Abstract
Subgroup identification has emerged as a popular statistical tool to access the heterogeneity in treatment effects based on specific characteristics of patients. Recently, a threshold linear mixed-effects model was proposed to identify a subgroup of patients who may benefit from treatment concerning longitudinal outcomes based on whether a continuous biomarker exceeds an unknown cut-point. This model assumes, however, normal distributions to both random effects and error terms and, therefore, may be sensitive to outliers in the longitudinal outcomes. In this paper, we propose a robust subgroup identification method for longitudinal data by developing a robust threshold t linear mixed-effects model, where random effects and within-subject errors follow a multivariate t distribution, with unknown degree-of-freedoms. The likelihood function is, however, difficult to directly maximize because the density function of a non-central t distribution is too complicated to compute and an indicator function is included in the definition of the mode. We, therefore, propose a smoothed expectation conditional maximization algorithm based on a gamma-normal hierarchical structure and the smooth approximation of an indicator function to make inferences on the parameters in the model. Simulation studies are conducted to investigate the performance and robustness of the proposed method. As an application, the proposed method is used to identify a subgroup of patients with advanced colorectal cancer who may have a better quality of life when treated by cetuximab.



Similar content being viewed by others
Availability of data and material
Not applicable
Code availability
Not applicable
References
Bai X, Chen K, Yao W (2016) Mixture of linear mixed models using multivariate t distribution. J Stat Comput Simul 86(4):771–787
Berger J, Wang X, Shen L (2014) A bayesian approach to subgroup identification. J Biopharmaceutical Stat 24:110–29
Chen B, Jiang W, Tu D (2014) A hierarchical bayes model for biomarker subset effects in clinical trials. Comput Stat Data Anal 71:324–334
de Alencar FH, Galarza CE, Matos LA, Lachos VH (2020) Censmfm: finite mixture of multivariate censored/missing data. R Package Version 211 http://CRANR-projectorg/package=CensMFM
Efron B, Hinkley DV (1978) Assessing the accuracy of the maximum likelihood estimator: observed versus expected Fisher information. Biometrika 65(3):457–483
Efron B, Tibshirani RJ (1993) An Introduction to the Bootstrap. No. 57 in Monographs on Statistics and Applied Probability, Chapman & Hall/CRC
Foster JC, Taylor JM, Ruberg SJ (2011) Subgroup identification from randomized clinical trial data. Stat Med 30(24):2867–2880
Gavanji P, Chen B, Jiang W (2018) Residual bootstrap test for interactions in biomarker threshold models with survival data. Stat Biosci 10:202–216
Ge X, Peng Y, Tu D (2020) A threshold linear mixed model for identification of treatment-sensitive subsets in a clinical trial based on longitudinal outcomes and a continuous covariate. Stat Methods Med Res 29(10):2919–2931
Hartley HO, Rao JNK (1967) Maximum-likelihood estimation for the mixed analysis of variance model. Biometrika 54(1/2):93–108
He Y, Lin H, Tu D (2018) A single-index threshold cox proportional hazard model for identifying a treatment-sensitive subset based on multiple biomarkers. Stat Med 37(23):3267–3279
Imai K, Ratkovic M (2013) Estimating treatment effect heterogeneity in randomized program evaluation. Ann Appl Stat. https://doi.org/10.1214/12-AOAS593
Jiang W, Freidlin B, Simon R (2007) Biomarker-adaptive threshold design: a procedure for evaluating treatment with possible biomarker-defined subset effect. J Natl Cancer Inst 99:1036–43. https://doi.org/10.1093/jnci/djm022
Jonker DJ, O’Callaghan CJ, Karapetis CS, Zalcberg JR, Tu D, Au HJ, Berry SR, Krahn M, Price T, Simes RJ, Tebbutt NC, van Hazel G, Wierzbicki R, Langer C, Moore MJ (2007) Cetuximab for the treatment of colorectal cancer. New Engl J Med 357(20):2040–2048
