1
|
Maciukiewicz M, Marshe VS, Hauschild AC, Foster JA, Rotzinger S, Kennedy JL, Kennedy SH, Müller DJ, Geraci J. GWAS-based machine learning approach to predict duloxetine response in major depressive disorder. J Psychiatr Res 2018; 99:62-68. [PMID: 29407288 DOI: 10.1016/j.jpsychires.2017.12.009] [Citation(s) in RCA: 47] [Impact Index Per Article: 7.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/19/2017] [Revised: 10/31/2017] [Accepted: 12/14/2017] [Indexed: 12/22/2022]
Abstract
Major depressive disorder (MDD) is one of the most prevalent psychiatric disorders and is commonly treated with antidepressant drugs. However, large variability is observed in terms of response to antidepressants. Machine learning (ML) models may be useful to predict treatment outcomes. A sample of 186 MDD patients received treatment with duloxetine for up to 8 weeks were categorized as "responders" based on a MADRS change >50% from baseline; or "remitters" based on a MADRS score ≤10 at end point. The initial dataset (N = 186) was randomly divided into training and test sets in a nested 5-fold cross-validation, where 80% was used as a training set and 20% made up five independent test sets. We performed genome-wide logistic regression to identify potentially significant variants related to duloxetine response/remission and extracted the most promising predictors using LASSO regression. Subsequently, classification-regression trees (CRT) and support vector machines (SVM) were applied to construct models, using ten-fold cross-validation. With regards to response, none of the pairs performed significantly better than chance (accuracy p > .1). For remission, SVM achieved moderate performance with an accuracy = 0.52, a sensitivity = 0.58, and a specificity = 0.46, and 0.51 for all coefficients for CRT. The best performing SVM fold was characterized by an accuracy = 0.66 (p = .071), sensitivity = 0.70 and a sensitivity = 0.61. In this study, the potential of using GWAS data to predict duloxetine outcomes was examined using ML models. The models were characterized by a promising sensitivity, but specificity remained moderate at best. The inclusion of additional non-genetic variables to create integrated models may improve prediction.
Collapse
Affiliation(s)
- Malgorzata Maciukiewicz
- Pharmacogenetic Research Clinic, Center for Addiction and Mental Health, Toronto, ON, Canada
| | - Victoria S Marshe
- Pharmacogenetic Research Clinic, Center for Addiction and Mental Health, Toronto, ON, Canada; Institute of Medical Science, Faculty of Medicine, University of Toronto, Toronto, ON, Canada
| | - Anne-Christin Hauschild
- IBM Life Sciences Discovery Centre, Princess Margaret Cancer Centre, Toronto, ON, Canada; Department of Computer Science, University of Toronto, Toronto, ON, Canada; University Health Network, Toronto, ON, Canada
| | - Jane A Foster
- University Health Network, Toronto, ON, Canada; Department of Psychiatry and Behavioral Neurosciences, McMaster University, Hamilton, ON, Canada
| | - Susan Rotzinger
- University Health Network, Toronto, ON, Canada; Department of Psychiatry, University of Toronto, Toronto, ON, Canada
| | - James L Kennedy
- Pharmacogenetic Research Clinic, Center for Addiction and Mental Health, Toronto, ON, Canada; Institute of Medical Science, Faculty of Medicine, University of Toronto, Toronto, ON, Canada; Department of Psychiatry, University of Toronto, Toronto, ON, Canada
| | - Sidney H Kennedy
- Institute of Medical Science, Faculty of Medicine, University of Toronto, Toronto, ON, Canada; University Health Network, Toronto, ON, Canada; Department of Psychiatry, University of Toronto, Toronto, ON, Canada; Department of Psychiatry, St. Michael's Hospital, Toronto, ON, Canada
| | - Daniel J Müller
- Pharmacogenetic Research Clinic, Center for Addiction and Mental Health, Toronto, ON, Canada; Institute of Medical Science, Faculty of Medicine, University of Toronto, Toronto, ON, Canada; Department of Psychiatry, University of Toronto, Toronto, ON, Canada.
| | - Joseph Geraci
- Department of Molecular Medicine, Queen's University, Kingston, ON, Canada
| |
Collapse
|
4
|
Abstract
The wide interindividual variability in clinical response and tolerability of antipsychotic medications has led investigators to postulate that these variabilities may be under genetic control. Although not always consistent, there are promising indications from emergent pharmacogenetic studies that efficacy of antipsychotic medications for the various symptom domains of psychopathology in schizophrenia may be genetically regulated. This is an encouraging approach. Moreover, there are also suggestive findings that the side-effect profiles of second-generation antipsychotic medications and their propensity to cause weight gain and glucose and lipid abnormalities as well as tardive dyskinesia may be related to pharmacogenetic factors in this patient population. Ultimately, such approaches could drive choices of antipsychotic medication based on the likelihood of clinical response and development of side effects in light of a particular patient's genetic profile. In the future, this targeted approach (personalized medicine) may become informative for clinicians choosing an antipsychotic medication for an individual patient with schizophrenia.
Collapse
Affiliation(s)
- Adriana Foster
- Department of Psychiatry and Health Behavior, Medical College of Georgia, 997 St Sebastian, Augusta, GA 30912, USA.
| | | | | |
Collapse
|
5
|
Abstract
Most psychiatric disorders, including major depressive disorder (MDD), are known to involve complex interactions between genetic and environmental influences that impact the development and severity of symptomatology. Health care practitioner competencies have been expanded to include application of genetic knowledge in mental health. Yet this information is difficult to decipher and apply. To assist with these challenges, this article synthesizes recent literature related to the genetics of MDD and illustrates the genetic pathways for major depression.
