Structural neuroimaging biomarkers for obsessive-compulsive disorder in the ENIGMA-OCD consortium: medication matters

ENIGMA-OCD Working Group

doi:10.1038/s41398-020-01013-y

Structural neuroimaging biomarkers for obsessive-compulsive disorder in the ENIGMA-OCD consortium: medication matters

ENIGMA-OCD Working Group

Research output: Contribution to journal › Article › peer-review

58 Scopus citations

Abstract

No diagnostic biomarkers are available for obsessive-compulsive disorder (OCD). Here, we aimed to identify magnetic resonance imaging (MRI) biomarkers for OCD, using 46 data sets with 2304 OCD patients and 2068 healthy controls from the ENIGMA consortium. We performed machine learning analysis of regional measures of cortical thickness, surface area and subcortical volume and tested classification performance using cross-validation. Classification performance for OCD vs. controls using the complete sample with different classifiers and cross-validation strategies was poor. When models were validated on data from other sites, model performance did not exceed chance-level. In contrast, fair classification performance was achieved when patients were grouped according to their medication status. These results indicate that medication use is associated with substantial differences in brain anatomy that are widely distributed, and indicate that clinical heterogeneity contributes to the poor performance of structural MRI as a disease marker.

Original language	English
Article number	342
Journal	Translational Psychiatry
Volume	10
Issue number	1
DOIs	https://doi.org/10.1038/s41398-020-01013-y
State	Published - 1 Dec 2020
Externally published	Yes

Funding

Funders	Funder number
HSM Hochspezialisierte Medizin of the Kanton Zurich
Seventh Framework Programme
EU FP7s
Michael Smith Foundation for Health Research
Netherlands Brain Foundation
Generalitat de Catalunya
Wellcome Trust Centre for Mitochondrial Research
Dana Foundation
Transcept Pharmaceuticals
Netherlands Society for Scientific Research
Eli Lilly and Company
Helse Vest Health Authority
National Institutes of Health
Provincial Health Services Authority
South London and Maudsley NHS Foundation Trust
Harmann Müller foundation
Medical Research Foundation
Young Investigator Grant
The Wellcome Trust DBT India Alliance
National Alliance for Research on Schizophrenia and Depression
Young Investigator Award
Alberta Innovates
Canadian Institutes of Health Research
National Research Foundation
Biohaven Pharmaceuticals
Ontario Brain Institute
Biogen, Inc.
Valera Health
International OCD Foundation
National Natural Science Foundation of China
SA MRC
National Institute of Mental Health	R33MH107589, K23MH115206, R01MH107419, R01MH081864, K23MH092397, K23MH082176, K23-MH09239, R21MH093889, K23MH104515, R01MH085900, R21MH101441, R01MH111794
Wellcome-DBT India Alliance	81371340, 500236/Z/11/Z
National Center for Advancing Translational Sciences	KL2TR000069, UL1TR000067
Wellcome and Royal Society	211155/Z/18/Z
Deutsche Forschungsgemeinschaft	KO 3744/7-1
National Institute on Drug Abuse	R21DA045218, R21DA046029
Ministero della Salute	RC13-14-15-16-17-18A
Japan Society for the Promotion of Science	19K03309, 18K15523, 18K07608, 16K19778, 16K04344, 26461762
Department of Science and Technology, Ministry of Science and Technology, India	SR/S0/HS/0016/2011
Bundesministerium für Bildung und Forschung	BMBF-01GW0724
Helse Vest	911754, 911880
South London and Maudsley Trust	064846
ZonMw	920-03-542, 917.15.318
Department of Biotechnology, Ministry of Science and Technology, India	BT/PR13334/Med/30/259/2009, BT/06/IYBA/2012
NHMRC	1140764
Ministry of Education, Culture, Sports, Science and Technology	16K10253, 25461732, 22591262
Wellcome Trust	211155
Fundació la Marató de TV3	091710, 01/2010
Fundação de Amparo à Pesquisa do Estado de São Paulo	2018/04654-9, 2018/21934-5, 2011/21357-9
Agència de Gestió d'Ajuts Universitaris i de Recerca	2017 SGR 1247
Jacobs Foundation	2017-1261-04
Fundação para a Ciência e a Tecnologia	PDE/BDE/113601/2015
European Regional Development Fund	FI17/00294, PI14/00419, PI16/00950, 2017 SGR 881
Norges Forskningsråd	243675
European Commission	278948
Hersenstichting	2010(1)-50
Shanghai Key Laboratory of Psychotic Disorders	13dz2260500
Hartmann Müller-Stiftung für Medizinische Forschung	1460
National Natural Science Foundation of China-Yunnan Joint Fund	81560233
Japan Agency for Medical Research and Development	JP19dm0307002
Instituto de Salud Carlos III	FI17/00294, CPII16/00048, PI16/00889, PI13/01958, PI14/00419, PI040829
Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung	114076, SNSF 320030_130237
Nederlandse Organisatie voor Wetenschappelijk Onderzoek	940-37-018, 912-02-050, 916.86.038, 907-00-012, 916.86.036
National Institute of Biomedical Imaging and Bioengineering	U54EB020403
Brain and Behavior Research Foundation	27023
Massachusetts General Hospital	K23-MH092397

