UBC Faculty Research and Publications

Discovering biomarkers for antidepressant response: protocol from the Canadian biomarker integration… Lam, Raymond W; Milev, Roumen; Rotzinger, Susan; Andreazza, Ana C; Blier, Pierre; Brenner, Colleen; Daskalakis, Zafiris J; Dharsee, Moyez; Downar, Jonathan; Evans, Kenneth R; Farzan, Faranak; Foster, Jane A; Frey, Benicio N; Geraci, Joseph; Giacobbe, Peter; Feilotter, Harriet E; Hall, Geoffrey B; Harkness, Kate L; Hassel, Stefanie; Ismail, Zahinoor; Leri, Francesco; Liotti, Mario; MacQueen, Glenda M; McAndrews, Mary P; Minuzzi, Luciano; Müller, Daniel J; Parikh, Sagar V; Placenza, Franca M; Quilty, Lena C; Ravindran, Arun V; Salomons, Tim V; Soares, Claudio N; Strother, Stephen C; Turecki, Gustavo; Vaccarino, Anthony L; Vila-Rodriguez, Fidel; Kennedy, Sidney H Apr 16, 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12888_2016_Article_785.pdf [ 829.28kB ]
Metadata
JSON: 52383-1.0308586.json
JSON-LD: 52383-1.0308586-ld.json
RDF/XML (Pretty): 52383-1.0308586-rdf.xml
RDF/JSON: 52383-1.0308586-rdf.json
Turtle: 52383-1.0308586-turtle.txt
N-Triples: 52383-1.0308586-rdf-ntriples.txt
Original Record: 52383-1.0308586-source.json
Full Text
52383-1.0308586-fulltext.txt
Citation
52383-1.0308586.ris

Full Text

STUDY PROTOCOL Open AccessDiscovering biomarkers for antidepressantresponse: protocol from the Canadianbiomarker integration network indepression (CAN-BIND) and clinicalcharacteristics of the first patient cohortRaymond W. Lam1, Roumen Milev2, Susan Rotzinger3,4, Ana C. Andreazza4,5, Pierre Blier6, Colleen Brenner7,Zafiris J. Daskalakis4,5, Moyez Dharsee8, Jonathan Downar3,4, Kenneth R. Evans8,9, Faranak Farzan4,5,Jane A. Foster3,10, Benicio N. Frey10, Joseph Geraci3, Peter Giacobbe3,4, Harriet E. Feilotter8,9, Geoffrey B. Hall10,Kate L. Harkness11, Stefanie Hassel12, Zahinoor Ismail13, Francesco Leri14, Mario Liotti15, Glenda M. MacQueen13,Mary Pat McAndrews3, Luciano Minuzzi10, Daniel J. Müller4,5, Sagar V. Parikh16, Franca M. Placenza3,Lena C. Quilty4,5, Arun V. Ravindran4,5, Tim V. Salomons17, Claudio N. Soares18, Stephen C. Strother19,Gustavo Turecki20,21, Anthony L. Vaccarino8, Fidel Vila-Rodriguez1, Sidney H. Kennedy3,4,18* and on behalf of theCAN-BIND Investigator TeamAbstractBackground: Major Depressive Disorder (MDD) is among the most prevalent and disabling medical conditionsworldwide. Identification of clinical and biological markers (“biomarkers”) of treatment response could personalizeclinical decisions and lead to better outcomes. This paper describes the aims, design, and methods of a discoverystudy of biomarkers in antidepressant treatment response, conducted by the Canadian Biomarker IntegrationNetwork in Depression (CAN-BIND). The CAN-BIND research program investigates and identifies biomarkers thathelp to predict outcomes in patients with MDD treated with antidepressant medication. The primary objective ofthis initial study (known as CAN-BIND-1) is to identify individual and integrated neuroimaging, electrophysiological,molecular, and clinical predictors of response to sequential antidepressant monotherapy and adjunctive therapy inMDD.(Continued on next page)* Correspondence: sidney.kennedy@uhn.caRaymond W. Lam, Roumen Milev and Susan Rotzinger are Co-first authors.3University Health Network, 399 Bathurst Street, Toronto, ON M5T 2S8,Canada4Department of Psychiatry, University of Toronto, 250 College Street, 8thfloor, Toronto, ON M5T 1R8, CanadaFull list of author information is available at the end of the article© 2016 Lam et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Lam et al. BMC Psychiatry  (2016) 16:105 DOI 10.1186/s12888-016-0785-x(Continued from previous page)Methods: CAN-BIND-1 is a multisite initiative involving 6 academic health centres working collaboratively withother universities and research centres. In the 16-week protocol, patients with MDD are treated with a first-lineantidepressant (escitalopram 10–20 mg/d) that, if clinically warranted after eight weeks, is augmented with anevidence-based, add-on medication (aripiprazole 2–10 mg/d). Comprehensive datasets are obtained using clinicalrating scales; behavioural, dimensional, and functioning/quality of life measures; neurocognitive testing; genomic,genetic, and proteomic profiling from blood samples; combined structural and functional magnetic resonanceimaging; and electroencephalography. De-identified data from all sites are aggregated within a secure neuroinformaticsplatform for data integration, management, storage, and analyses. Statistical analyses will include multivariate andmachine-learning techniques to identify predictors, moderators, and mediators of treatment response.Discussion: From June 2013 to February 2015, a cohort of 134 participants (85 outpatients with MDD and 49 healthyparticipants) has been evaluated at baseline. The clinical characteristics of this cohort are similar to other studies of MDD.Recruitment at all sites is ongoing to a target sample of 290 participants. CAN-BIND will identify biomarkers of treatmentresponse in MDD through extensive clinical, molecular, and imaging assessments, in order to improve treatment practiceand clinical outcomes. It will also create an innovative, robust platform and database for future research.Trial registration: ClinicalTrials.gov identifier NCT01655706. Registered July 27, 2012.BackgroundDepressive disorders, including Major DepressiveDisorder (MDD), are highly prevalent and disablingconditions with substantial personal and societalcosts [1]. MDD is now the second leading cause ofdisability worldwide [2] and contributes to excessmortality associated with many comorbid medicalconditions [3]. Treatment of depressive disorders isbased on empirical data and evidence-based guide-lines, but treatment selection remains more of an artthan a science. Hence, the discovery of clinical andbiological markers, or biomarkers, of treatment re-sponse that would inform an individualized approachto depression treatment remains a research goal [4].A major challenge to identify predictors (baselinecharacteristics that predict response), moderators (base-line characteristics that predict differential response to aspecific treatment) and mediators (events or changes oc-curring during treatment that explains the response) isthat MDD is a complex, heterogeneous condition. A var-iety of neurobiological and environmental influences,both independently and in combination with one an-other, can alter the clinical expression of MDD in termsof symptoms, severity, episode duration, response totreatment, and functional outcomes. As a result, nosingle intervention is effective for all people with de-pression. Current diagnostic systems such as theDSM-5 [5] can reliably codify depressive symptoms ascriteria for MDD, but these symptoms are not uniqueto depression and, even if clustered together, may notrepresent a specific underlying disease process or atreatment substrate. Hence, the clinical entity termed“Major Depressive Disorder” represents only the final,external manifestations of an enormously complex,multi-level, multi-factorial process.The Canadian Biomarker Integration Network in De-pression (CAN-BIND) [6] was created with an aim touse an integrated approach to biomarker discovery.CAN-BIND draws on multidisciplinary expertise frominvestigative teams at 8 Canadian universities, all in ac-tive collaboration with the Ontario Brain Institute (OBI)[7] and Indoc Research (Toronto, ON, Canada). TheCanadian Institutes of Health Research, academic insti-tutions, and various industry partners provide additionalfunding and support (see Acknowledgements).The overall goal of CAN-BIND is to identify predictors,moderators, and mediators of treatment response andnon-response in people with MDD to guide clinicaldecision-making. CAN-BIND-1 uses an integrated clinical,neuroimaging and molecular approach with high-di-mensional mathematical modeling techniques to specific-ally search for (a) baseline predictors and moderators ofantidepressant and adjunctive agent response, (b) earlytreatment mediators of response (changes from baselineto 2 weeks), and (c) later treatment mediators of response(changes from baseline to 8 weeks, and to 16 weeks).MethodsOverview of protocolPatients with MDD are treated with open-label escitalo-pram 10–20 mg/d for 8 weeks. Responders (≥50 % re-duction in Montgomery-Åsberg Depression Rating Scale[MADRS] score) continue on escitalopram for another8 weeks, while non-responders have aripiprazole 2–10 mg/d added on to escitalopram for 8 weeks. Clinical,neuroimaging and molecular assessments are conductedat Baseline (Week 0) and Weeks 2 and 8; clinical and mo-lecular assessments also are conducted at Weeks 4, 10,and 16; and additional brief clinical evaluations are com-pleted at Weeks 6, 12, and 14. Clinical characterizationLam et al. BMC Psychiatry  (2016) 16:105 Page 2 of 13assesses a broad palette of symptoms, functional out-comes, cognitive performance, personality dimensions,and