Collapse
Affiliation(s)
- Betty L Elder
- Wichita State University, School of Nursing, Wichita, Kansas, USA.
| | | |
Collapse
|
6
|
Luyten P, Vliegen N, Van Houdenhove B, Blatt SJ. Equifinality, multifinality, and the rediscovery of the importance of early experiences: pathways from early adversity to psychiatric and (functional) somatic disorders. PSYCHOANALYTIC STUDY OF THE CHILD 2009; 63:27-60. [PMID: 19449788 DOI: 10.1080/00797308.2008.11800798] [Citation(s) in RCA: 14] [Impact Index Per Article: 0.9] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 10/18/2022]
Abstract
Most current mainstream research, diagnostic assessment, and treatment strategies focus on specific psychiatric disorders--on diagnoses that are based on manifest symptoms within a categorical, atheoretical approach. This disorder-centered approach has been antithetical to psychoanalytic views, which are fundamentally person centered, focusing on the dynamics of individual lives. Growing realization of the high comorbidity among psychiatric disorders has led to the need to include developmental considerations and hierarchical models in the classification and treatment of psychopathology. In addition, this realization has led to a renewed interest in the principles of equifinality and multifinality--that a given end state can be the result of different developmental paths and that similar developmental factors may lead to dissimilar outcomes. In this chapter these developments are illustrated by research on the impact of early adversity, a central domain in psychoanalytic thought. Findings from various strands of research in the neurosciences and genetic research, in particular, suggest that early adversity leads to vulnerability for a wide variety of both psychiatric and (functional) somatic disorders. These findings have contributed to the rediscovery of the importance of early experiences more generally and to the need for a broad developmental perspective. In this context, we also discuss the danger of reductionism associated with the growing influence and popularity of affective neuroscience and genetics as well as the vital role a psychoanalytic perspective might play in countering this reductionism by reestablishing the importance of meaning and meaning making in understanding and treating patients with a history of early adversity. In particular, we focus on the importance of narrative and mental representations in the development of the capacity for mentalization in these patients.
Collapse
|
7
|
Abstract
Emergent pharmacogenetic studies indicate that the efficacy of antipsychotic medications in schizophrenia may be predicted through genetic analysis. There also is evidence that the side-effect profiles of second-generation antipsychotic medications and their propensity to cause weight gain, glucose and lipid abnormalities, and tardive dyskinesia may be predicted by pharmacogenetic analysis in this patient population. In the future, this targeted approach with the choice of antipsychotic medication based on the likelihood of clinical response and development of side effects in light of a particular patient's genetic status may gain hold as new treatments are developed with even fewer side effects.
Collapse
|
8
|
Luyten P, Blatt SJ, Van Houdenhove B, Corveleyn J. Depression research and treatment: Are we skating to where the puck is going to be? Clin Psychol Rev 2006; 26:985-99. [PMID: 16473443 DOI: 10.1016/j.cpr.2005.12.003] [Citation(s) in RCA: 77] [Impact Index Per Article: 4.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/15/2005] [Accepted: 12/15/2005] [Indexed: 11/30/2022]
Abstract
This paper critically reviews empirical findings regarding current key assumptions underlying the nature and treatment of depression which heavily rely on the DSM approach. This review shows that empirical evidence provides little support for these assumptions. In response to these findings, an etiologically based, biopsychosocial, dynamic interactionism model of depression is proposed. This model could foster further integration in research on depression and assist in the development of guidelines for the treatment of depression that are better informed by research findings and more congruent with complex clinical realities.
Collapse
Affiliation(s)
- Patrick Luyten
- Department of Psychology, University of Leuven, Tiensestraat 102, 3000 Leuven, Belgium.
| | | | | | | |
Collapse
|
9
|
Abstract
This article gives an overview of genetic research approaches and their application to delusional disorder. Most studies have been based on small samples and have had other methodological limitations, so it is not clear whether there is a genetic contribution to the aetiology of delusional disorder. It is unlikely that delusional disorder is strongly related genetically to affective disorder or schizophrenia, but more subtle relationships cannot be ruled out. The rarity of multiply affected families prohibits linkage studies and, to date, molecular genetic investigations have been mainly limited to small association studies of dopamine receptor polymorphisms. A range of considerably larger, epidemiologically rigorous studies is required, but the uncommonness and other features of the disorder put strong limitations on the prospects for ascertaining adequate samples.
Collapse
Affiliation(s)
- Alastair G Cardno
- Academic Unit of Psychiatry and Behavioural Sciences, School of Medicine, University of Leeds, 15 Hyde Terrace, Leeds LS2 9LT, UK.
| | | |
Collapse
|
10
|
Abstract
There is considerable evidence that genetic factors play a major role in the etiology of unipolar depression. Investigations into vulnerability genes for unipolar depression are underway and for more broadly defined depression-related traits, such as anxiety, neuroticism, and harm avoidance. This review discusses some of the core issues related to study design and molecular genetic methodology, followed by an overview of recent molecular genetic findings for unipolar depression. The research to date has identified regions within certain chromosomes that may contain risk genes. Improved study design and the use of new molecular techniques hold promise for the identification of more specific vulnerability genes for unipolar depression.
Collapse
Affiliation(s)
- Patricia Huezo-Diaz
- MRC Social Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, De Crespigny Park, Denmark Hill, London, UK.
| | | | | |
Collapse
|