×

Optimal treatment regimes for competing risk data using doubly robust outcome weighted learning with bi-level variable selection. (English) Zbl 07345440

Summary: 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.

MSC:

62-XX Statistics

Software:

Rglpk; cmprsk
PDF BibTeX XML Cite
Full Text: DOI

References:

[1] Ahn, K. W.; Banerjee, A.; Sahr, N.; Kim, S., Group and within-group variable selection for competing risks data, Lifetime Data Anal., 24, 3, 407-424 (2018) · Zbl 1468.62370
[2] Alousi, A.; Wang, T.; Hemmer, M. T.; Spellman, S. R.; Arora, M.; Couriel, D. R.; Pidala, J.; Anderlini, P.; Boyiadzis, M.; Bredeson, C. N., 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, Biol. Blood Marrow Transplant., 25, 2, 270-278 (2019)
[3] Anasetti, C.; Logan, B. R.; Lee, S. J., Peripheral-blood stem cells versus bone marrow from unrelated donors, New Engl. J. Med., 367, 16, 1487-1496 (2012)
[4] Andersen, P. K.; Klein, J. P.; Rosthøj, S., Generalised linear models for correlated pseudo-observations, with applications to multi-state models, Biometrika, 90, 1, 15-27 (2003) · Zbl 1034.62062
[5] Bai, X.; Tsiatis, A. A.; Lu, W.; Song, R., Optimal treatment regimes for survival endpoints using a locally-efficient doubly-robust estimator from a classification perspective, Lifetime Data Anal., 23, 4, 585-604 (2017) · Zbl 1468.62372
[6] Bakoyannis, G.; Touloumi, G., Practical methods for competing risks data: a review, Stat. Methods Med. Res., 21, 3, 257-272 (2012) · Zbl 1242.62116
[7] Bensinger, W. I., Allogeneic transplantation: peripheral blood versus bone marrow, Curr. Opinion Oncol., 24, 2, 191 (2012)
[8] Cox, D. R., Regression models and life-tables, J. R. Stat. Soc. Ser. B Stat. Methodol., 34, 2, 187-202 (1972)
[9] Emura, T.; Shih, J.-H.; Ha, I. D.; Wilke, R. A., Comparison of the marginal hazard model and the sub-distribution hazard model for competing risks under an assumed copula, Stat. Methods Med. Res., 29, 8, 2307-2327 (2020)
[10] Fine, J. P.; Gray, R. J., A proportional hazards model for the subdistribution of a competing risk, J. Amer. Statist. Assoc., 94, 446, 496-509 (1999) · Zbl 0999.62077
[11] Hager, R.; Tsiatis, A. A.; Davidian, M., Optimal two-stage dynamic treatment regimes from a classification perspective with censored survival data, Biometrics, 74, 4, 1180-1192 (2018)
[12] Han, P., Multiply robust estimation in regression analysis with missing data, J. Amer. Statist. Assoc., 109, 507, 1159-1173 (2014) · Zbl 1368.62279
[13] Han, P.; Wang, L., Estimation with missing data: beyond double robustness, Biometrika, 100, 2, 417-430 (2013) · Zbl 1284.62260
[14] He, P.; Eriksson, F.; Scheike, T. H.; Zhang, M.-J., A proportional hazards regression model for the subdistribution with covariates-adjusted censoring weight for competing risks data, Scand. J. Stat., 43, 1, 103-122 (2016) · Zbl 1364.62239
[15] Huang, J.; Breheny, P.; Ma, S., A selective review of group selection in high-dimensional models, Statist. Sci., 27, 4, 481-499 (2012) · Zbl 1331.62347
[16] Huang, J.; Ma, S.; Xie, H.; Zhang, C. H., A group bridge approach for variable selection, Biometrika, 96, 2, 339-355 (2009) · Zbl 1163.62050
[17] Jiang, R.; Lu, W.; Song, R.; Hudgens, M. G.; Naprvavnik, S., Doubly robust estimation of optimal treatment regimes for survival data-with application to an hiv/aids study, Ann. Appl. Stat., 11, 3, 1763-1786 (2017) · Zbl 1380.62255
[18] Logan, B. R.; Zhang, M.-J.; Klein, J. P., Marginal models for clustered time-to-event data with competing risks using pseudovalues, Biometrics, 67, 1, 1-7 (2011) · Zbl 1216.62175
[19] Lu, W.; Zhang, H. H.; Zeng, D., Variable selection for optimal treatment decision, Stat. Methods Med. Res., 22, 5, 493-504 (2013)
[20] Pidala, J.; Lee, S. J.; Ahn, K. W., Nonpermissive HLA-DPB1 mismatch increases mortality after myeloablative unrelated allogeneic hematopoietic cell transplantation, Blood, 21, 16, 2596 (2014)
[21] Rubin, D., Bayesian inference for causal effects: the role of randomization, Ann. Statist., 6, 34-58 (1978) · Zbl 0383.62021
[22] Shi, C.; Fan, A.; Song, R.; Lu, W., High-dimensional A-learning for optimal dynamic treatment regimes, Ann. Statist., 46, 3, 925-957 (2018) · Zbl 1398.62029
[23] Song, R.; Kosorok, M.; Zeng, D.; Zhao, Y.; Laber, E.; Yuan, M., On sparse representation for optimal individualized treatment selection with penalized outcome weighted learning, Stat, 4, 1, 59-68 (2015)
[24] Thao, L. T.P.; Geskus, R., A comparison of model selection methods for prediction in the presence of multiply imputed data, Biom. J., 61, 2, 343-356 (2019) · Zbl 1419.62456
[25] Theussl, S.; Hornik, K., Rglpk: R/GNU linear programming kit interface (2019), URL: https://CRAN.R-project.org/package=Rglpk. R package version 0.6-4
[26] Tsiatis, A., A nonidentifiability aspect of the problem of competing risks, Proc. Natl. Acad. Sci., 72, 1, 20-22 (1975) · Zbl 0299.62066
[27] Tsiatis, A. A., Dynamic Treatment Regimes Statistical Methods for Precision Medicine (2019), Chapman & Hall/CRC
[28] Wang, S.; Wu, T. T., Doubly regularized Cox regression for high-dimensional survival data with group structures, Stat. Interface, 6, 2, 175-186 (2013) · Zbl 1327.62423
[29] Yavuz, I.; Chng, Y.; Wahed, A. S., Estimating the cumulative incidence function of dynamic treatment regimes, J. Roy. Statist. Soc. Ser. A, 181, 1, 85-106 (2018)
[30] Zhang, B.; Tsiatis, A. A.; Davidian, M.; Zhang, M.; Laber, E., Estimating optimal treatment regimes from a classification perspective, Stat, 1, 103-114 (2012)
[31] Zhao, Y. Q.; Zeng, D.; Laber, E. B.; Song, R.; Yuan, M.; Kosorok, M. R., Doubly robust learning for estimating individualized treatment with censored data, Biometrika, 102, 1, 151-168 (2015) · Zbl 1345.62092
[32] Zhou, J.; Zhang, J.; Lu, W.; Li, X., On restricted optimal treatment regime estimation for competing risks data, Biostatistics, 0, 1, 1-16 (2019)
[33] Zou, H., The adaptive lasso and its oracle properties, J. Amer. Statist. Assoc., 101, 476, 1418-1429 (2006) · Zbl 1171.62326
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.