recent and past life events.ParticipantsParticipants are recruited at 6 clinical centres: Vancouver(Djavad Mowafaghian Centre for Brain Health), Calgary(Hotchkiss Brain Institute), Toronto (2 sites: UniversityHealth Network and Centre for Addiction and MentalHealth), Hamilton (St. Joseph’s Healthcare Hamilton),and Kingston (Providence Care, Mental Health Services).Research Ethics Boards at each site approved the study.Recruitment draws upon outpatient-referral networks,community-based advertising, and dedicated knowledgetranslation (KT) activities.Table 1 lists the inclusion and exclusion criteria forthe patients with MDD. Healthy comparison participantsare 18–60 years of age, with no psychiatric or unstablemedical diagnosis, and sufficient fluency in English tocomplete study procedures. They are matched to the pa-tient group by sex and age distribution.ProcedureAt the Screening Visit, eligible participants provide writ-ten, informed consent for all study procedures. Each pa-tient undergoes screening evaluations that include a fullpsychiatric consultation to confirm a diagnosis of MDDusing the Mini International Neuropsychiatric Interview(MINI) [8]. Previous medication history (type, dose, andduration) is collected using the Antidepressant Treat-ment History Form (ATHF) [9]. A detailed medical his-tory and listing of concomitant treatments, if any, arerecorded. A reproductive/menstrual history is obtainedfrom female patients. Patients undergo medical work-upthat includes physical examination, height and bodyweight measurements, clinical laboratory assays, and 12-lead electrocardiography (if indicated). Social back-ground is collected using standardized reporting formsfor demographic characteristics, handedness, ethnicity,education (expressed as years of formal schooling), mari-tal status, occupational status, job classification, andhousehold income. Participants are screened for contra-indications to magnetic resonance imaging (MRI).Figure 1 shows an overview of the protocol. At theBaseline Visit (Visit 1, Week 0), extensive clinical assess-ments are conducted, and blood and urine samples areobtained for molecular analysis. Participants alsoundergo the first of 3 sessions of structural and func-tional neuroimaging and electroencephalography (EEG),as described below. All patients start treatment withescitalopram and receive standardized clinical manage-ment based on the CANMAT clinical guidelines [10].Neuroimaging/EEG, molecular and clinical assessmentsare conducted again at Weeks 2 and 8. Additional bloodsamples are collected at Week 4 for pharmacogenetic ana-lysis, and at Weeks 2, 10, and 16 for medication levels.Blood chemistry screening, urinalysis, and body weightmeasurements are repeated at Week 16. At the conclusionof the study, patients are discharged into standard clinicalcare by a family physician and/or regular psychiatrist. Pa-tients can also elect to enrol in a long-term naturalisticfollow-up study that includes wellness monitoring usingelectronic mental health (e-Mental Health) tools.Healthy comparison participants attend 5 study visits:Screening, Baseline, Week 2, Week 8, and Week 16.They complete the same assessments as patients but donot receive any treatment.TreatmentsIn Phase 1, patients receive flexibly dosed escitalopram(10–20 mg/d) for 8 weeks. Patients are started at 10 mg/dTable 1 Inclusion and exclusion criteria for patientsInclusion Criteria• Outpatients 18 to 60 years of age.• DSM-IV-TR criteria for MDE in MDD, as confirmed by the MINI.• Depressive episode duration ≥3 months.• Free of psychotropic medications for at least 5 half-lives (i.e. 1 weekfor most antidepressants, 5 weeks for fluoxetine) before baseline.• Score ≥24 on the MADRS.• Fluent in English, sufficient to complete the interviews andself-report questionnaires.Exclusion Criteria• Diagnosis of Bipolar I or Bipolar II disorder.• Any other psychiatric diagnosis that is considered the primary diagnosis.• Any significant personality disorder diagnosis (e.g., borderline,antisocial) that might interfere with participation in the protocol,defined by clinician judgment.• High suicidal risk, defined by clinician judgment.• Substance dependence/abuse in the past 6 months.• Significant neurological disorders, head trauma, or other unstablemedical conditions.• Pregnant or breastfeeding.• Psychosis in the current episode.• High risk for hypomanic switch (i.e., history of antidepressant-induced hypomania).• Failed 4 or more adequate pharmacologic interventions(as determined by the ATHF).• Previously failed or showed intolerance to escitalopram oraripiprazole.• Started psychological treatment within the past 3 months with theintent of continuing treatment.• Contraindications to MRI.DSM-IV-TR, Diagnostic and Statistical Manual of Mental Disorders, FourthEdition, Text Revision; MDE, Major Depressive Episode; MDD, Major DepressiveDisorder; MINI, Mini International Neuropsychiatric Interview; MADRS,Montgomery-Åsberg Depression Rating Scale; ATHF, Antidepressant TreatmentHistory FormLam et al. BMC Psychiatry  (2016) 16:105 Page 3 of 13and increased to 20 mg/d at Week 2 if they do not achieve≥20 % reduction in MADRS score from baseline, and atWeek 4 for those who do not achieve ≥50 % reduction inMADRS.In Phase 2, patients who achieve ≥50 % reduction inbaseline MADRS score at Week 8 are considered “re-sponders” and continue on their effective dose of escita-lopram for a further 8 weeks. Patients who do notachieve ≥50 % reduction in MADRS are considered“non-responders” and receive flexibly dosed aripiprazole(2–10 mg/d), added to escitalopram, for a further8 weeks. Dose increases of aripiprazole are recom-mended if patients do not achieve ≥50 % reduction inMADRS after 2 or 4 weeks.This standardized algorithm reflects usual clinicalpractice, is consistent with evidence-based treatmentguidelines [11–13], and promotes consistency across allsites. Doses can be decreased at the discretion of thetreating psychiatrist if patients do not tolerate higherdoses. Treatment is open-label, and no randomizationprocedures are used. Medication adherence is monitoredwith pill counts at each visit.Concomitant non-psychotropic medications for stableconditions are allowed at the discretion of the studypsychiatrist, who can also permit use of vitamins, supple-ments, oral contraceptives, and non-prescription analge-sics. Patients on pre-existing stable doses are allowed tocontinue on zopiclone up to 7.5 mg prn or lorazepam1–2 mg prn, to a maximum of 3 doses/week.AssessmentsClinical platformThe clinical assessments (Table 2) are selected based ontheoretical and clinical utility and to minimize respondentburden as much as possible. Raters received training, andinter-rater reliability was established using recorded inter-views. Clinician-rated symptom and clinical measures in-clude the Montgomery-Asberg Depression Rating Scale(MADRS), using the structured interview guide (SIGMA)to enhance reliability [14, 15], and Clinical Global Impres-sion, Severity and Improvement scales (CGI-S and CGI-I)[16]. Depressive and other associated symptoms are ex-plored in greater detail using the Depression InventoryDevelopment (DID) semi-structured interview, as part ofthe International Society for CNS Drug Development’s ini-tiative to refine and validate a new measurement tool foruse in clinical trials of MDD [17]. Patient-rated symptomscales include the Quick Inventory for Depressive Symp-tomatology (QIDS-SR) [18] and the Generalized AnxietyDisorder (GAD-7) scale [19]. Manic and hypomanicsymptoms are assessed with the clinician-rated YoungMania Rating Scale (YMRS) [20] and the patient-ratedHypomania Check-List (HCL-32) [21].Self-rated scales are used to assess functional impairment(Sheehan Disability Scale; SDS) [22] and occupational func-tioning (Lam Employment Absence and Productivity Scale;LEAPS) [23]. Quality of life is measured both generally(World Health Organization Quality of Life Assessment;WHOQoL-BREF) [24] and with a specific focus on depres-sion (Quality of Life, Enjoyment and Satisfaction Question-naire; Q-LES-Q) [25], using these 2 self-rated measures.We also assessed other behavioural and dimensionalconstructs, including aversive and incentive motivation(Behavioural Inhibition System/Behavioural ActivationSystem; BIS/BAS) [26], anhedonia (Dimensional AnhedoniaRating Scale; DARS; and Snaith-Hamilton Pleasure Scale;SHAPS) [27, 28], personality (NEO Five-Factor Inventory;NEO-FFI) [29] and pain (Brief Pain Inventory; BPI-SF) [30].Fig. 1 CAN-BIND-1 Clinical ProtocolLam et al. BMC Psychiatry  (2016) 16:105 Page 4 of 13Sleep, circadian rhythms, and seasonality are assessed usingthe Pittsburgh Sleep Quality Index (PSQI) [31], the BiologicalRhythms Interview of Assessment in Neuropsychiatry(BRIAN) [32], and the Seasonal Pattern Assessment Ques-tionnaire (SPAQ) [33]. Physical activity is measured usingthe self-rated International Physical Activity Questionnaire(IPAQ) [34], while dietary habits are assessed through aseries of