gms | German Medical Science

63. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

02. - 06.09.2018, Osnabrück

A comparison of subgroup identification methods in clinical drug development

Meeting Abstract

Search Medline for

  • Cynthia Huber - Universitätsmedizin Göttingen, Göttingen, Deutschland
  • Norbert Benda - Bundesinstitut für Arzneimittel und Medizinprodukte, Bonn, Deutschland
  • Tim Friede - Universitätsmedizin Göttingen, Göttingen, Deutschland

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 63. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Osnabrück, 02.-06.09.2018. Düsseldorf: German Medical Science GMS Publishing House; 2018. DocAbstr. 131

doi: 10.3205/18gmds027, urn:nbn:de:0183-18gmds0277

Published: August 27, 2018

© 2018 Huber et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at



With the advances in genomic sequencing, predictive biomarkers have become a useful tool for the development of personalized medicine. Predictive biomarkers can be used to select subsets of patients, which are most likely to benefit from a treatment. A number of approaches for subgroup identification were proposed over the last years. Although overviews of subgroup identification methods are available [1], [2], systematic comparisons of their performance in simulation studies are rare.

Interaction trees [3], model-based recursive partitioning [4], subgroup identification based on differential effect [5], simultaneous threshold interaction modeling algorithm [6] and adaptive refinement by directed peeling (ARDP) [7] were proposed for subgroup identification. We compared these methods in a Monte Carlo simulation study. All methods, besides ARDP, aim at identifying subgroups with differential treatment effects. ARDP, in contrast, identifies a sequence of nested subgroups with enhanced treatment effects.

In order to identify a target population for subsequent trials a dichotomization of the identified subgroups is needed. Therefore, we propose a subgroup criterion leading to a target subgroup consisting of the identified subgroups with an estimated treatment difference no less than a prespecified threshold. In our simulation study we evaluated these methods by considering measures for binary classification, like sensitivity and specificity. Moreover, we evaluate the Type I error rate, e.g. the proportion of incorrectly selecting a subgroup as target population. In settings with large effects or huge sample sizes most methods perform well. For more realistic settings in drug development involving data from a single trial none of the methods seems suitable for selecting a target population.

The methods and the subgroup criterion are illustrated by an application in amyotrophic lateral sclerosis.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.

This contribution has already been published [8].


Lipkovich I, Dmitrienko A, D’Agostino RB Sr. Tutorial in biostatistics: data-driven subgroup identification and analysis in clinical trials. Statistics in Medicine. 2017;36(1):136-96.
Thomas O, et al. Methods for identification and confirmation of targeted subgroups in clinical trials: A systematic review. Journal of Biopharmaceutical Statistics. 2016;26(1):99-119.
Xiaogang S, et al. Subgroup Analysis via Recursive Partitioning. Journal of Machine Learning Research. 2009;10:141-58.
Seibold H, Zeileis A, Hothorn T. Model-Based Recursive Partitioning for Subgroup Analyses. The International Journal of Biostatistics. 2016;12(1):45-63.
Lipkovich I, et al. Subgroup identification based on differential effect search—A recursive partitioning method for establishing response to treatment in patient subpopulations. Statistics in Medicine. 2011;30(21):2601–21.
Dusseldorp E, Conversano C, Van Os B. Combining an Additive and Tree-Based Regression Model Simultaneously: STIMA. Journal of Computational and Graphical Statistics. 2010;19(3):514-30.
Shilpa P, et al. Identifying back pain subgroups: developing and applying approaches using individual patient data collected within clinical trials. Programme Grants for Applied Research. 2016;4(10):111-34.
Huber C, Benda N, Friede T. A comparison of subgroup identification methods in clinical drug development. 64. Biometrisches Kolloquium an der Goethe-Universität Frankfurt. 2018. External link