Individualized treatment effects with censored data via fully nonparametric Bayesian accelerated failure time models.

Individuals often respond differently to identical treatments, and characterizing such variability in treatment response is an important aim in the practice of personalized medicine. In this article, we describe a nonparametric accelerated failure time model that can be used to analyze heterogeneous treatment effects (HTE) when patient outcomes are time-to-event. By utilizing Bayesian additive regression trees and a mean-constrained Dirichlet process mixture model, our approach offers a flexible model for the regression function while placing few restrictions on the baseline hazard. Our nonparametric method leads to natural estimates of individual treatment effect and has the flexibility to address many major goals of HTE assessment. Moreover, our method requires little user input in terms of model specification for treatment covariate interactions or for tuning parameter selection. Our procedure shows strong predictive performance while also exhibiting good frequentist properties in terms of parameter coverage and mitigation of spurious findings of HTE. We illustrate the merits of our proposed approach with a detailed analysis of two large clinical trials (N = 6769) for the prevention and treatment of congestive heart failure using an angiotensin-converting enzyme inhibitor. The analysis revealed considerable evidence for the presence of HTE in both trials as demonstrated by substantial estimated variation in treatment effect and by high proportions of patients exhibiting strong evidence of having treatment effects which differ from the overall treatment effect.

[1]  Lancelot F. James,et al.  Gibbs Sampling Methods for Stick-Breaking Priors , 2001 .

[2]  Menggang Yu,et al.  Regularized outcome weighted subgroup identification for differential treatment effects , 2015, Biometrics.

[3]  Hajime Yamato.  Characteristic Functions of Means of Distributions Chosen from a Dirichlet Process , 1984 .

[4]  H. Chipman,et al.  Bayesian CART Model Search , 1998 .

[5]  J. M. Taylor,et al.  Subgroup identification from randomized clinical trial data , 2011, Statistics in medicine.

[6]  E. Lesaffre,et al.  Accelerated Failure Time Model for Arbitrarily Censored Data With Smoothed Error Distribution , 2005 .

[7]  Els Goetghebeur,et al.  Model evaluation based on the sampling distribution of estimated absolute prediction error , 2007 .

[8]  Eric B. Laber,et al.  Doubly Robust Learning for Estimating Individualized Treatment with Censored Data. , 2015, Biometrika.

[9]  Daniel L. Oberski,et al.  Identification of predicted individual treatment effects in randomized clinical trials , 2018, Statistical methods in medical research.

[10]  Adam Kapelner,et al.  Prediction with missing data via Bayesian Additive Regression Trees , 2013, ArXiv.

[11]  Trevor J. Hastie,et al.  Confidence intervals for random forests: the jackknife and the infinitesimal jackknife , 2013, J. Mach. Learn. Res..

[12]  Yu-Chuan Chen,et al.  Ensemble survival trees for identifying subpopulations in personalized medicine. , 2016, Biometrical journal. Biometrische Zeitschrift.

[13]  Purushottam W. Laud,et al.  Nonparametric survival analysis using Bayesian Additive Regression Trees (BART) , 2016, Statistics in medicine.

[14]  Adam Kapelner,et al.  Bayesian Additive Regression Trees With Parametric Models of Heteroskedasticity , 2014 .

[15]  Stefan Wager,et al.  Estimation and Inference of Heterogeneous Treatment Effects using Random Forests , 2015, Journal of the American Statistical Association.

[16]  T. Hanson.  Modeling censored lifetime data using a mixture of gammas baseline , 2006 .

[17]  James M. Robins,et al.  Semiparametric estimation of an accelerated failure time model with time-dependent covariates , 1992 .

[18]  H. Chipman,et al.  BART: Bayesian Additive Regression Trees , 2008, 0806.3286.

[19]  Jennifer L. Hill,et al.  Bayesian Nonparametric Modeling for Causal Inference , 2011 .

[20]  J. Berger,et al.  A Bayesian Approach to Subgroup Identification , 2014, Journal of biopharmaceutical statistics.

[21]  David J. Hand,et al.  Good methods for coping with missing data in decision trees , 2008, Pattern Recognit. Lett..

[22]  T. Louis,et al.  Triple‐goal estimates in two‐stage hierarchical models , 1998 .

[23]  David B. Dunson,et al.  Semiparametric Bayes hierarchical models with mean and variance constraints , 2010, Comput. Stat. Data Anal..

[24]  Thomas A. Louis,et al.  Nonparametric analysis of an accelerated failure time model , 1981 .

[25]  I. Lipkovich,et al.  Tutorial in biostatistics: data‐driven subgroup identification and analysis in clinical trials , 2017, Statistics in medicine.

[26]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[27]  T. Cai,et al.  Identifying predictive markers for personalized treatment selection , 2016, Biometrics.

[28]  W Shen,et al.  Innovations in bayes and empirical bayes methods: estimating parameters, populations and ranks. , 1999, Statistics in medicine.

[29]  Lynn Kuo,et al.  Bayesian semiparametric inference for the accelerated failure‐time model , 1997 .

[30]  Lee-Jen Wei,et al.  The accelerated failure time model: a useful alternative to the Cox regression model in survival analysis. , 1992, Statistics in medicine.

[31]  Lu Tian,et al.  A Simple Method for Detecting Interactions between a Treatment and a Large Number of Covariates , 2012, 1212.2995.

[32]  Xiaogang Su,et al.  Subgroup Analysis via Recursive Partitioning , 2009 .

[33]  David M Kent,et al.  Limitations of applying summary results of clinical trials to individual patients: the need for risk stratification. , 2007, JAMA.

[34]  W. Loh,et al.  A regression tree approach to identifying subgroups with differential treatment effects , 2014, Statistics in medicine.

[35]  Donglin Zeng,et al.  Estimating Individualized Treatment Rules Using Outcome Weighted Learning , 2012, Journal of the American Statistical Association.

[36]  Emmanuel Lesaffre,et al.  BAYESIAN ACCELERATED FAILURE TIME MODEL FOR CORRELATED INTERVAL-CENSORED DATA WITH A NORMAL MIXTURE AS ERROR DISTRIBUTION , 2007 .

[37]  Michael Branson,et al.  Bayesian models for subgroup analysis in clinical trials , 2011, Clinical trials.

[38]  H. Weisberg,et al.  Post hoc subgroups in clinical trials: Anathema or analytics? , 2015, Clinical trials.

[39]  Salim Yusuf,et al.  Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. , 1991, The New England journal of medicine.

[40]  W. Johnson,et al.  Modeling Regression Error With a Mixture of Polya Trees , 2002 .

[41]  Ronald Christensen,et al.  Modelling accelerated failure time with a Dirichlet process , 1988 .

[42]  Kim-Anh Do,et al.  Bayesian ensemble methods for survival prediction in gene expression data , 2011, Bioinform..

[43]  Lihui Zhao,et al.  Predicting the restricted mean event time with the subject's baseline covariates in survival analysis. , 2014, Biostatistics.