brief questions.Assessment of environmental stressors includes adult pat-terns of attachment and recent stressful life events that arecaptured with self-report scales, including the Experience inClose Relationships (ECR-R) [35] scale, and the List ofThreatening Experiences (LTE) [36], respectively.Two clinical interviews are conducted by centralized,trained raters at Queen’s University and facilitated by Medeo,a secure telehealth/videohealth service (Medeo, Vancouver,BC, Canada). The Childhood Experience of Care and Abuse(CECA) [37], administered at Week 4 to patients and atWeek 2 to healthy participants, is a semi-structured inter-view that carefully assesses childhood maltreatment. The LifeEvents and Difficulties Schedule (LEDS) [38], administeredat Week 16 to all participants, is a semi-structured context-ual interview and rating system that assesses stressful lifeevents that have occurred within 6 months of depression on-set. Interviews are audiotaped and transcribed so that re-sponses can be coded based on standard, manualizedconventions.CNS Vital Signs (CNS VS) is a computerized test batteryused to assess memory, reaction time and psychomotorspeed, complex attention, and cognitive flexibility [39]. Thereliability of CNS VS is similar to that of conventional neuro-psychological tests [40]. The CNS VS has a robust normativedatabase and is sensitive to common causes of cognitive im-pairment. Analyses of cognitive data will be augmented withan estimate of premorbid intelligence obtained using the Na-tional Adult Reading Test (NART) [41].Medication side effects are documented with theclinician-rated Toronto Side Effects Scale (TSES) [42],which records the incidence, frequency, and severity ofcommon adverse events, and the Sexual FunctioningQuestionnaire (SexFX), which evaluates specific sexualfunctioning [43]. Adverse Events are classified as mild,moderate or severe.Neuroimaging platformfMRI Structural and functional neuroimaging data are ob-tained on 3.0 Tesla (3 T) magnetic resonance imaging (MRI)systems using multicoll phased-array head coils. Among the6 clinical sites, 4 different models of scanners are used, thusmandating an extensive and ongoing standardization andquality control process to ensure that data are comparableand usable [44]. The four models of scanners include Discov-ery MR750 3.0T (GE Healthcare, Little Chalfont, Bucking-hamshire, UK), Signa HDxt 3.0T (GE Healthcare, LittleChalfont, Buckinghamshire, UK), MAGNETOM TrioTim(Siemens Healthcare, Erlangen, Germany), and Achieva 3.0T(Philips Healthcare, Best, Netherlands). Cross-site T1 pilotingincluded a “human phantom”, who travelled to each site foranatomical scans, and a manganese chloride (MnCl2)-basedphantom model for progressive quantitative assessment ofhydration based on signal intensity linearity characteristics[45]. Since the study’s launch, each site has obtained monthlyscans of 2 geometric phantoms (a spherical agar phantomdeveloped by the Function Bioinformatics Research Net-work, and a custom-built cylindrical model using plasticLEGO blocks) [46, 47] to facilitate scanner calibration andtroubleshooting over the long term.Table 2 Clinical characterization assessmentsClinician-Administered AssessmentsMontgomery Asberg Depression Rating Scale (MADRS)Young Mania Rating Scale (YMRS)Clinical Global Impression (CGI)Depression Inventory Development (DID) InterviewToronto Side Effects Scale (TSES)Sexual Side Effects Questionnaire (SexFX)Childhood Experience of Care and Abuse (CECA)Life Events and Difficulties Schedule (LEDS)CNS Vital Signs (CNS-VS) computerized neuropsychological test batteryNational Adult Reading Test (NART)Self-Report AssessmentsQuick Inventory of Depressive Symptomatology, Self-Report (QIDS-SR)Generalized Anxiety Disorder 7-item scale (GAD-7)Hypomania Check-List (HCL-32)Brief Pain Inventory (BPI)Sheehan Disability Scale (SDS)Lam Employment Absence and Productivity Scale (LEAPS)Quality of Life, Enjoyment and Satisfaction Questionnaire (Q-LES-Q)World Health Organization Quality of Life Assessment (WHOQoL-BREF)Pittsburgh Sleep Quality Index (PSQI)Seasonal Pattern Assessment Questionnaire (SPAQ)Biological Rhythm Interview of Assessment in Neuropsychiatry (BRIAN)Dimensional Anhedonia Rating Scale (DARS)Snaith-Hamilton Pleasure Scale (SHAPS)Behavioural Inhibition/Behavioural Activation System (BIS/BAS)NEO Five-Factor Inventory (NEO-FFI)Experiences in Close Relationships (ECR-R) questionnaireList of Threatening Experiences (LTE)International Physical Activity Questionnaire (IPAQ)Brief Diet QuestionnaireLam et al. BMC Psychiatry  (2016) 16:105 Page 5 of 13Following surveys and localization, the examinationprotocol includes a whole-brain T1-weighted turbo gradi-ent echo sequence (9 min) at 1 mm3 resolution, repetitiontime (TR) = 6.2–1900 ms, echo time (TE) = 2.7–3.5 ms, flipangle = 8–15°, inversion time (TI) = 0–1100 ms, field ofview (FOV) = 220–256 mm, matrix 2562–5122, 170–180contiguous slices at 1 mm thickness. A small Vitamin Ecapsule is placed as a stereotactic marker at the right tem-ple to confirm subject orientation during image review [48].Secondly, a whole-brain diffusion tensor imaging (DTI)series is obtained using 30 gradient directions at 2 b-values(500 and 1000 s/mm2) (5 min) with an additional 3 im-ages at b = 0 s/mm2 for tensor construction at a finalvoxel resolution of 2.3 x 2.3 x 5 mm (1 min). Func-tional MRI includes a 10-min resting state scan witheyes open using a fixation cross [49], obtained usinga whole-brain T2*-sensitive blood oxygenation level-dependent (BOLD) echo planar imaging (EPI) series.At the inception of the study, a BOLD EPI series was usedduring the Emotional Face Categorization/Conflict Task (2runs of 7 min each) [50, 51]. After enrolling a cohort of 107patients and 52 healthy participants, we substituted a rewardparadigm and an implicit go no-go task. These tasks allowfor the assessment of reward networks and attentional biasesfor affectively laden stimuli [52]. The substitution, ratherthan addition, of the functional tasks was necessary to main-tain a feasible total time for participants in the scanner. Thenew tasks involve a 12-min hedonic function task [53, 54],where participants receive feedback and earn small monetaryrewards while choosing correct responses among sets of vis-ual stimuli and a 10-min affective go/no-go task to a seriesof stimuli that contain emotional content [55]. Stimulussizes, instructions to participants, and support materials arestandardized across sites. All behavioural data are capturedusing E-Prime software version 2.0 or higher (PsychologySoftware Tools, Sharpsburg, PA, USA).Electroencephalography The fMRI tasks were trans-lated into analogous versions for electroencephalography(EEG), which is carried out at 4 sites, again using severalequipment models (Biosemi ActiveTwo [BioSemi, Amsterdam,Netherlands], BrainVision Recorder/QuickAmp [Brain Prod-ucts, Munich, Germany], Compumedics NeuroScan [Compu-medics USA, Charlotte, NC, USA] , EGI Geodesic [ElectricalGeodesics, Eugene, OR, USA]) that require cross-site consult-ation and standardization. All sites used a minimum of 64-channel caps and conductive electrode gels, with 53 commonelectrodes identified across the different systems. EEGs are digi-tized continuously (bandpass 0.04–100 Hz with 1000 Hz sam-pling rate) and all electrodes are referenced to the vertex(Cz) electrode.At each session, participants complete a 10-min fixationresting state sequence with eyes open, followed by a 10-minresting state sequence with eyes closed. As with fMRI, theEmotional Face Categorization/Conflict Task was used forEEG when the study began, but was later changed to a taskquerying anhedonia in MDD and an affective go/no-go task,also using E-Prime software. Ocular artifacts, such as blinks,saccades, and lateral movements, are removed by independ-ent component analysis (ICA) as implemented in EEGLAB[56] and performed using MATLAB (MathWorks, Natick,MA, USA).Molecular platformBlood and urine samples for genomic and proteomic analysesare collected from all participants at Weeks 0, 2, 8 and 16.Following blood draws at Weeks 0, 2, and 8, total RNA is im-mediately isolated from leukocytes and stabilized using theLeukoLOCK filter apparatus (Thermo Fisher Scientific, Wal-tham, MA, USA), which also depletes globin mRNA to im-prove the utility of samples for expression profiling and otherapplications. For patients with MDD, additional blood sam-ples are collected at Week 4 for pharmacogenetic analysisand at Weeks 2, 10, and 16 for drug levels.Standard Operating Procedures (SOPs) for sample re-ceipt and accessioning, nucleic acid extraction, quality as-sessment, data tracking, DNA and RNA preparation forassays, scanning, and feature extraction are in place.Objectives of the molecular platform are to investigate1) candidate biomarkers for disease state or drug responsein baseline samples in patients and healthy participants, 2)global DNA alterations that may correlate with diseasestate