Karapetis C, Khambata-Ford S, Jonker D, O’Callaghan C, Tu D, Tebbutt N, Simes J, Chalchal H, Shapiro J, Robitaille S, Price T, Shepherd L, Au HJ, Langer C, Moore M, Zalcberg J (2008) K-ras mutations and benefit from cetuximab in advanced colorectal cancer. New Engl J Med 359:1757–65
Lange KL, Little RJA, Taylor JMG (1989) Robust statistical modeling using the t distribution. J Am Stat Assoc 84(408):881–896
Lipkovich I, Dmitrienko A, Denne J, Enas G (2011) Subgroup identification based on differential effect search (sides)—a recursive partitioning method for establishing response to treatment in patient subpopulations. Stat Med 30:2601–21. https://doi.org/10.1002/sim.4289
Loh WY, Fu H, Man M, Champion V, Yu M (2016) Identification of subgroups with differential treatment effects for longitudinal and multiresponse variables. Stat Med 35(26):4837–4855
Loh WY, Cao L, Zhou P (2019) Subgroup identification for precision medicine: a comparative review of 13 methods. WIREs Data Min Knowld Discov 9(5):e1326
Lu W, Qin G, Zhu Z, Tu D (2021) Multiply robust subgroup identification for longitudinal data with dropouts via median regression. J Multivar Anal. https://doi.org/10.1016/j.jmva.2020.104691
Peel D, McLachlan GJ (2000) Robust mixture modelling using the t distribution. Stat Comput 10(4):339–348
Pinheiro JC, Liu C, Wu YN (2001) Efficient algorithms for robust estimation in linear mixed-effects models using the multivariate t distribution. J Comput Graph Stat 10(2):249–276
Su X, Tsai CL, Wang H, Nickerson D, Li B (2009) Subgroup analysis via recursive partitioning. J Mach Learn Res 10:141–158
Su X, Zhou T, Yan X, Fan J, Yang S (2008) Interaction trees with censored survival data. Int J Biostat 4(1)
Wells C, Tu D, Siu L, Shapiro J, Jonker D, Karapetis C, Simes J, Liu G, Price T, Tebbutt N, O’Callaghan C (2018) Outcomes of older patients (\(>=\)70 years) treated with targeted therapy in metastatic chemorefractory colorectal cancer: A retrospective analysis of ncic ctg co.17 and co.20. Clinical Colorectal Cancer 18
Xing Y, Yu T, Wu Y, Roy M, Kim J, Lee C (2006) An expectation-maximization algorithm for probabilistic reconstructions of full-length isoforms from splice graphs. Nucleic Acids Res 34:3150–60
Xu Y, Yu M, Zhao YQ, Li Q, Wang S, Shao J (2015) Regularized outcome weighted subgroup identification for differential treatment effects: outcome weighted subgroup identification. Biometrics. https://doi.org/10.1111/biom.12322
Funding
This work was partially supported by the National Natural Science Foundation of China (11871164), Shanghai Special Program : Clinical Multidisciplinary Treatment System and Systems Epidemiology Research, Three-year Action Program of Shanghai Municipality for Strengthening the Construction of Public Health System (GWV-10.1-XK05) Big Data and Artificial Intelligence Application, Shanghai Municipal Science and Technology Major Project (ZD2021CY001 ), and Natural Science and Engineering Council of Canada.
Author information
Authors and Affiliations
Contributions
Not applicable
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
(1) Proofs of Eqs. (15) and (17).
Since the conjugate prior follows a gamma distribution and likelihood a normal distribution, it can be shown that the posterior distribution(\(\tau _{i}|Y_{i}\)) follows a gamma distribution. Similarly, the posterior distribution(\(\alpha _{i}|Y_{i}, \tau _{i}\)) can be shown following a normal distribution. Denote f() as a density function. The following are detailed derivations for Eqs. (15) and (17).
Derivation of Eq. (15): Since
where
Therefore,
Derivation of Eq. (17): Since
where
Thus,
(2) See Table 8
(3) See Table 9
Rights and permissions
About this article
Cite this article
Zhang, R., Qin, G. & Tu, D. A robust threshold t linear mixed model for subgroup identification using multivariate T distributions. Comput Stat 38, 299–326 (2023). https://doi.org/10.1007/s00180-022-01229-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-022-01229-0