Personalized prediction of antidepressant v. placebo response

evidence from the EMBARC study

Christian A. Webb, Madhukar H. Trivedi, Zachary D. Cohen, Daniel G. Dillon, Jay C. Fournier, Franziska Goer, Maurizio Fava, Patrick J. McGrath, Myrna Weissman, Ramin Parsey, Phil Adams, Joseph M. Trombello, Crystal Cooper, Patricia Deldin, Maria A. Oquendo, Melvin G. McInnis, Quentin Huys, Gerard Bruder, Benji T. Kurian, Manish Jha & 2 others Robert J. DeRubeis, Diego A. Pizzagalli

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Background: Major depressive disorder (MDD) is a highly heterogeneous condition in terms of symptom presentation and, likely, underlying pathophysiology. Accordingly, it is possible that only certain individuals with MDD are well-suited to antidepressants. A potentially fruitful approach to parsing this heterogeneity is to focus on promising endophenotypes of depression, such as neuroticism, anhedonia, and cognitive control deficits. Methods: Within an 8-week multisite trial of sertraline v. placebo for depressed adults (n = 216), we examined whether the combination of machine learning with a Personalized Advantage Index (PAI) can generate individualized treatment recommendations on the basis of endophenotype profiles coupled with clinical and demographic characteristics. Results: Five pre-treatment variables moderated treatment response. Higher depression severity and neuroticism, older age, less impairment in cognitive control, and being employed were each associated with better outcomes to sertraline than placebo. Across 1000 iterations of a 10-fold cross-validation, the PAI model predicted that 31% of the sample would exhibit a clinically meaningful advantage [post-treatment Hamilton Rating Scale for Depression (HRSD) difference ⩾3] with sertraline relative to placebo. Although there were no overall outcome differences between treatment groups (d = 0.15), those identified as optimally suited to sertraline at pre-treatment had better week 8 HRSD scores if randomized to sertraline (10.7) than placebo (14.7) (d = 0.58). Conclusions: A subset of MDD patients optimally suited to sertraline can be identified on the basis of pre-treatment characteristics. This model must be tested prospectively before it can be used to inform treatment selection. However, findings demonstrate the potential to improve individual outcomes through algorithm-guided treatment recommendations.

Original languageEnglish (US)
Pages (from-to)1-10
Number of pages10
JournalPsychological Medicine
DOIs
StateAccepted/In press - Jul 2 2018

Fingerprint

Antidepressive Agents
Sertraline
Placebos
Major Depressive Disorder
Depression
Endophenotypes
Therapeutics
Anhedonia
Demography

Keywords

  • Antidepressant
  • depression
  • endophenotype
  • machine learning
  • placebo
  • precision medicine
  • prediction

ASJC Scopus subject areas

  • Applied Psychology
  • Psychiatry and Mental health

Cite this

Webb, C. A., Trivedi, M. H., Cohen, Z. D., Dillon, D. G., Fournier, J. C., Goer, F., ... Pizzagalli, D. A. (Accepted/In press). Personalized prediction of antidepressant v. placebo response: evidence from the EMBARC study. Psychological Medicine, 1-10. https://doi.org/10.1017/S0033291718001708

Personalized prediction of antidepressant v. placebo response : evidence from the EMBARC study. / Webb, Christian A.; Trivedi, Madhukar H.; Cohen, Zachary D.; Dillon, Daniel G.; Fournier, Jay C.; Goer, Franziska; Fava, Maurizio; McGrath, Patrick J.; Weissman, Myrna; Parsey, Ramin; Adams, Phil; Trombello, Joseph M.; Cooper, Crystal; Deldin, Patricia; Oquendo, Maria A.; McInnis, Melvin G.; Huys, Quentin; Bruder, Gerard; Kurian, Benji T.; Jha, Manish; DeRubeis, Robert J.; Pizzagalli, Diego A.

In: Psychological Medicine, 02.07.2018, p. 1-10.