or drug response using baseline samples, and 3) dy-namic molecules in pre- and post- treatment samples toidentify pathways of drug response and select additionaltargets for investigation by targeted methods.De-identified specimens are transferred to and storedat the Douglas Mental Health University Institute biore-pository in accordance with regulatory guidelines andbest practices for biobanking. A subset of samples istransferred to other CAN-BIND sites for specific ana-lyses. Material transfer agreements between sites wereestablished at the outset of the study.Targeted analyses include DNA single-nucleotide poly-morphism (SNP) open-array analysis to identify and se-quence variants that correlate with disease state or drugresponse [57], along with studies using establishedmethods for profiling of mRNA and miRNA sequences,histone modifications, and methylation status across thegenome [58]. Proteomic analysis by selected-reactionmonitoring mass spectrometry (SRM-MS) with state-of-the-art hybrid triple quadrupole/linear ion trap liquid-chromatography mass spectrometry (5500 and 4000QTRAPs), will be used for relative quantification ofhigh-interest plasma proteins within biological pathwayswith purported relationships to MDD [59].Exploratory analyses include RNA-Seq for miRNA fromblood and from plasma using either a HiSeq (Illumina,Lam et al. BMC Psychiatry  (2016) 16:105 Page 6 of 13San Diego, CA, USA) or Proton (Thermo Fisher Scientific,Waltham, MA, USA) platform. DNA oxidative damage toguanine will be evaluated by measuring levels of 8-hydroxy-2-deoxyguanosine (8-OhdG) using a competitive ELISA ana-lysis kit (StressMarq BioSciences, Victoria, BC, Canada).Cytosine oxidation will be measured using an ELISA-basedassay to assess the levels of 5-hydroxymethylcytosine(MethylFlash™Methylated DNA Quantification Kit; Epi-gentek Group, Farmingdale, NY, USA). Global DNAmethylation (i.e., 5-methylcytosine) will be evaluatedusing an ELISA-based method (Sigma-Aldrich, Darm-stadt, Germany). Inflammatory markers in blood will bemeasured using standard antibody-based immunoassays.Data managementEach site has entered a standardized Participation Agree-ment with OBI to facilitate transfer of both raw andprocessed/de-identified data, in accordance with OBI’s Gov-ernance Policy and with any specific conditions required byeach institution’s local legislative and/or ethical policies.De-identified electronic data from all sites are aggregatedfor data analyses using multiple bioinformatics approaches.The OBI’s Centre for Ontario Data Exploration (“Brain-CODE”, https://www.braincode.ca/) is an online neuroin-formatics platform that allows researchers to collaborateacross distances and work more efficiently, and ultimately,to promote new discoveries to improve patient care. Brain-CODE provides the rare capability of supporting scientificinquiry and analytics across multiple brain diseases andmodalities by integrating clinical, imaging, pathology, andgenomics data. Standard variable definitions and formatsare used so that investigators collect data consistentlyacross studies and modalities. This will reduce variability indata collection and facilitate comparisons across diseases,merging of data sets and meta-analyses.In storing sensitive patient information, Brain-CODEemploys sophisticated security systems and utilizes Priv-acy by Design (PbD) to protect privacy by embedding itinto the design specifications of technologies, businesspractices, and physical infrastructure [7]. The OBI alsoworks with several research ethics boards and related or-ganizations to streamline the project review and ap-proval process; and develops opportunities to link orintegrate Brain-CODE with other provincial, national, orinternational databases to augment its analytical power.With this unique complement of capabilities, Brain-CODE is the first database of its kind in the world.Brain-CODE is deployed at the High Performance Com-puting Virtual Laboratory (HPCVL) data centre atQueen’s University in Kingston, Ontario. The HPCVL is aCompute Canada high-performance computing consor-tium that supports regulatory-compliant (e.g., ICH E6, 21CFR Part 11, HIPAA, PIPEDA) processes for securingprivacy of healthcare data [60]. All data are collected,processed, maintained, and stored in Canada.Online data/images can be accessed only on securewebsites via restricted portals requiring unique user-names and passwords for each member of the studyteam. User profiles are assigned only to study personnelrequiring access to enter/verify data, and credentials foreach user are vetted by the program manager.Specific data-collection platforms with Brain-CODEinclude:1) Brain-CODE Subject Registry, a secure portal usedby study centres in Ontario to link encryptedpersonal health information from study participantsto provincial health/administrative databases;2) OpenClinica Enterprise, a regulatory-compliant,web-based electronic data capture (EDC) systemand database for demographic and historical data,diagnostic information, and clinician-rated assessmentsand scales (OpenClinica, Waltham, MA, USA);3) LimeSurvey e-PRO Questionnaires, an open-sourcesurvey tool used by participants for direct entry ofthe 20 self-reported measures using a laptop ortablet while attending clinic visits (LimeSurvey,Hamburg, Germany);4) SPReD (originally the Stroke Patient RecoveryResearch Database), a comprehensive onlinerepository powered by the open-source ExtensibleNeuroimaging Archiving Toolkit (XNAT) imaginginformatics platform [61]. Structural and functionalMRI data are first converted to DICOM (Digital Im-aging and Communications in Medicine) format ateach site prior to uploading [62], whereas EEG dataare uploaded as raw files for subsequentstandardization to 58 channels and conversion into auniversally readable format in EEGLAB. Supplementaryresults, such as behavioural and physiologicaldata, and session notes, are uploaded through aspecial sub-process. All files must be correctlylabeled in accordance with SPReD’s organizationalstructure and naming conventions; an automatedpipeline scans for errors on a daily basis; and5) BASE (BioArray Software Environment) andLabKey, open-source laboratory-informationmanagement systems that enable tracking andmanagement of proteomics and genomicsworkflows, experiments, and raw data [63, 64].Clinical data support is provided by Indoc Research forOpenClinica and LimeSurvey, and by Rotman Research In-stitute/Baycrest Health Sciences for SPReD. For clinical andself-rated measures, source data verification is completedusing a remote, risk-based monitoring system that includescentralized processes for cleaning and extracting data.Lam et al. BMC Psychiatry  (2016) 16:105 Page 7 of 13Future CAN-BIND projects (see below) will use secure,customizable Research Electronic Data Capture (REDCap)tools for all clinical and self-reported data [65].MRI data undergoes manual quality control (QC) pro-cedures where trained raters check for visual artifactsand then grade scans as usable or rejected; in the lattercase, rescanning is recommended if possible within thestudy timeline. For fMRI data, preprocessing includesconversion from DICOM to NIfTI (NeuroimagingInformatics Technology Initiative) format [66] priorto identifying the volume with the least motion, towhich the remaining volumes are registered.Converted EEG data are resampled to 512 Hz withchannels re-referenced to Cz before digital filters are ap-plied (high pass = 0.5 Hz, low pass = 100 Hz) and triggercodes are standardized.For a period following study closure, CAN-BIND dataare protected by OBI for the exclusive use of the investi-gative team and its collaborators. However, in the future,de-identified CAN-BIND data may be shared by OBIwith other collaborators and third parties for researchpurposes. These datasets could be made available toother clinical research teams with similar datasets forcomparison of treatment outcomes in other psychiatricconditions (such as dementia, schizophrenia, and bipolardisorder). Eligible third parties will be recognized re-searchers or organizations who have submitted detailedstudy plans and ethics boards approvals to the OBI. Be-fore OBI discloses the de-identified data to any thirdparty, it will enter into an agreement with the latter toprotect confidentiality and ensure correct usage of data.AnalysesWe will examine several outcomes, including respondersat Week 8 who maintained response at Week 16, re-sponders at Week 8 who are non-responders at Week16, non-responders at Week 8 who respond at Week 16with augmentation, and non-responders at Weeks 8 and16, among other outcomes. A reasonably large dataset isrequired to study adequately the complex clinical, neu-roimaging, and molecular factors that contribute totreatment outcomes. Our sample size target for thisphase of the CAN-BIND program is 200 patients and 90healthy participants (290 in total).To identify standalone features that differ significantly be-tween responders and non-responders, either