Access to Document

10.1038/s41398-020-01013-y

Cite this

@article{fb170a80206d4cd29f989ce68127b64e,

title = "Structural neuroimaging biomarkers for obsessive-compulsive disorder in the ENIGMA-OCD consortium: medication matters",

abstract = "No diagnostic biomarkers are available for obsessive-compulsive disorder (OCD). Here, we aimed to identify magnetic resonance imaging (MRI) biomarkers for OCD, using 46 data sets with 2304 OCD patients and 2068 healthy controls from the ENIGMA consortium. We performed machine learning analysis of regional measures of cortical thickness, surface area and subcortical volume and tested classification performance using cross-validation. Classification performance for OCD vs. controls using the complete sample with different classifiers and cross-validation strategies was poor. When models were validated on data from other sites, model performance did not exceed chance-level. In contrast, fair classification performance was achieved when patients were grouped according to their medication status. These results indicate that medication use is associated with substantial differences in brain anatomy that are widely distributed, and indicate that clinical heterogeneity contributes to the poor performance of structural MRI as a disease marker.",

author = "{ENIGMA-OCD Working Group} and Bruin, {Willem B.} and Luke Taylor and Thomas, {Rajat M.} and Shock, {Jonathan P.} and Paul Zhutovsky and Yoshinari Abe and Pino Alonso and Ameis, {Stephanie H.} and Alan Anticevic and Arnold, {Paul D.} and Francesca Assogna and Francesco Benedetti and Beucke, {Jan C.} and Boedhoe, {Premika S.W.} and Irene Bollettini and Anushree Bose and Silvia Brem and Brennan, {Brian P.} and Buitelaar, {Jan K.} and Rosa Calvo and Yuqi Cheng and Cho, {Kang Ik K.} and Sara Dallaspezia and Damiaan Denys and Ely, {Benjamin A.} and Feusner, {Jamie D.} and Fitzgerald, {Kate D.} and Fouche, {Jean Paul} and Fridgeirsson, {Egill A.} and Patricia Gruner and G{\"u}rsel, {Deniz A.} and Hauser, {Tobias U.} and Yoshiyuki Hirano and Hoexter, {Marcelo Q.} and Hao Hu and Chaim Huyser and Iliyan Ivanov and Anthony James and Fern Jaspers-Fayer and Norbert Kathmann and Christian Kaufmann and Kathrin Koch and Masaru Kuno and Gerd Kvale and Kwon, {Jun Soo} and Yanni Liu and Christine Lochner and Luisa L{\'a}zaro and Paulo Marques and Stein, {Dan J.}",

note = "Publisher Copyright: {\textcopyright} 2020, The Author(s).",

year = "2020",

month = dec,

day = "1",

doi = "10.1038/s41398-020-01013-y",

language = "אנגלית",

volume = "10",

journal = "Translational Psychiatry",

issn = "2158-3188",

publisher = "Nature Publishing Group",

number = "1",

}

TY - JOUR

T1 - Structural neuroimaging biomarkers for obsessive-compulsive disorder in the ENIGMA-OCD consortium

T2 - medication matters

AU - ENIGMA-OCD Working Group

AU - Bruin, Willem B.

