TITLE: Topics on Oncology
SPEAKER: Professor Ying Lu, Stanford University
Moderator: Naitee Ting


Identification of subgroups with differential treatment effects is increasingly important nowadays when effective therapies for modern diseases (such as cancer) are hard to find while the amount of data grows in leaps and bounds. The good news is that several statistical methods for subgroup identification are available. The bad news, for the practitioner, is that there is little guidance to choose among them. As a result, many in the biopharma industry are doing their own in-house evaluation of the methods.

The aims of the tutorial are (1) to review the motivation, objectives, and limitations of the existing algorithms and (2) to present the results of a comprehensive head-to-head comparison of the methods. The algorithms include:

  1. BLASSO: Bayesian Lasso (Gu et al.,   2013)
  2. FindIt: Finding heterogeneous treatment effects (Imai and Ratkovic, 2013; Egami et al., 2017)
  3. GUIDE: Generalized unbiased interaction detection and estimation (Loh et al., 2015, 2016, 2018)
  4. IT: Interaction trees (Su et al., 2008, 2009, 2011)
  5. MOB: Model-based recursive partitioning (Zeileis et al., 2008; Seibold et al., 2016, 2017)
  6. QUINT: Qualitative interaction trees (Dusseldorp and Van Mechelen, 2014; Dusseldorp et al., 2016a, b)
  7. ROWSi: Regularized outcome weighted subgroup identification (Xu et al., 2015)
  8. SIDES: Subgroup identification based on differential effect search (Lipkovich et al., 2011; Lipkovich and Dmitrienko, 2014)
  9. SubgrpID: Patient subgroup identification for clinical drug development (Chen et al., 2015; Huang et al., 2017)
  10. VT: Virtual twins (Foster et al., 2011; Vieille, 2016)

Real and simulated data are used to compare the algorithms in terms of their probability of false positives (Type I error), power, bias in variable selection (probability of selecting the wrong biomarkers and hence wrong subgroups), bias in estimates of treatment effects, and computational speed. Included in the discussion are conceptual and inferential questions such as: “When there are multiple subgroups, which is the right one?” “Is the sample size large enough?” “Is the subgroup statistically significant?” “Are the treatment effects statistically significant?”



  • Chen, G., Zhong, H., Belousov, A. and Devanarayan, V. (2015) A PRIM approach to predictive-signature development for patient stratification. Statistics in Medicine, 34, 317–342.
  • Dusseldorp, E., Doove, L., van de Put, J. and Van Mechelen, I. (2016a) quint: Qualitative Interaction Trees. R package version 1.2.1.
  • Dusseldorp, E., Doove, L. and Van Mechelen, I. (2016b) Quint: An R package for identification of subgroups of clients who differ in which treatment alternative is best for them. Behavior Research Methods, 48, 650–663.
  • Dusseldorp, E. and Van Mechelen, I. (2014) Qualitative interaction trees: a tool to identify qualitative treatment-subgroup interactions. Statistics in Medicine, 33, 219–237.
  • Egami, N., Ratkovic, M. and Imai, K. (2017) FindIt: Finding Heterogeneous Treatment Effects. R package version 1.1.2.
  • Foster, J. C., Taylor, J. M. G. and Ruberg, S. J. (2011) Subgroup identification from randomized clinical trial data. Statistics in Medicine, 30, 2867–2880.
  • Gu, X., Yin, G. and Lee, J. J. (2013) Bayesian two-step Lasso strategy for biomarker selection in personalized medicine development for time-to-event endpoints. Contemporary Clinical Trials, 36, 642–650.
  • Huang, X., Sun, Y., Trow, P., Chatterjee, S., Chakravarty, A., Tian, L. and Devanarayan, V. (2017) Patient subgroup identification for clinical drug development. Statistics in Medicine, 36, 1414–1428.
  • Imai, K. and Ratkovic, M. (2013) Estimating treatment effect heterogeneity in randomized program evaluation. Annals of Applied Statistics, 7, 443– 470.
  • Lipkovich, I. and Dmitrienko, A. (2014) Strategies for identifying predictive biomarkers and subgroups with enhanced treatment effect in clinical trials using SIDES. Journal of Biopharmaceutical Statistics, 24, 130–153.
  • Lipkovich, I., Dmitrienko, A., Denne, J. and Enas, G. (2011) Subgroup identification based on differential effect search—a recursive partitioning method for establishing response to treatment in patient subpopulations. Statistics in Medicine, 30, 2601–2621.
  • Loh, W.-Y., Fu, H., Man, M., Champion, V. and Yu, M. (2016) Identification of subgroups with differential treatment effects for longitudinal and multiresponse variables. Statistics in Medicine, 35, 4837–4855.
  • Loh, W.-Y., He, X. and Man, M. (2015) A regression tree approach to identifying subgroups with differential treatment effects. Statistics in Medicine, 34, 1818–1833.
  • Loh, W.-Y., Man, M. and Wang, S. (2018) Subgroups from regression trees with adjustment for prognostic effects and post-selection inference. Statistics in Medicine. To appear.
  • Seibold, H., Zeileis, A. and Hothorn, T. (2016) Model-based recursive partitioning for subgroup analyses. International Journal of Biostatistics, 12, 45–63.
    • (2017) Individual treatment effect prediction for amyotrophic lateral sclerosis patients. Statistical Methods in Medical Research. DOI: 10.1177/0962280217693034.
  • Su, X., Meneses, K., McNees, P. and Johnson, W. O. (2011) Interaction trees: exploring the differential effects of an intervention programme for breast cancer survivors. Journal of the Royal Statistical Society, Ser. C, 60, 457–474.
  • Su, X., Tsai, C. L., Wang, H., Nickerson, D. M. and Bogong, L. (2009) Subgroup analysis via recursive partitioning. Journal of Machine Learning Research, 10, 141–158.
  • Su, X., Zhou, T., Yan, X., Fan, J. and Yang, S. (2008) Interaction trees with censored survival data. International Journal of Biostatistics, 4.
  • Vieille, F. (2016) aVirtualTwins: Adaptation of Virtual Twins Method from Jared Foster. R package version 1.0.0.
  • Xu, Y., Yu, M., Zhao, Y.-Q., Li, Q., Wang, S. and Shao, J. (2015) Regularized outcome weighted subgroup identification for differential treatment effects. Biometrics, 71, 645–653.
  • Zeileis, A., Hothorn, T. and Hornik, K. (2008) Model-based recursive partitioning. Journal of Computational and Graphical Statistics, 17, 492–514.


This entry was posted in . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *