Optimal treatment regimes for competing risk data using doubly robust outcome weighted learning with bi-level variable selection

The goal of the optimal treatment regime is maximizing treatment benefits via personalized treatment assignments based on the observed patient and treatment characteristics. Parametric regression-based outcome learning approaches require exploring complex interplay between the outcome and treatment assignments adjusting for the patient and treatment covariates, yet correctly specifying such relationships is challenging. Thus, a robust method against misspecified models is desirable in practice. Parsimonious models are also desired to pursue a concise interpretation and to avoid including spurious predictors of the outcome or treatment benefits. These issues have not been comprehensively addressed in the presence of competing risks. Recognizing that competing risks and group variables are frequently present, we propose a doubly robust estimation with adaptive L 1 penalties to select important variables at both group and within-group levels for competing risks data. The proposed method is applied to hematopoietic cell transplantation data to personalize the graft source choice for treatment-related mortality (TRM). While the existing medical literature attempts to find a uniform solution ignoring the heterogeneity of the graft source effects on TRM, the analysis results show the effect of the graft source on TRM could be different depending on the patient-specific characteristics.

[1]  H. Zou.  The Adaptive Lasso and Its Oracle Properties , 2006 .

[2]  Mei‐jie Zhang,et al.  A Proportional Hazards Regression Model for the Subdistribution with Covariates‐adjusted Censoring Weight for Competing Risks Data , 2016, Scandinavian journal of statistics, theory and applications.

[3]  Jian Huang,et al.  A Selective Review of Group Selection in High-Dimensional Models. , 2012, Statistical science : a review journal of the Institute of Mathematical Statistics.

[4]  J. Kanda,et al.  Peripheral Blood versus Bone Marrow from Unrelated Donors: Bone Marrow Allografts Have Improved Long-Term Overall and Graft-versus-Host Disease-Free, Relapse-Free Survival. , 2019, Biology of blood and marrow transplantation : journal of the American Society for Blood and Marrow Transplantation.

[5]  I. Ha,et al.  Comparison of the marginal hazard model and the sub-distribution hazard model for competing risks under an assumed copula , 2019, Statistical methods in medical research.

[6]  Donglin Zeng,et al.  On sparse representation for optimal individualized treatment selection with penalized outcome weighted learning , 2015, Stat.

[7]  Marie Davidian,et al.  Optimal two‐stage dynamic treatment regimes from a classification perspective with censored survival data , 2018, Biometrics.

[8]  Peisong Han,et al.  Multiply Robust Estimation in Regression Analysis With Missing Data , 2014 .

[9]  Wenbin Lu,et al.  On restricted optimal treatment regime estimation for competing risks data. , 2019, Biostatistics.

[10]  Lu Wang,et al.  Estimation with missing data: beyond double robustness , 2013 .

[11]  Mei-Jie Zhang,et al.  Marginal Models for Clustered Time‐to‐Event Data with Competing Risks Using Pseudovalues , 2011, Biometrics.

[12]  Giota Touloumi,et al.  Practical methods for competing risks data: A review , 2012, Statistical methods in medical research.

[13]  Medhat Askar,et al.  Nonpermissive HLA-DPB1 mismatch increases mortality after myeloablative unrelated allogeneic hematopoietic cell transplantation. , 2014, Blood.

[14]  Wenbin Lu,et al.  Variable selection for optimal treatment decision , 2013, Statistical methods in medical research.

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

[16]  A. Wahed,et al.  Estimating the cumulative incidence function of dynamic treatment regimes , 2018 .

[17]  Wenbin Lu,et al.  DOUBLY ROBUST ESTIMATION OF OPTIMAL TREATMENT REGIMES FOR SURVIVAL DATA-WITH APPLICATION TO AN HIV/AIDS STUDY. , 2017, The annals of applied statistics.

[18]  P. Westervelt,et al.  Peripheral-blood stem cells versus bone marrow from unrelated donors. , 2012, The New England journal of medicine.

[19]  Kwang Woo Ahn,et al.  Group and within-group variable selection for competing risks data , 2018, Lifetime data analysis.

[20]  J. Klein,et al.  Generalised linear models for correlated pseudo‐observations, with applications to multi‐state models , 2003 .

[21]  Robert Gray,et al.  A Proportional Hazards Model for the Subdistribution of a Competing Risk , 1999 .

[22]  Chengchun Shi,et al.  HIGH-DIMENSIONAL A-LEARNING FOR OPTIMAL DYNAMIC TREATMENT REGIMES. , 2018, Annals of statistics.

[23]  Wenbin Lu,et al.  Optimal treatment regimes for survival endpoints using a locally-efficient doubly-robust estimator from a classification perspective , 2017, Lifetime data analysis.

[24]  D.,et al.  Regression Models and Life-Tables , 2022 .

[25]  R. Geskus,et al.  A comparison of model selection methods for prediction in the presence of multiply imputed data , 2018, Biometrical journal. Biometrische Zeitschrift.

[26]  Sijian Wang,et al.  Doubly regularized Cox regression for high-dimensional survival data with group structures , 2013 .

[27]  W. Bensinger.  Allogeneic transplantation: peripheral blood vs. bone marrow , 2012, Current opinion in oncology.

[28]  Donald B. Rubin,et al.  Bayesian Inference for Causal Effects: The Role of Randomization , 1978 .

[29]  Min Zhang,et al.  Estimating optimal treatment regimes from a classification perspective , 2012, Stat.

[30]  Cun-Hui Zhang,et al.  A group bridge approach for variable selection , 2009, Biometrika.