AU - Taylor, Luke

AU - Thomas, Rajat M.

AU - Shock, Jonathan P.

AU - Zhutovsky, Paul

AU - Abe, Yoshinari

AU - Alonso, Pino

AU - Ameis, Stephanie H.

AU - Anticevic, Alan

AU - Arnold, Paul D.

AU - Assogna, Francesca

AU - Benedetti, Francesco

AU - Beucke, Jan C.

AU - Boedhoe, Premika S.W.

AU - Bollettini, Irene

AU - Bose, Anushree

AU - Brem, Silvia

AU - Brennan, Brian P.

AU - Buitelaar, Jan K.

AU - Calvo, Rosa

AU - Cheng, Yuqi

AU - Cho, Kang Ik K.

AU - Dallaspezia, Sara

AU - Denys, Damiaan

AU - Ely, Benjamin A.

AU - Feusner, Jamie D.

AU - Fitzgerald, Kate D.

AU - Fouche, Jean Paul

AU - Fridgeirsson, Egill A.

AU - Gruner, Patricia

AU - Gürsel, Deniz A.

AU - Hauser, Tobias U.

AU - Hirano, Yoshiyuki

AU - Hoexter, Marcelo Q.

AU - Hu, Hao

AU - Huyser, Chaim

AU - Ivanov, Iliyan

AU - James, Anthony

AU - Jaspers-Fayer, Fern

AU - Kathmann, Norbert

AU - Kaufmann, Christian

AU - Koch, Kathrin

AU - Kuno, Masaru

AU - Kvale, Gerd

AU - Kwon, Jun Soo

AU - Liu, Yanni

AU - Lochner, Christine

AU - Lázaro, Luisa

AU - Marques, Paulo

AU - Stein, Dan J.

PY - 2020/12/1

Y1 - 2020/12/1

N2 - No diagnostic biomarkers are available for obsessive-compulsive disorder (OCD). Here, we aimed to identify magnetic resonance imaging (MRI) biomarkers for OCD, using 46 data sets with 2304 OCD patients and 2068 healthy controls from the ENIGMA consortium. We performed machine learning analysis of regional measures of cortical thickness, surface area and subcortical volume and tested classification performance using cross-validation. Classification performance for OCD vs. controls using the complete sample with different classifiers and cross-validation strategies was poor. When models were validated on data from other sites, model performance did not exceed chance-level. In contrast, fair classification performance was achieved when patients were grouped according to their medication status. These results indicate that medication use is associated with substantial differences in brain anatomy that are widely distributed, and indicate that clinical heterogeneity contributes to the poor performance of structural MRI as a disease marker.

AB - No diagnostic biomarkers are available for obsessive-compulsive disorder (OCD). Here, we aimed to identify magnetic resonance imaging (MRI) biomarkers for OCD, using 46 data sets with 2304 OCD patients and 2068 healthy controls from the ENIGMA consortium. We performed machine learning analysis of regional measures of cortical thickness, surface area and subcortical volume and tested classification performance using cross-validation. Classification performance for OCD vs. controls using the complete sample with different classifiers and cross-validation strategies was poor. When models were validated on data from other sites, model performance did not exceed chance-level. In contrast, fair classification performance was achieved when patients were grouped according to their medication status. These results indicate that medication use is associated with substantial differences in brain anatomy that are widely distributed, and indicate that clinical heterogeneity contributes to the poor performance of structural MRI as a disease marker.

UR - http://www.scopus.com/inward/record.url?scp=85092536771&partnerID=8YFLogxK

U2 - 10.1038/s41398-020-01013-y

DO - 10.1038/s41398-020-01013-y

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

C2 - 33033241

AN - SCOPUS:85092536771

SN - 2158-3188

VL - 10

JO - Translational Psychiatry

JF - Translational Psychiatry

IS - 1

M1 - 342

ER -

Structural neuroimaging biomarkers for obsessive-compulsive disorder in the ENIGMA-OCD consortium: medication matters

Abstract

Funding

Access to Document

Other files and links

Fingerprint

Cite this