Jue Hou1, Chamindi Seneviratne2, Xiaogang Su3, Jeremy Taylor4, Bankole Johnson2, Xin-Qun Wang5, Heping Zhang6, Henry R Kranzler7, Joseph Kang8, Lei Liu8. 1. Department of Statistics, University of Illinois, Urbana, Illinois. 2. Department of Psychiatry, Institute for Genome Sciences, University of Maryland, Baltimore, Maryland. 3. Department of Mathematical Sciences, University of Texas at El Paso (UTEP), El Paso, Texas. 4. Department of Biostatistics, University of Michigan, Ann Arbor, Michigan. 5. Department of Public Health Sciences, University of Virginia, Charlottesville, Virginia. 6. Department of Biostatistics, Yale University, New Haven, Connecticut. 7. Department of Psychiatry, University of Pennsylvania, Philadelphia, Pennsylvania. 8. Department of Preventive Medicine, Northwestern University, Chicago, Illinois.
Abstract
BACKGROUND: Identification of patient subgroups to enhance treatment effects is an important topic in personalized (or tailored) alcohol treatment. Recently, several recursive partitioning methods have been proposed to identify subgroups benefiting from treatment. These novel data mining methods help to address the limitations of traditional regression-based methods that focus on interactions. METHODS: We propose an exploratory approach, using recursive partitioning methods, for example, interaction trees (IT) and virtual twins (VT), to flexibly identify subgroups in which the treatment effect is likely to be large. We apply these tree-based methods to a pharmacogenetic trial of ondansetron. RESULTS: Our methods identified several subgroups based on patients' genetic and other prognostic covariates. Among the 251 subjects with complete genotype information, the IT method identified 118 with specific genetic and other prognostic factors, resulting in a 17.2% decrease in the percentage of heavy drinking days (PHDD). The VT method identified 88 subjects with a 21.8% decrease in PHDD. Overall, the VT subgroup achieved a good balance between the treatment effect and the group size. CONCLUSIONS: A data mining approach is proposed as a valid exploratory method to identify a sufficiently large subgroup of subjects that is likely to receive benefit from treatment in an alcohol dependence pharmacotherapy trial. Our results provide new insights into the heterogeneous nature of alcohol dependence and could help clinicians to tailor treatment to the biological profile of individual patients, thereby achieving better treatment outcomes.
RCT Entities:
BACKGROUND: Identification of patient subgroups to enhance treatment effects is an important topic in personalized (or tailored) alcohol treatment. Recently, several recursive partitioning methods have been proposed to identify subgroups benefiting from treatment. These novel data mining methods help to address the limitations of traditional regression-based methods that focus on interactions. METHODS: We propose an exploratory approach, using recursive partitioning methods, for example, interaction trees (IT) and virtual twins (VT), to flexibly identify subgroups in which the treatment effect is likely to be large. We apply these tree-based methods to a pharmacogenetic trial of ondansetron. RESULTS: Our methods identified several subgroups based on patients' genetic and other prognostic covariates. Among the 251 subjects with complete genotype information, the IT method identified 118 with specific genetic and other prognostic factors, resulting in a 17.2% decrease in the percentage of heavy drinking days (PHDD). The VT method identified 88 subjects with a 21.8% decrease in PHDD. Overall, the VT subgroup achieved a good balance between the treatment effect and the group size. CONCLUSIONS: A data mining approach is proposed as a valid exploratory method to identify a sufficiently large subgroup of subjects that is likely to receive benefit from treatment in an alcohol dependence pharmacotherapy trial. Our results provide new insights into the heterogeneous nature of alcohol dependence and could help clinicians to tailor treatment to the biological profile of individual patients, thereby achieving better treatment outcomes.
Authors: Raymond F Anton; Stephanie S O'Malley; Domenic A Ciraulo; Ron A Cisler; David Couper; Dennis M Donovan; David R Gastfriend; James D Hosking; Bankole A Johnson; Joseph S LoCastro; Richard Longabaugh; Barbara J Mason; Margaret E Mattson; William R Miller; Helen M Pettinati; Carrie L Randall; Robert Swift; Roger D Weiss; Lauren D Williams; Allen Zweben Journal: JAMA Date: 2006-05-03 Impact factor: 56.272
Authors: Heping Zhang; Richard S Legro; Jeffrey Zhang; Leon Zhang; Xiang Chen; Hao Huang; Peter R Casson; William D Schlaff; Michael P Diamond; Stephen A Krawetz; Christos Coutifaris; Robert G Brzyski; Gregory M Christman; Nanette Santoro; Esther Eisenberg Journal: Hum Reprod Date: 2010-08-17 Impact factor: 6.918
Authors: Bankole A Johnson; Nassima Ait-Daoud; Chamindi Seneviratne; John D Roache; Martin A Javors; Xin-Qun Wang; Lei Liu; J Kim Penberthy; Carlo C DiClemente; Ming D Li Journal: Am J Psychiatry Date: 2011-01-19 Impact factor: 18.112
Authors: Bankole A Johnson; Norman Rosenthal; Julie A Capece; Frank Wiegand; Lian Mao; Karen Beyers; Amy McKay; Nassima Ait-Daoud; Raymond F Anton; Domenic A Ciraulo; Henry R Kranzler; Karl Mann; Stephanie S O'Malley; Robert M Swift Journal: JAMA Date: 2007-10-10 Impact factor: 56.272
Authors: David J Hinton; Marely Santiago Vázquez; Jennifer R Geske; Mario J Hitschfeld; Ada M C Ho; Victor M Karpyak; Joanna M Biernacka; Doo-Sup Choi Journal: Sci Rep Date: 2017-05-31 Impact factor: 4.379
Authors: Kim C M Bul; Lisa L Doove; Ingmar H A Franken; Saskia Van der Oord; Pamela M Kato; Athanasios Maras Journal: PLoS One Date: 2018-03-15 Impact factor: 3.240