at baseline orover the course of the study, parametric two-sample t-testswill be used for features with normally distributed data.Otherwise, the non-parametric Mann–Whitney U test willbe used. A significance threshold of α = 0.05 and multipletesting corrections will be required. Given an input list ofup to 25,000 features, of which at least 250 are differentiallyexpressed between responders and non-responders, assum-ing an overall false discovery rate of 5 %, and applying two-sample t-testing procedures, a minimum of 49 subjects percategory are required for power of at least 90 % to correctlyidentify a given feature which differs significantly betweenthe categories. If nonparametric testing procedures areused, a minimum of 52 subjects per category are requiredto achieve the same power. Target recruitment for patientswith MDD is 200. An estimated response rate of 60 %would yield 120 responders and 80 non-responders. Theplanned sample size is sufficiently powered for both para-metric and non-parametric univariate testing. Multivariateanalysis and the development of prognostic signatures ofescitalopram response will follow univariate analysis.Data analyses will be conducted using a suite of com-mercial and open-source software tools installed onhigh-end workstations at HPCVL. While data from eachanalytical platform will be independently analyzed usingthe methods described above, important relationships(either correlational or causal) between modalities canbe detected through an integration of data collectedacross assessment platforms for a given study subject.To manage the integrated analyses of this complex dataset, we have established a Data Science Advisory Teamconsisting of the Principal Investigators, domain-specificexperts (e.g., molecular or imaging team members), in-formatics advisors, and operations support. This team willoversee processes for data cleaning, preprocessing, inte-gration, quality control and overall analytics plans for eachof the core research activities. Cross-domain as well asdomain-specific data interrogation will be conducted byWorking Groups responsible for applying specific analyticapproaches according to their expertise and contributingto a master Statistical Analysis Plan. Conventional univari-ate and multivariate statistical approaches will addresshypothesis-driven investigations, while a variety of analyt-ics pipelines with both supervised and unsupervised tech-niques will allow exploratory analyses. Machine-learningapproaches, including principle component analysis, ran-dom forests, and support vector machines will be used toidentify subgroups of patients with shared symptoms and/or biological features that could have clinical relevance.Progress to dateBetween 2013 and 2015, 159 participants (107 patients and52 healthy participants) were screened and 134 (85 patientsand 49 healthy participants) entered at baseline in the ini-tial protocol with the Emotional Face Categorization/Con-flict Task during MRI and EEG sessions, as noted above.Full neuroimaging data (to Week 8) were obtained from 71patients and 45 healthy participants.Table 3 lists the clinical characteristics of this first cohort(N= 134) that were evaluable at baseline. Independent-samples t-tests showed no significant demographic differ-ences between patients and healthy participants. The pa-tients were moderately depressed at baseline, as reflectedLam et al. BMC Psychiatry  (2016) 16:105 Page 8 of 13by MADRS, QIDS-SR, and CGI-S mean scores, as well asmoderately impaired as assessed by psychosocial (SDS) andwork (LEAPS) functioning scales.DiscussionThe CAN-BIND study is a unique biomarker-discovery ini-tiative in its mandate, scope, organizational structure, andintegrative approach. First, the study collects systematicoutcome data in the clinical, neuroimaging, and moleculardomains in an agnostic fashion, with no preconceivedidentification of biomarkers for depression. Second, dataare aggregated and integrated using a secure informaticsplatform that facilitates collaborative data sharing andrigorous quality-control processes across sites. Third,cross-domain and domain-specific analyses are undertakenusing high-dimensional mathematical models supple-mented with conventional statistical tools.While CAN-BIND-1 focuses on pharmacotherapy,subsequent protocols will address biomarkers in treatmentresponse using other evidence-based interventions forMDD, including transcranial magnetic stimulation, cogni-tive behaviour therapy, cognitive remediation, and others.Additional studies using CAN-BIND platforms to examineintegrated biomarkers involve youth at risk for seriousmental illness, suicide risk, and effects of childhood mal-treatment on stress sensitivity and reward responsivity.CAN-BIND also operates a dedicated reverse-translationTable 3 Clinical characteristics of the first cohort (N = 134)Characteristic Patients withMDD (N = 85)Healthy participants(N= 49)Female:Male, N (%) 50:35 (59 %:41 %) 32:17 (65 %:35 %)Age in years, mean (SD), range 36.1 (12.5), 19–61 32.5 (10.2), 20–57Ethnicitya, N (%)Aboriginal 0 0Arab 3 (4 %) 1 (2 %)Asian 9 (11 %) 9 (18 %)Black 1 (1 %) 0Latin American/Hispanic 5 (6 %) 2 (4 %)White 59 (69 %) 35 (71 %)Other 5 (6 %) 2 (4 %)Mixed 3 (4 %) 0Marital status, N (%)Never Married 48 (57 %) 26 (53 %)Separated 7 (8 %) 1 (2 %)Married 16 (19 %) 12 (25 %)Divorced 7 (8 %) 3 (6 %)Domestic Partnership 5 (6 %) 6 (12 %)Widowed 2 (2 %) 1 (2 %)Occupational status, N (%)Working now 45 (53 %) 28 (57 %)Disabled (permanent ortemporary)13 (7 %) 0Temporary leave 5 (3 %) 0Looking, unemployed 10 (5 %) 3 (6 %)Student 8 (4 %) 15 (31 %)Retired 1 (1 %) 0If employed, number of hoursscheduled to work over the past2 weeks: mean (SD)50.0 (27.2) 53.8 (25.8)If employed, number of hoursmissed due to symptoms overthe past 2 weeks: mean (SD)10.5 (18.2) 0.1 (0.6)Education, years: mean (SD) 14.1 (2.0) 15.9 (2.9)Age of onset of MDD, years:mean (SD), range 20.6 (10.7), 5–55 n/aSingle episode:Recurrent, N (%) 24:61 (29 %:71 %) n/aNo. previous episodes, mean (SD) 4.3 (2.8) n/aCurrent episode duration≤12 months 39 (46 %) n/a1–2 years 13 (15 %)>2 years 30 (35 %)Unknown/not reported 3 (4 %)Median duration (range), months 14.5 (3–151)Comorbiditiesb,cSubstance-Related Disorders 7 (8 %) n/aTable 3 Clinical characteristics of the first cohort (N = 134)(Continued)Anxiety Disorders 67 (79 %)Eating Disorders 1 (1 %)Stable medical conditions 52 (61 %)Use of antidepressants duringcurrentepisode, N (%) 60 (71 %) n/aNo. antidepressants used,mean (SD)1.7 (1.5)Baseline MADRS, mean (SD) 29.9 (6.0) 0.4 (1.0)Baseline YMRS, mean (SD) 2.2 (1.8) 0.3 (0.7)Baseline CGI Severity, mean (SD) 4.7 (0.8) 1.0 (0.0)Baseline QIDS-SR, mean (SD) 15.9 (4.3) 2.1 (1.7)Baseline SDS, mean (SD) 16.6 (7.5) 0.0 (0.0)Baseline LEAPS, mean (SD) 14.1 (6.1) 1.9 (2.7)aCategories adapted from ethnic-origin groups listed in national censusquestionnaires [67]bBased on DSM-IV-TR, as determined by the Mini InternationalNeuropsychiatric InterviewcPercentages may not add up to 100 % because patients may have more than1 comorbid conditionMDD Major depressive disorder, SD Standard deviation, MADRS MontgomeryÅsberg Depression Rating Scale, YMRS Young Mania Rating Scale, CGI ClinicalGlobal Impression, QIDS-SR Quick Inventory of Depressive Symptomatology,Self-Report, SDS Sheehan Disability Scale, LEAPS Lam Employment Absenceand Productivity ScaleLam et al. BMC Psychiatry  (2016) 16:105 Page 9 of 13platform with various pre-clinical groups that investigateneuropharmacology of hedonic function in laboratory rats,zebrafish high-throughput screening, electrophysiologicaland behavioural impact of medications, and microRNA/in-flammatory markers. CAN-BIND activities are broadlydisseminated through multimodal, iterative knowledgetranslation/knowledge exchange initiatives that includeongoing collaboration with patients, family members,and other stakeholders in the community. Over time,CAN-BIND will expand with new collaborations withacademic health centres, industry partners, and mentalhealth networks.ConclusionCAN-BIND is a large Canadian collaborative researchendeavor that is attempting to discover integrated clin-ical, imaging and molecular biomarkers of treatment re-sponse in MDD. It may also identify clinically relevantsubtypes of depression and further our knowledge of thepathogenesis and pathophysiology of MDD. Given themultifaceted study design, we also expect to find novelpsychobiological insights that will lead to the generationof new hypotheses to be validated in future studies.Ethics approval and consent to participateResearch described in this article has been ethics ap-proved at each participating clinical centre. The ethicscommittees include: University of British ColumbiaClinical Research Ethics Board (Vancouver); Universityof Calgary Conjoint Health Research Ethics Board(Calgary); University Health Network Research EthicsBoard (Toronto); Centre for Addiction and MentalHealth Research Ethics Board (Toronto); Hamilton Inte-grated