Precision medicine is a medical paradigm that focuses on making effective treatment decision based on individual patient characteristics. When there are a large amount of patient information, such as patient’s genetic information, medical records and clinical measurements, available, it is of interest to select the covariates which have interactions with the treatment, for example, in determining the individualized treatment regime where only a subset of covariates with treatment interactions involves in decision making. We propose a marginal feature ranking and screening procedure for measuring interactions between the treatment and covariates. The method does not require imposing a specific model structure on the regression model and is applicable in a high dimensional setting. Theoretical properties in terms of consistency in ranking and selection are established. We demonstrate the finite sample performance of the proposed method by simulation and illustrate the applications with two real data examples from clinical trials.
SongRKosorokMZengD, et al. On sparse representation for optimal individualized treatment selection with penalized outcome weighted learning. Stat2015; 4: 59–68.
2.
QiZLiuDFuH, et al. Multi-armed angle-based direct learning for estimating optimal individualized treatment rules with various outcomes. J Am Stat Assoc2020; 115: 678–691.
TianLAlizadehAGentlesA, et al. A simple method for estimation interactions between a treatment and a large number of covariates. J Am Stat Assoc2014; 109: 1517–1532.
15.
SiriwardhanaCDattaSKulasekeraKB. Selection of the optimal personalized treatment from multiple treatments with multivariate outcome measures. J Biopharm Stat2020; 30: 462–480.
16.
QiZLiuY. D-learning to estimate optimal individual treatment rules. Electron J Stat2018; 12: 1601–1638.
17.
TibshiraniR. Regression shrinkage and selection via the Lasso. J R Stat Soc Series B1996; 58: 267–288.
18.
FanJLiR. Variable selection via nonconcave penalized likelihood and its oracle properties. J Am Stat Assoc2001; 96: 1348–1360.
19.
CandesETaoT. The dantzig selector: statistical estimation when is much larger than . Ann Stat2007; 35: 2313–2351.
20.
ZouHHastieT. Regularization and variable selection via the elastic net. J R Stat Soc Ser B2005; 67: 301–320.
21.
ZouH. The adaptive Lasso and its oracle properties. J Am Stat Assoc2006; 101: 1418–1429.
GeorgeEIMcCullochRE. Variable selection via Gibbs sampling. J Am Stat Assoc1993; 88: 881–889.
24.
HolmesCHeldL. Bayesian auxiliary variable models for binary and multinomial regression. Bayesian Anal2006; 1: 145–168.
25.
BrownPJVannucciMFearnT. Multivariate Bayesian variable selection and prediction. J R Stat Soc Ser B1998; 60: 627–641.
26.
BottoloLRichardsonS. Evolutionary stochastic search for Bayesian model exploration. Bayesian Anal2010; 11: 583–618.
27.
BottoloLBanterleMRichardsonS, et al. A computationally efficient Bayesian seemingly unrelated regressions model for high-dimensional quantitative trait loci discovery. J R Stat Soc: Ser C2021; 70: 886–908.
28.
ZhaoZBanterleMLewinA, et al. Structured Bayesian variable selection for multiple correlated response variables and high-dimensional predictors. arXiv 2021; 2021: 2101.05899.
GunterLZhuJMurphyS. Variable selection for qualitative interactions. Stat Meth2011; 8: 42–55.
31.
FosterJCTaylorJMRubergSJ. Subgroup identification from randomized clinical trial data. Stat Med2011; 30: 2867–2880.
32.
LipkovichIDmitrienkoADenneJ, et al. Subgroup identification based on differential effect search—a recursive partitioning method for establishing response to treatment in patient subpopulations. Stat Med2011; 30: 2601–2621.
33.
LipkovichIDmitrienkoAPatraK, et al. Subgroup identification in clinical trials by stochastic sidescreen methods. Stat Biopharm Res2017; 9: 368–378.
34.
