , 2000 .

[30]  Robert L. Strawderman,et al.  Higher-Order Asymptotic Approximation: Laplace, Saddlepoint, and Related Methods , 2000 .

[31]  M. Kosorok,et al.  Q-LEARNING WITH CENSORED DATA. , 2012, Annals of statistics.

[32]  M. Kosorok,et al.  Reinforcement Learning Strategies for Clinical Trials in Nonsmall Cell Lung Cancer , 2011, Biometrics.

[33]  R. Gill,et al.  Cox's regression model for counting processes: a large sample study : (preprint) , 1982 .

[34]  Phil Ansell,et al.  Regret‐Regression for Optimal Dynamic Treatment Regimes , 2010, Biometrics.

[35]  Zhiqiang Tan,et al.  Comment: Understanding OR, PS and DR , 2007, 0804.2969.

[36]  Jian Huang,et al.  Variable selection in the accelerated failure time model via the bridge method , 2010, Lifetime data analysis.

[37]  James M. Robins,et al.  Causal Inference from Complex Longitudinal Data , 1997 .

[38]  H. White.  A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity , 1980 .

[39]  D. Lin,et al.  Nonparametric Analysis of Recurrent Events and Death , 2000, Biometrics.

[40]  Peter Dayan,et al.  Technical Note: Q-Learning , 2004, Machine Learning.

[41]  Jianqing Fan,et al.  Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties , 2001 .

[42]  J. Robins.  The control of confounding by intermediate variables. , 1989, Statistics in medicine.

[43]  Richard Kay,et al.  Comparing proportional hazards and accelerated failure time models: an application in influenza , 2006, Pharmaceutical statistics.

[44]  Terence Tao,et al.  The Dantzig selector: Statistical estimation when P is much larger than n , 2005, math/0506081.

[45]  D. Rubin,et al.  Ignorability and Coarse Data , 1991 .

[46]  P. Thall,et al.  Adaptive therapy for androgen-independent prostate cancer: a randomized selection trial of four regimens. , 2007, Journal of the National Cancer Institute.

[47]  Yu Cheng,et al.  Quantile Regression for Doubly Censored Data , 2012, Biometrics.

[48]  H. Sung,et al.  Evaluating multiple treatment courses in clinical trials. , 2000, Statistics in medicine.

[49]  Limin Peng,et al.  Regression Modeling of Semicompeting Risks Data , 2007, Biometrics.

[50]  N. Keiding,et al.  Estimation of dynamic treatment strategies for maintenance therapy of children with acute lymphoblastic leukaemia: an application of history‐adjusted marginal structural models , 2012, Statistics in medicine.

[51]  Tristan Zajonc,et al.  Bayesian Inference for Dynamic Treatment Regimes: Mobility, Equity, and Efficiency in Student Tracking , 2010 .

[52]  Lei Liu,et al.  A Joint Frailty Model for Survival and Gap Times Between Recurrent Events , 2007, Biometrics.

[53]  Peter Dayan,et al.  Q-learning , 1992, Machine Learning.

[54]  Susan A. Murphy,et al.  A-Learning for approximate planning , 2004 .

[55]  R. Wolfe,et al.  Shared Frailty Models for Recurrent Events and a Terminal Event , 2004, Biometrics.

[56]  S A Murphy,et al.  Screening Experiments for Developing Dynamic Treatment Regimes , 2009, Journal of the American Statistical Association.

[57]  D. Cox.  Regression Models and Life-Tables , 1972 .

[58]  Xuelin Huang,et al.  Analysis of multi‐stage treatments for recurrent diseases , 2012, Statistics in medicine.

[59]  J M Robins,et al.  Marginal Mean Models for Dynamic Regimes , 2001, Journal of the American Statistical Association.

[60]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[61]  A A Tsiatis,et al.  Efficient Estimation of the Distribution of Quality‐Adjusted Survival Time , 1999, Biometrics.

[62]  S. Keleş,et al.  Recurrent events analysis in the presence of time‐dependent covariates and dependent censoring , 2004 .