Research Ethics Board (Hamilton); Queen’s Univer-sity Health Sciences and Affiliated Teaching HospitalsResearch Ethics Board (Kingston). Participants providedwritten, informed consent for all study procedures.Consent for publicationNot applicable - this manuscript does not contain any in-dividual persons’ data.Abbreviations21 CFR Part 11: title 21 of the code of federal regulations [U.S.]; electronicrecords; electronic signatures, part 11; ATHF: antidepressant treatment historyform; BASE: bioArray software environment; BIS/BAS: behavioural inhibitionsystem/behavioural activation system; BOLD: blood oxygenation level-dependent; BPI-SF: brief pain inventory; Brain-CODE: Ontario brain institutecentre for Ontario data exploration; BRIAN: biological rhythm interview ofassessment in neuropsychiatry; CAN-BIND: Canadian biomarker integrationnetwork in depression; CANMAT: Canadian network for mood and anxietytreatments; CBT: cognitive-behavioural therapy; CECA: childhood experienceof care and abuse; CGI: clinical global impression; CNS VS: central nervoussystem vital signs; Cz: vertex [in EEG]; DARS: dimensional anhedonia ratingscale; DICOM: digital imaging and communications in medicine;DID: depression inventory development; DNA: deoxyribonucleic acid; DSM-5: diagnostic and statistical manual of mental disorders, fifth edition; DSM-IV-TR: diagnostic and statistical manual of mental disorders, fourth edition, textrevision; DTI: diffusion-tensor imaging; ECR-R: experiences in closerelationships; EEG: electroencephalography; ELISA: enzyme-linkedimmunosorbent assay; EPI: echo planar imaging; e-PRO: electronic version ofpatient-reported outcome; fMRI: functional magnetic resonance imaging;FOV: field of view; GAD: generalized anxiety disorder; GE: general electric;HCL-32: hypomania check-list, 32-item version; HIPAA: health insuranceportability and accountability act [U.S.]; HPCVL: high performance computingvirtual laboratory; Hz: hertz; ICH E6: international conference onharmonisation of technical requirements for registration of pharmaceuticalsfor human use, guidance on efficacy, 6 (good clinical practice);IPAQ: international physical activity questionnaire; iTBS: intermittent theta-burststimulation; KT/KE: knowledge transfer/knowledge exchange; LEAPS: lamemployment absence and productivity scale; LEDS: life events and difficultiesschedule; LTE: list of threatening experiences; MADRS: montgomery-asbergdepression rating scale; MDD: major depressive disorder; MDE: major depressiveepisode; MINI: mini international neuropsychiatric interview; miRNA: microRNA;mm: millimetres; MRI: magnetic resonance imaging; mRNA: messenger RNA;ms: milliseconds; NART: national adult reading test; NEO-FFI: NEO five-factor in-ventory; NIfTI: neuroimaging informatics technology initiative; OBI: Ontario braininstitute; QC: quality control; QIDS-SR: quick inventory of depressivesymptomatology, self-report; Q-LES-Q: quality of life, enjoyment and satisfactionquestionnaire; QTRAP: quadruple/linear ion trap; PbD: privacy by design;PIPEDA: personal information protection and electronic documents act;PRN: pro re nata [i.e., in medicine, “as needed”]; PSQI: Pittsburgh sleep qualityindex; REDCap: research electronic data capture; RNA: ribonucleic acid;rTMS: repetitive transcranial magnetic stimulation; SDS: sheehan disability scale;SexFX: sexual side effects questionnaire; SHAPS: snaith-hamilton pleasure scale;SIGMA: structured interview guide for the MADRS; SMI: serious mental illness;SNP: single-nucleotide polymorphism; SOP: standard operating procedures;SPAQ: seasonal pattern assessment questionnaire; SPReD: stroke patientrecovery research database; SRM-MS: selected-reaction monitoring massspectrometry; TE: echo time [in MRI]; TR: repetition time [in MRI]; TSES: Torontoside effects scale; WHOQoL-BREF: world health organization quality of lifeassessment; XNAT: extensible neuroimaging archiving toolkit; YMRS: youngmania rating scale.Competing interestsRWL has received speaker and consultant honoraria or research funds fromAstraZeneca, Brain Canada, Bristol-Myers Squibb, the Canadian Institutes ofHealth Research (CIHR), the Canadian Network for Mood and Anxiety Treatments,the Canadian Psychiatric Association, Eli Lilly, Janssen, Lundbeck, LundbeckInstitute, Medscape, Otsuka, Pfizer, Servier, St. Jude Medical, Takeda, the UniversityHealth Network Foundation, and Vancouver Coastal Health Research Institute.ACA has received funds form CIHR, OMHF, Ministry of Research andInnovation of Ontario, CAMH Foundation.In the last 5 years, ZJD received research and equipment in-kind support foran investigator-initiated study through Brainsway Inc. ZJD has served on theadvisory board for Sunovion, Hoffmann-La Roche Limited and Merck andreceived speaker support from Eli Lilly. He has also received support fromthe Ontario Mental Health Foundation (OMHF), the Canadian Institutes ofHealth Research (CIHR), the Brain and Behaviour Research Foundation andthe Temerty Family and Grant Family and through the Centre for Addictionand Mental Health (CAMH) Foundation and the Campbell Institute.BNF has received grant/research support from Alternative Funding PlanInnovations Award, Brain and Behavior Research Foundation, CanadianInstitutes of Health Research, Hamilton Health Sciences Foundation, J.P. BickellFoundation, Ontario Brain Institute, Ontario Mental Health Foundation, Societyfor Women’s Health Research, Teresa Cascioli Charitable Foundation, Eli Lilly andPfizer, and has received consultant and/or speaker fees from AstraZeneca,Bristol-Myers Squibb, Canadian Psychiatric Association, CANMAT, Daiichi Sankyo,Lundbeck, Pfizer, Servier and Sunovion.ZI has received research funding from CIHR, Canadian Consortium forNeurodegeneration and Aging, National Institute of Aging, Joan and CliffordHatch Foundation, Katthy Taylor Chair in Vascular Dementia, Ontario AFPInnovation Fund and University of Calgary Department of Psychiatry, as wellas consultant honoraria from BMS/Otsuka, Janssen, Lundbeck, Pfizer andSunovion.DJM has received research funds from CIHR, CFI, NARSAD, OMHF, NIH andthe University of Toronto.LM has received Grants/Research Support from Alternative Funding PlanInnovations Award, Brain & Behavioral Foundation, Canadian Institutes ofLam et al. BMC Psychiatry  (2016) 16:105 Page 10 of 13Health Research, Hamilton Health Sciences Foundation, Ontario BrainInstitute, and Ontario Mental Health Foundation, and Speakers Bureau/Honoraria from Bristol-Myers Squibb, Lundbeck, Canadian Psychiatric Associ-ation, and the Canadian Network for Mood and Anxiety Treatments.LCQ has received research funds from Ontario Mental Health Foundation,Gambling Research Exchange Ontario, Campbell Family Mental HealthResearch Institute, American Psychiatric Association, National Institutes ofHealth, and Canadian Consortium for Gambling Research.AVR has received speaker and consultant honoraria or research funds fromBristol Myers Squibb, Canadian Depression Research and InterventionNetwork, Canadian Foundation for Innovation and the Ministry of EconomicDevelopment and Innovation, Canadian Institutes of Health Research, GrandChallenges Canada, Janssen, Lundbeck, Ontario Mental Health Foundation,Pfizer and Sunovion.ALV has received consultant honoraria from Janssen and Roche.PB has received grant funding and/or honoraria for lectures and/orparticipation in advisory boards for Astra Zeneca, Bristol Myers Squibb, EliLilly, Forest, Euthymics, Janssen, Lundbeck, Merck, Otsuka, Pfizer, Pierre Fabre,Servier, Shire, Takeda, and Valeant.FF has received funding from NARSAD and NSERC.SVP has been a consultant to Takeda, Bristol Myers Squibb, Lundbeck; hashad a research contract with Assurex; has equity in Mensante.SCS is Chief Scientific Officer of ADM Diagnostics, LLC, Chicago.GT has an Investigator-initiated grant from Pfizer Canada, honorarium fromBristol-Meyers Squibb Canada and Janssen Canada.PG has grants/research support from CIHR, NIH; and speaker's bureau/honoraria from Bristol-Meyers-Squibb, Lundbeck, St.Jude Medical.JD has received research support from the Canadian Institutes of HealthResearch, Brain Canada, the National Institutes of Health, the Klarman FamilyFoundation, the Edgestone Foundation, and the Toronto General andWestern Hospital Foundation, as well as travel stipends from Lundbeck andANT Neuro, and in-kind equipment support for an investigator-initiated studyfrom MagVenture.FVR has received grant funding from Canadian Institutes of Health Researchand Brain CanadaGMM has been on advisory board or speaker for Lundbeck, Lilly, Pfizer, JanssenRM has received speaker and consultant honoraria or research funds fromAstraZeneca, Bristol-Myers Squibb, the Canadian Institutes of Health Research(CIHR), the Canadian Network for Mood and Anxiety Treatments, Eli Lilly,Janssen, Lundbeck, Ontario Mental Health Foundation, Otsuka, Paladin, Pfizer,Sunovion, and the University Health Network Foundation.SHK has received funding from Allergan, Brain