Research output: Contribution to journalArticle

Webb, CA, Trivedi, MH, Cohen, ZD, Dillon, DG, Fournier, JC, Goer, F, Fava, M, McGrath, PJ, Weissman, M, Parsey, R, Adams, P, Trombello, JM, Cooper, C, Deldin, P, Oquendo, MA, McInnis, MG, Huys, Q, Bruder, G, Kurian, BT, Jha, M, DeRubeis, RJ & Pizzagalli, DA 2018, 'Personalized prediction of antidepressant v. placebo response: evidence from the EMBARC study', Psychological Medicine, pp. 1-10. https://doi.org/10.1017/S0033291718001708
Webb, Christian A. ; Trivedi, Madhukar H. ; Cohen, Zachary D. ; Dillon, Daniel G. ; Fournier, Jay C. ; Goer, Franziska ; Fava, Maurizio ; McGrath, Patrick J. ; Weissman, Myrna ; Parsey, Ramin ; Adams, Phil ; Trombello, Joseph M. ; Cooper, Crystal ; Deldin, Patricia ; Oquendo, Maria A. ; McInnis, Melvin G. ; Huys, Quentin ; Bruder, Gerard ; Kurian, Benji T. ; Jha, Manish ; DeRubeis, Robert J. ; Pizzagalli, Diego A. / Personalized prediction of antidepressant v. placebo response : evidence from the EMBARC study. In: Psychological Medicine. 2018 ; pp. 1-10.
@article{db2372cf0a834425b5fef8f1425d7258,
title = "Personalized prediction of antidepressant v. placebo response: evidence from the EMBARC study",
abstract = "Background: Major depressive disorder (MDD) is a highly heterogeneous condition in terms of symptom presentation and, likely, underlying pathophysiology. Accordingly, it is possible that only certain individuals with MDD are well-suited to antidepressants. A potentially fruitful approach to parsing this heterogeneity is to focus on promising endophenotypes of depression, such as neuroticism, anhedonia, and cognitive control deficits. Methods: Within an 8-week multisite trial of sertraline v. placebo for depressed adults (n = 216), we examined whether the combination of machine learning with a Personalized Advantage Index (PAI) can generate individualized treatment recommendations on the basis of endophenotype profiles coupled with clinical and demographic characteristics. Results: Five pre-treatment variables moderated treatment response. Higher depression severity and neuroticism, older age, less impairment in cognitive control, and being employed were each associated with better outcomes to sertraline than placebo. Across 1000 iterations of a 10-fold cross-validation, the PAI model predicted that 31{\%} of the sample would exhibit a clinically meaningful advantage [post-treatment Hamilton Rating Scale for Depression (HRSD) difference ⩾3] with sertraline relative to placebo. Although there were no overall outcome differences between treatment groups (d = 0.15), those identified as optimally suited to sertraline at pre-treatment had better week 8 HRSD scores if randomized to sertraline (10.7) than placebo (14.7) (d = 0.58). Conclusions: A subset of MDD patients optimally suited to sertraline can be identified on the basis of pre-treatment characteristics. This model must be tested prospectively before it can be used to inform treatment selection. However, findings demonstrate the potential to improve individual outcomes through algorithm-guided treatment recommendations.",
keywords = "Antidepressant, depression, endophenotype, machine learning, placebo, precision medicine, prediction",
author = "Webb, {Christian A.} and Trivedi, {Madhukar H.} and Cohen, {Zachary D.} and Dillon, {Daniel G.} and Fournier, {Jay C.} and Franziska Goer and Maurizio Fava and McGrath, {Patrick J.} and Myrna Weissman and Ramin Parsey and Phil Adams and Trombello, {Joseph M.} and Crystal Cooper and Patricia Deldin and Oquendo, {Maria A.} and McInnis, {Melvin G.} and Quentin Huys and Gerard Bruder and Kurian, {Benji T.} and Manish Jha and DeRubeis, {Robert J.} and Pizzagalli, {Diego A.}",
year = "2018",
month = "7",
day = "2",
doi = "10.1017/S0033291718001708",
language = "English (US)",
pages = "1--10",
journal = "Psychological Medicine",
issn = "0033-2917",
publisher = "Cambridge University Press",

}

TY - JOUR

T1 - Personalized prediction of antidepressant v. placebo response

T2 - evidence from the EMBARC study

AU - Webb, Christian A.

AU - Trivedi, Madhukar H.

AU - Cohen, Zachary D.

AU - Dillon, Daniel G.

AU - Fournier, Jay C.

AU - Goer, Franziska

AU - Fava, Maurizio

AU - McGrath, Patrick J.