ChenWGhoshDRaghunathanTE, et al. On Bayesian methods of exploring qualitative interactions for targeted treatment. Stat Med2012; 31: 3693–3707.
35.
LuWZhangHHZengD. Variable selection for optimal treatment decision. Stat Meth Med Res2013; 22: 493–504.
36.
LohWYHeXManM. A regression tree approach to identifying subgroups with differential treatment effects. Stat Med2015; 34: 1818–1833.
37.
SechidisKPapangelouKMetcalfePD, et al. Distinguishing prognostic and predictive biomarkers: an information theoretic approach. Bioinformatics2018; 34: 3365–3376.
38.
ZhuLLiLLiR, et al. Model-free feature screening for ultrahigh-dimensional data. J Am Stat Assoc2011; 106: 1464–1475.
39.
Puig-PeiroRMasonAMestre-FerrandizJ, et al. Relative effectiveness in breast cancer treatment: a health production approach. Int J Technol Assess Health Care2015; 31: 371–379.
40.
ItalianoA. Prognostic or predictive? it’s time to get back to definitions!. J Clin Oncol2011; 29: 4718.
41.
FanJLvJ. Sure independence screening for ultrahigh dimensional feature space (with discussion). J R Stat Soc Ser B2008; 70: 849–911.
42.
CarrollRJFanJGijbelsI, et al. Generalized partially linear single-index models. J Am Stat Assoc1997; 92: 477–489.
43.
FanJSongR. Sure independence screening in generalized linear models with NP-dimensionality. Ann Stat2010; 38: 3567–3604.
44.
SiriwardhanaCZhaoMDattaS, et al. A probability based method for selecting the optimal personalized treatment from multiple treatments. Stat Meth Med Res2017; 28: 749–760.
45.
LiKC. Sliced inverse regression for dimension reduction. J Am Stat Assoc1991; 86: 316–327.
46.
FangKKotzSNgKW. Symmetric Multivariate and Related Distributions. New York: Springer, 1990.
47.
HallPLiKC. On almost linearity of low dimensional projections from high dimensional data. Ann Stat1993; 21: 867–889.
48.
CaiTLiuWLuoX. A constrained minimization approach to sparse precision matrix estimation. J Am Stat Assoc2011; 106: 594–607.
49.
WangH. Forward regression for ultra-high dimensional variable screening. J Am Stat Assoc2009; 104: 1512–1524.
HammerSMKatzensteinDAHughesMD, et al. A trial comparing nucleoside monotherapy with combination therapy in HIV-infected adults with CD4 cell counts from 200 to 500 per cubic millimeter. New Eng J Med1996; 335: 1081–1090.
52.
FanJMaYDaiW. Nonparametric independence screening in sparse ultra-high-dimensional varying coefficient models. J Am Stat Assoc2014; 109: 1270–1284.
53.
ChampionVSkinnerCSHuiS, et al. The effect of telephone versus print tailoring for mammography adherence. Patient Educ Couns2007; 65: 416–423.
WangXDaiJY. TwoPhaseInd: an R package for estimating gene-treatment interactions and discovering predictive markers in randomized clinical trials. Bioinformatics2016; 32: 3348–3350.
56.
DaiJYZhangXCWangCY, et al. Augmented case-only designs for randomized clinical trials with failure time endpoints. Biometrics2016; 72: 30–38.
57.
WangJPatelAWasonJ, et al. Two-stage penalized regression screening to detect biomarker-treatment interactions in randomized clinical trials. Biometrics2022; 78: 141–150.
58.
VerdaguerHSaurìTMacarullaT. Predictive and prognostic biomarkers in personalized gastrointestinal cancer treatment. J Gastrointest Oncol2017; 8: 405–417.
59.
LeeJWangL. Prognostic and predictive molecular biomarkers in chronic lymphocytic leukemia. J Mol Diagn2020; 22: 1114–1125.
60.
SerflingRJ. Approximation Theorems of Mathematical Statistics. New York: Wiley, 1980.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