Canada, Bristol-Meyers Squibb,Canadian Institutes of Health Research, Janssen, Lundbeck, Lundbeck Insti-tute, Medscape, Ontario Brain Institute, Otsuka, Pfizer, Servier, St. Jude Med-ical, Sunovion.SR, HEF, GBH, KLH, ML, JAF, FL, JG, FMP, CB, MPM, KRE, SH, CNS, MD and TVSlist no competing interests.Authors’ contributionsSHK, RWL, RM, SR, PB, KRE, BNF, PG, GMM, SVP, AVR, CNS, and GT madesubstantial contributions to overall study conception and design and toongoing data collection and analyses. ACA, MD, KRE, JAF, JG, HEF, DJM, andGT contributed to development of the molecular platform and continue tocontribute to data collection and/or analysis related to the platform. GBH,SH, LM, TVM, SCS, JD, FVR contributed to the development of the fMRIplatform and continue to contribute to data collection and/or analysisrelated to the platform. MPM, CB, FF, ML, JG contributed to the developmentof the EEG platform and continue to contribute to data analysis related tothe platform. MD, KRE and ALV contributed to the development of the datamanagement platform. SHK, RWL, RM, SR, ZJD, JD, KRE, JAF, BNF, PG, DJM,TVM, KLH, ZI, PB, FL, GMM, MPM, LM, SVP, FMP, LCQ, AVR, CNS, GT, ALV, andFVR contributed to the development of the clinical platform and continue tocontribute to data collection and/or analysis related to the platform. RWLdrafted the manuscript. All authors were involved in critical revisions of themanuscript. All authors read and approved the final manuscript.AcknowledgementsCAN-BIND is an Integrated Discovery Program carried out in partnershipwith, and financial support from, the Ontario Brain Institute, an independentnon-profit corporation, funded partially by the Ontario government. Theopinions, results and conclusions are those of the authors and no endorsementby the Ontario Brain Institute is intended or should be inferred. Additionalfunding is provided by the Canadian Institutes of Health Research (CIHR),Lundbeck, Bristol-Myers Squibb,, and Servier. Funding and/or in kind support isalso provided by the investigators’ universities and academic institutions. Allstudy medications are independently purchased at wholesale market values.Members of the CAN-BIND Investigator Team are listed here: www.canbind.ca/our-team/.Author details1University of British Columbia and Vancouver Coastal Health Authority, 2255Wesbrook Mall, Vancouver, BC V6T 2A1, Canada. 2Queen’s University,Providence Care, Mental Health Services 752 King Street West, Postal Bag603, Kingston, ON K7L 7X3, Canada. 3University Health Network, 399 BathurstStreet, Toronto, ON M5T 2S8, Canada. 4Department of Psychiatry, Universityof Toronto, 250 College Street, 8th floor, Toronto, ON M5T 1R8, Canada.5Centre for Addiction and Mental Health, 1001 Queen St. W, Toronto, ONM6J 1A8, Canada. 6University of Ottawa Institute of Mental Health Research,1145 Carling Avenue, Ottawa, ON K1Z 7K4, Canada. 7Loma Linda University,24851 Circle Dr, Loma Linda, CA 92354, USA. 8Indoc Research, 258 AdelaideSt. East, Suite 200, Toronto, ON M5A 1N1, Canada. 9Department of Pathologyand Molecular Medicine, Queen’s University, 88 Stuart Street, Kingston, ONK7L 3N6, Canada. 10McMaster University, and St. Joseph’s HealthcareHamilton, 1280 Main Street West, Hamilton, ON L8S4L8, Canada.11Department of Psychology, Queen’s University, Kingston, ON K7L 3N6,Canada. 12Aston University, Aston Triangle, Birmingham, West Midlands B47ET, UK. 13University of Calgary Hotchkiss Brain Institute, 2500 University DrNW, Calgary, AB T2N 1N4, Canada. 14University of Guelph, 50 Stone Rd E,Guelph, ON N1G 2W1, Canada. 15Simon Fraser University, 8888 University Dr,Burnaby, BC V5A 1S6, Canada. 16Universisty of Michigan, 500S State St, AnnArbor, MI48109USA. 17University of Reading, Earley Gate, Whiteknights,Reading RG6 6AL, UK. 18St. Michael’s Hospital, 193 Yonge St, Toronto, ONM5B 1M4, Canada. 19Rotman Research Institute at Baycrest Centre, 3560Bathurst Street, Toronto, ON M6A 2E1, Canada. 20McGill University , 845 RueSherbrooke O, Montréal, QC H3A 0G4, Canada. 21Douglas Mental HealthUniversity Institute Frank B. Common (FBC) F-3145, 6875 LaSalle Boulevard,Montréal, QC H4H 1R3, Canada.Received: 4 March 2016 Accepted: 18 March 2016References1. Ferrari AJ, Charlson FJ, Norman RE, Flaxman AD, Patten SB, Vos T, WhitefordHA. The epidemiological modelling of major depressive disorder: applicationfor the Global Burden of Disease Study 2010. PLoS One. 2013;8(7):e69637.2. Global Burden of Disease Study 2013 Collaborators. Global, regional, andnational incidence, prevalence, and years lived with disability for 301 acuteand chronic diseases and injuries in 188 countries, 1990-2013: a systematicanalysis for the Global Burden of Disease Study 2013. Lancet. 2015;386(9995):743–800.3. Walker ER, McGee RE, Druss BG. Mortality in mental disorders and globaldisease burden implications: a systematic review and meta-analysis. JAMAPsychiatry. 2015;72:334–41.4. Collins PY, Patel V, Joestl SS, March D, Insel TR, Daar AS. Scientific AdvisoryBoard and the Executive Committee of the Grand Challenges on Global MentalHealth. Grand challenges in global mental health. Nature. 2011;475:27–30.5. American Psychiatric Association. Diagnostic and Statistical Manual ofMental Disorders. 5th ed. Arlington: American Psychiatric Association; 2013.6. Kennedy SH, Downar J, Evans KR, Feilotter H, Lam RW, MacQueen GM,Milev R, Parikh SV, Rotzinger S, Soares C. The Canadian BiomarkerIntegration Network in Depression (CAN-BIND): advances in responseprediction. Curr Pharm Des. 2012;18(36):5976–89.7. Stuss DT. The Ontario Brain Institute: completing the circle. Can J Neurol Sci.2014;41(6):683–93.8. Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, Hergueta T,Baker R, Dunbar GC. The Mini-International Neuropsychiatric Interview (M.I.N.I.):the development and validation of a structured diagnostic psychiatric interviewfor DSM-IV and ICD-10. J Clin Psychiatry. 1998;59(20):22–33. quiz 34-57.9. Sackeim HA. The definition and meaning of treatment-resistant depression.J Clin Psychiatry. 2001;62(16):10–7.10. Lam RW, Kennedy SH, Grigoriadis S, McIntyre RS, Milev R, Ramasubbu R, ParikhSV, Patten SB, Ravindran AV. Canadian Network for Mood and AnxietyLam et al. BMC Psychiatry  (2016) 16:105 Page 11 of 13Treatments (CANMAT). Canadian Network for Mood and Anxiety Treatments(CANMAT) clinical guidelines for the management of major depressive disorderin adults. III. Pharmacotherapy J Affect Disord. 2009;117(1):S26–43.11. Berman RM, Fava M, Thase ME, Trivedi MH, Swanink R, McQuade RD, CarsonWH, Adson D, Taylor L, Hazel J, Marcus RN. Aripiprazole augmentation in majordepressive disorder: a double-blind, placebo-controlled study in patients withinadequate response to antidepressants. CNS Spectr. 2009;14(4):197–206.12. Berman RM, Marcus RN, Swanink R, McQuade RD, Carson WH, Corey-LislePK, Khan A. The efficacy and safety of aripiprazole as adjunctive therapy inmajor depressive disorder: a multicenter, randomized, double-blind,placebo-controlled study. J Clin Psychiatry. 2007;68(6):843–53.13. Thase ME, Trivedi MH, Nelson JC, Fava M, Swanink R, Tran QV, Pikalov A,Yang H, Carlson BX, Marcus RN, Berman RM. Examining the efficacy ofadjunctive aripiprazole in major depressive disorder: a pooled analysis of 2studies. Prim Care Companion J Clin Psychiatry. 2008;10(6):440–7.14. Montgomery SA, Asberg M. A new depression scale designed to besensitive to change. Br J Psychiatry. 1979;134:382–9.15. Williams JB, Kobak KA. Development and reliability of a structured interviewguide for the Montgomery Asberg Depression Rating Scale (SIGMA). Br JPsychiatry. 2008;192(1):52–8.16. Guy W. ECDEU Assessment Manual for Psychopharmacology. Rockville: USDepartment of Health, Education, and Welfare Public Health ServiceAlcohol, Drug Abuse, and Mental Health Administration; 1976.17. Kalali A, Vaccarino A, Evans K, Kennedy S, Kobak K, Sills T, Sheehan D,Williams J, Greist J, Schatzberg A. P.2.a.010 The depression inventorydevelopment workgroup: Results from the second validation study. EuroNeuropsychopharm. 2008;18(4):S298.18. Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B, Klein DN, MarkowitzJC, Ninan PT, Kornstein S, Manber R, Thase ME, Kocsis JH, Keller MB. The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating(QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients withchronic major depression. Biol Psychiatry. 2003;54(5):573–83.19. Spitzer RL, Kroenke K, Williams JB, Löwe B. A brief measure for assessinggeneralized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166(10):1092–7.20. Young RC, Biggs JT, Ziegler VE, Meyer DA. A rating scale for mania:reliability, validity and sensitivity. Br J Psychiatry. 1978;133:429–35.21. Angst J, Adolfsson R, Benazzi F, Gamma A, Hantouche E, Meyer TD, SkepparP, Vieta E, Scott J. The HCL-32: towards a self-assessment tool forhypomanic symptoms in outpatients. J Affect Disord. 