AU - Weissman, Myrna

AU - Parsey, Ramin

AU - Adams, Phil

AU - Trombello, Joseph M.

AU - Cooper, Crystal

AU - Deldin, Patricia

AU - Oquendo, Maria A.

AU - McInnis, Melvin G.

AU - Huys, Quentin

AU - Bruder, Gerard

AU - Kurian, Benji T.

AU - Jha, Manish

AU - DeRubeis, Robert J.

AU - Pizzagalli, Diego A.

PY - 2018/7/2

Y1 - 2018/7/2

N2 - Background: Major depressive disorder (MDD) is a highly heterogeneous condition in terms of symptom presentation and, likely, underlying pathophysiology. Accordingly, it is possible that only certain individuals with MDD are well-suited to antidepressants. A potentially fruitful approach to parsing this heterogeneity is to focus on promising endophenotypes of depression, such as neuroticism, anhedonia, and cognitive control deficits. Methods: Within an 8-week multisite trial of sertraline v. placebo for depressed adults (n = 216), we examined whether the combination of machine learning with a Personalized Advantage Index (PAI) can generate individualized treatment recommendations on the basis of endophenotype profiles coupled with clinical and demographic characteristics. Results: Five pre-treatment variables moderated treatment response. Higher depression severity and neuroticism, older age, less impairment in cognitive control, and being employed were each associated with better outcomes to sertraline than placebo. Across 1000 iterations of a 10-fold cross-validation, the PAI model predicted that 31% of the sample would exhibit a clinically meaningful advantage [post-treatment Hamilton Rating Scale for Depression (HRSD) difference ⩾3] with sertraline relative to placebo. Although there were no overall outcome differences between treatment groups (d = 0.15), those identified as optimally suited to sertraline at pre-treatment had better week 8 HRSD scores if randomized to sertraline (10.7) than placebo (14.7) (d = 0.58). Conclusions: A subset of MDD patients optimally suited to sertraline can be identified on the basis of pre-treatment characteristics. This model must be tested prospectively before it can be used to inform treatment selection. However, findings demonstrate the potential to improve individual outcomes through algorithm-guided treatment recommendations.

AB - Background: Major depressive disorder (MDD) is a highly heterogeneous condition in terms of symptom presentation and, likely, underlying pathophysiology. Accordingly, it is possible that only certain individuals with MDD are well-suited to antidepressants. A potentially fruitful approach to parsing this heterogeneity is to focus on promising endophenotypes of depression, such as neuroticism, anhedonia, and cognitive control deficits. Methods: Within an 8-week multisite trial of sertraline v. placebo for depressed adults (n = 216), we examined whether the combination of machine learning with a Personalized Advantage Index (PAI) can generate individualized treatment recommendations on the basis of endophenotype profiles coupled with clinical and demographic characteristics. Results: Five pre-treatment variables moderated treatment response. Higher depression severity and neuroticism, older age, less impairment in cognitive control, and being employed were each associated with better outcomes to sertraline than placebo. Across 1000 iterations of a 10-fold cross-validation, the PAI model predicted that 31% of the sample would exhibit a clinically meaningful advantage [post-treatment Hamilton Rating Scale for Depression (HRSD) difference ⩾3] with sertraline relative to placebo. Although there were no overall outcome differences between treatment groups (d = 0.15), those identified as optimally suited to sertraline at pre-treatment had better week 8 HRSD scores if randomized to sertraline (10.7) than placebo (14.7) (d = 0.58). Conclusions: A subset of MDD patients optimally suited to sertraline can be identified on the basis of pre-treatment characteristics. This model must be tested prospectively before it can be used to inform treatment selection. However, findings demonstrate the potential to improve individual outcomes through algorithm-guided treatment recommendations.

KW - Antidepressant

KW - depression

KW - endophenotype

KW - machine learning

KW - placebo

KW - precision medicine

KW - prediction

UR - http://www.scopus.com/inward/record.url?scp=85049356702&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049356702&partnerID=8YFLogxK

U2 - 10.1017/S0033291718001708

DO - 10.1017/S0033291718001708

M3 - Article

SP - 1

EP - 10

JO - Psychological Medicine

JF - Psychological Medicine

SN - 0033-2917

ER -