2005;88(2):217–33.22. Leon AC, Olfson M, Portera L, Farber L, Sheehan DV. Assessing psychiatricimpairment in primary care with the Sheehan Disability Scale. Int JPsychiatry Med. 1997;27(2):93–105.23. Lam RW, Michalak EE, Yatham LN. A new clinical rating scale for workabsence and productivity: validation in patients with major depressivedisorder. BMC Psychiatry. 2009;9:78.24. The WHOQOL Group. The development of the World Health Organizationquality of life assessment instrument (the WHOQOL). In: Orley J, Kuyken W,editors. Quality of Life Assessment: International Perspectives. Heidelberg:Springer Verlag; 1994. p. 41–57.25. Endicott J, Nee J, Harrison W, Blumenthal R. Quality of Life Enjoyment andSatisfaction Questionnaire: a new measure. Psychopharmacol Bull. 1993;29(2):321–6.26. Carver CS, White TL. Behavioral inhibition, behavioral activation, andaffective responses to impending reward and punishment: The BIS/BASScales. J Pers Soc Psychol. 1994;67(2):319–33.27. Rizvi SJ, Quilty LC, Sproule BA, Cyriac A, Michael Bagby R, Kennedy SH.Development and validation of the Dimensional Anhedonia Rating Scale(DARS) in a community sample and individuals with major depression.Psychiatry Res. 2015;229(1-2):109–19.28. Snaith RP, Hamilton M, Morley S, Humayan A, Hargreaves D, Trigwell P. Ascale for the assessment of hedonic tone the Snaith-Hamilton PleasureScale. Br J Psychiatry. 1995;167(1):99–103.29. Costa PT, McCrae RR. Revised NEO Personality Inventory (NEO PI-R) and NEOFive-Factor Inventory (NEO-FFI) Professional Manual. Odessa: PsychologicalAssessment Resources, Inc.; 1992.30. Cleeland CS. Measurement of pain by subjective report. In: Chapman CR &Loeser JD (eds.), Issues in Pain Measurement. Advances in Pain Researchand Therapy, Volume.12. New York: Raven Press; 1989; 391–403.31. Buysse DJ, Reynolds 3rd CF, Monk TH, Berman SR, Kupfer DJ. The PittsburghSleep Quality Index: a new instrument for psychiatric practice and research.Psychiatry Res. 1989;28(2):193–213.32. Giglio LM, Magalhães PV, Andreazza AC, Walz JC, Jakobson L, Rucci P, RosaAR, Hidalgo MP, Vieta E, Kapczinski F. Development and use of a biologicalrhythm interview. J Affect Disord. 2009;118(1-3):161–5.33. Magnusson A. Validation of the Seasonal Pattern Assessment Questionnaire(SPAQ). J Affect Disord. 1996;40(3):121–9.34. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE,Pratt M, Ekelund U, Yngve A, Sallis JF, Oja P. International physical activityquestionnaire: 12-country reliability and validity. Med Sci Sports Exerc.2003;35(8):1381–95.35. Fraley RC, Waller NG, Brennan KA. An item response theory analysis of self-report measures of adult attachment. J Pers Soc Psychol. 2000;78(2):350–65.36. Brugha TS, Cragg D. The List of Threatening Experiences: the reliability andvalidity of a brief life events questionnaire. Acta Psychiatr Scand. 1990;82(1):77–81.37. Bifulco A, Brown GW, Harris TO. Childhood Experience of Care and Abuse(CECA): a retrospective interview measure. J Child Psychol Psychiatry. 1994;35(8):1419–35.38. Bifulco A, Brown G, Edwards A, Harris T, Neilson E, Richards C, Robinson R. LifeEvents and Difficulties Schedule (LEDS–2): Vol. 1. Life events manual. London:Royal Holloway and Bedford New College, University of London; 1989.39. Iverson GL, Brooks BL, Young AH. Identifying neurocognitive impairment indepression using computerized testing. Appl Neuropsychol. 2009;16(4):254–61.40. Gualtieri CT, Johnson LG. Reliability and validity of a computerizedneurocognitive test battery, CNS Vital Signs. Arch Clin Neuropsychol.2006;21(7):623–43.41. Nelson HE. National Adult Reading Test (NART): Test manual (Part1).Windsor: NFER-NELSON Publishing Company Ltd.; 1982.42. Vanderkooy JD, Kennedy SH, Bagby RM. Antidepressant side effects indepression patients treated in a naturalistic setting: a study of bupropion,moclobemide, paroxetine, sertraline, and venlafaxine. Can J Psychiatry. 2002;47(2):174–80.43. Kennedy SH, Eisfeld BS, Dickens SE, Bacchiochi JR, Bagby RM. Antidepressant-induced sexual dysfunction during treatment with moclobemide, paroxetine,sertraline, and venlafaxine. J Clin Psychiatry. 2000;61(4):276–81.44. Glover GH, Mueller BA, Turner JA, van Erp TG, Liu TT, Greve DN, Voyvodic JT,Rasmussen J, Brown GG, Keator DB, Calhoun VD, Lee HJ, Ford JM, MathalonDH, Diaz M, O'Leary DS, Gadde S, Preda A, Lim KO, Wible CG, Stern HS,Belger A, McCarthy G, Ozyurt B, Potkin SG. Function biomedical informaticsresearch network recommendations for prospective multicenter functionalMRI studies. J Magn Reson Imaging. 2012;36(1):39–54.45. Kurmis AP, Barber C, Slavotinek JP, Fazzalari NL. A MnCl2-based MR signalintensity linear response phantom. Radiol Technol. 2007;79(2):119–25.46. Friedman L, Glover GH. Report on a multicenter fMRI quality assuranceprotocol. J Magn Reson Imaging. 2006;23(6):827–39.47. Fonov VS, Janke A, Caramanos Z, Arnold DL, Narayanan S, Pike GB, Collins DL.Improved precision in the measurement of longitudinal global and regionalvolumetric changes via a novel MRI gradient distortion characterization andcorrection technique. In: Laio H, editor. Medical Imaging and Augmented Reality.Lecture Notes in Computer Science, vol. 6326. Berlin Heidelberg: Springer; 2010.p. 324–33.48. Alzheimer's Disease Neuroimaging Insitiative MRI Technical ProceduresManual. [Internet]. 2013 [cited 2016 February 5]. Available from: http://www.adni-info.org/Scientists/doc/ADNI2MRITrainingManualVersion5_Oct2014_CLEAN.pdf.49. Raichle ME, Snyder AZ. A default mode of brain function: a brief history ofan evolving idea. Neuroimage. 2007;37(4):1083–90. discussion 1097-9.50. Egner T, Etkin A, Gale S, Hirsch J. Dissociable neural systems resolve conflict fromemotional versus nonemotional distracters. Cereb Cortex. 2008;18(6):1475–84.51. Etkin A, Egner T, Peraza DM, Kandel ER, Hirsch J. Resolving emotionalconflict: a role for the rostral anterior cingulate cortex in modulating activityin the amygdala. Neuron. 2006;51(6):871–82.52. Elliott R, Rubinsztein JS, Sahakian BJ, Dolan RJ. The neural basis of mood-congruent processing biases in depression. Arch Gen Psychiatry. 2002;59(7):97–604.53. Hall GB, Milne AM, MacQueen GM. An fMRI study of reward circuitry inpatients with minimal or extensive history of major depression. Eur ArchPsychiatry Clin Neurosci. 2014;264(3):187–98.54. Vrieze E, Pizzagalli DA, Demyttenaere K, Hompes T, Sienaert P, de Boer P,Schmidt M, Claes S. Reduced reward learning predicts outcome in majordepressive disorder. Biol Psychiatry. 2013;73(7):639–45.55. Albert J, López-Martín S, Carretié L. Emotional context modulates responseinhibition: neural and behavioral data. Neuroimage. 2010;49(1):914–21.Lam et al. BMC Psychiatry  (2016) 16:105 Page 12 of 1356. Delorme A, Makeig S. EEGLAB: an open source toolbox for analysis ofsingle-trial EEG dynamics. J Neurosci Methods. 2004;134:9–21.57. Trevino V, Falciani F, Barrera-Saldaña HA. DNA microarrays: a powerfulgenomic tool for biomedical and clinical research. Mol Med. 2007;13(9-10):527–41.58. Einat P. Methodologies for high-throughput expression profiling ofmicroRNAs. Methods Mol Biol. 2006;342:139–57.59. Chiu CL, Randall S, Molloy MP. Recent progress in selected reaction monitoringMS-driven plasma protein biomarker analysis. Bioanalysis. 2009;1(4):847–55.60. High Performance Computing Virtual Lab [Internet]. Kingston, ON,Canada: High Performance Computing Virtual Lab; 2016 [accessed 2016Jan 30]. Available from:https://www.hpcvl.org/security/security-policy61. Gee T, Kenny S, Price CJ, Seghier ML, Small SL, Leff AP, Pacurar A, StrotherSC. Data warehousing methods and processing infrastructure for brainrecovery research. Archives Italiennes de Biologie. 2010; 148(3):207–17.62. National Electrical Manufacturers Association. NEMA PS3 / ISO 12052, DigitalImaging and Communications in Medicine (DICOM) Standard. Virginia; 2016.63. Vallon-Christersson J, Nordborg N, Svensson M, Häkkinen J. BASE – 2ndgeneration software for microarray data management and analysis. BMCBioinformatics. 2009;10:330.64. Nelson EK, Piehler B, Eckels J, Rauch A, Bellew M, Hussey P, Ramsay S, NatheC, Lum K, Krouse K, Stearns D, Connolly B, Skillman T, Igra M. LabKey Server:an open source platform for scientific data integration, analysis andcollaboration. BMC Bioinformatics. 2011;12:71.65. Harris PA, Taylor T, Thielke R, Payne J, Gonzalez N, Conde JG. Researchelectronic data capture (REDCap) - A metadata-driven methodology andworkflow process for providing translational research informatics support.J Biomed Inform. 2009;42(2):377–81.66. Mark J. NIfTI-1 Data Format: Neuroimaging Informatics Technology Initiative[Internet]. 2007 [cited 2016 February 5]. Available from: http://nifti.nimh.nih.gov/nifti-1.67. Statistics Canada; Ottawa, ON; Canada. Ethnic Origin Reference Guide, 2006Census. Catalogue no. 97-562-GWE2006025. http://www12.statcan.gc.ca/census-recensement/2006/ref/rp-guides/ethnic-ethnique-eng.cfm. Lastmodified: 4 May 2011.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Lam et al. BMC Psychiatry  (2016) 16:105 Page 13 of 13

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0308586/manifest

Comment

Related Items