STUDY PROTOCOL published: 18 January 2022 doi: 10.3389/fpsyt.2021.807839 Focusing on Comorbidity—A Novel Meta-Analytic Approach and Protocol to Disentangle the Specific Neuroanatomy of Co-occurring Mental Disorders Lydia Fortea 1,2,3 , Anton Albajes-Eizagirre 1 , Yuan-Wei Yao 4,5,6 , Edu Soler 1 , Edited by: Norma Verdolini 1,2,3,7 , Alexander O. Hauson 8,9 , Adriana Fortea 1,2,3,10,11 , Santiago Madero 1,12 , Stefan Borgwardt, Aleix Solanes 1,13 , Scott C. Wollman 8 , Maria Serra-Blasco 2,14,15 , Toby Wise 16 , University of Lübeck, Germany Steve Lukito 17 , Maria Picó-Pérez 18,19,20 , Christina Carlisi 17,21 , JinTao Zhang 22 , Reviewed by: PingLei Pan 23 , Álvar Farré-Colomés 1,24 , Danilo Arnone 25,26 , Matthew J. Kempton 16,27 , Felix Brandl, Carles Soriano-Mas 2,28,29 , Katya Rubia 17 , Luke Norman 17,30,31 , Paolo Fusar-Poli 27,32,33,34 , Technical University of David Mataix-Cols 35,36 , Marc Valentí 1,2,3,7,11 , Esther Via 37,38 , Narcis Cardoner 2,13,39 , Munich, Germany Marco Solmi 32,40,41,42 , Jae I. Shin 43 , Eduard Vieta 1,2,3,7,11 and Joaquim Radua 1,2,27,35* Andreea Oliviana Diaconescu, Krembil Centre for Neuroinformatics, 1 Institut d’Investigacions Biomèdiques August Pi I Sunyer (IDIBAPS), Barcelona, Spain, 2 Centro de Investigación Biomédica Centre for Addiction and Mental en Red de Salud Mental (CIBERSAM), Instituto de Salud Carlos III, Madrid, Spain, 3 Department of Medicine, University of Health, Canada Barcelona, Barcelona, Spain, 4 Department of Education and Psychology, Freie Universität Berlin, Berlin, Germany, 5 Einstein Center for Neurosciences Berlin, Charité - Universitätsmedizin Berlin, Berlin, Germany, 6 Berlin School of Mind and Brain, *Correspondence: Humboldt-Universität zu Berlin, Berlin, Germany, 7 Bipolar and Depressive Disorders Unit, Hospital Clinic, Barcelona, Spain, Joaquim Radua 8 Clinical Psychology PhD Program, California School of Professional Psychology, San Diego, CA, United States, radua@clinic.cat 9 Department of Psychiatry, University of California San Diego, La Jolla, CA, United States, 10 Fundació Clínic per a la Recerca Biomèdica (FCRB), Barcelona, Spain, 11 Psychiatric and Psychology Service, Hospital Clinic, Barcelona, Spain, Specialty section: 12 Schizophrenia Unit, Hospital Clinic, Barcelona, Spain, 13 Department of Psychiatry and Forensic Medicine, Autonomous This article was submitted to University of Barcelona, Barcelona, Spain, 14 Department of Psychology, Abat Oliba CEU (“Centro de Estudios Universitarios”) Neuroimaging and Stimulation, University, Barcelona, Spain, 15 Programa E-Health ICOnnecta’t, Institut Català d’Oncologia, Barcelona, Spain, 16 Department a section of the journal of Neuroimaging, Institute of Psychiatry, Psychology, and Neuroscience, King’s College London, London, United Kingdom, Frontiers in Psychiatry 17 Department of Child and Adolescent Psychiatry, Institute of Psychiatry, Psychology, and Neuroscience, King’s College London, London, United Kingdom, 18 Live and Health Sciences Research Institute (ICVS), University of Minho, Received: 02 November 2021 Braga, Portugal, 19 ICVS/3B’s, PT Government Associate Laboratory, Braga, Portugal, 20 Clinical Academic Center - Braga, Accepted: 13 December 2021 Braga, Portugal, 21 Division of Psychology and Language Sciences, University College London, London, United Kingdom, Published: 18 January 2022 22 State Key Laboratory of Cognitive Neuroscience and Learning and IDG/McGovern Institute for Brain Research, Beijing Citation: Normal University, Beijing, China, 23 Department of Neurology, Department of Addictive Behavior and Addiction Medicine, Fortea L, Albajes-Eizagirre A, Yao Y-W, Central Institute of Mental Health, Affiliated Yancheng Hospital of Southeast University, Yancheng, China, 24 Medical Faculty Soler E, Verdolini N, Hauson AO, Mannheim, University of Heidelberg, Mannheim, Germany, 25 Department of Psychological Medicine, Institute of Psychiatry, Fortea A, Madero S, Solanes A, Psychology, and Neuroscience, King’s College London, London, United Kingdom, 26 Department of Psychiatry and Wollman SC, Serra-Blasco M, Wise T, Behavioral Science, College of Medicine and Health Sciences, United Arab Emirates University (UAEU), Al Ain, United Lukito S, Picó-Pérez M, Carlisi C, Arab Emirates, 27 Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College Zhang J, Pan P, Farré-Colomés Á, London, London, United Kingdom, 28 Psychiatry and Mental Health Group, Neuroscience Program, Bellvitge Biomedical Arnone D, Kempton MJ, Research Institute (IDIBELL), L’Hospitalet de Llobregat, Barcelona, Spain, 29 Department of Psychobiology and Methodology Soriano-Mas C, Rubia K, Norman L, of Health Sciences, Universitat Autònoma de Barcelona, Barcelona, Spain, 30 Department of Psychiatry, University of Fusar-Poli P, Mataix-Cols D, Valentí M, Michigan, Ann Arbor, MI, United States, 31 The Social and Behavioral Research Branch, National Human Genome Research Via E, Cardoner N, Solmi M, Shin JI, Institute, National Institute of Health, Bethesda, MD, United States, 32 Early Psychosis: Interventions and Clinical-Detection Vieta E and Radua J (2022) Focusing (EPIC) Lab, Department of Psychosis Studies, Institute of Psychiatry, Psychology, London, United Kingdom, 33 Department of on Comorbidity—A Novel Brain and Behavioral Sciences, University of Pavia, Pavia, Italy, 34 Outreach and Support in South London (OASIS) Service, Meta-Analytic Approach and Protocol South London and Maudsley NHS Foundation Trust, London, United Kingdom, 35 Department of Clinical Neuroscience, to Disentangle the Specific Center for Psychiatry Research, Karolinska Institutet, Stockholm, Sweden, 36 Stockholm Health Care Services, Stockholm Neuroanatomy of Co-occurring County Council, Stockholm, Sweden, 37 Child and Adolescent Psychiatry and Psychology Department, Hospital Sant Joan Mental Disorders. de Déu, Barcelona, Spain, 38 Child and Adolescent Mental Health Research Group, Institut de Recerca Sant Joan de Déu, Front. Psychiatry 12:807839. Barcelona, Spain, 39 Mental Health Department, Hospital Universitari Parc Taulí, Institut d’Investigació i Innovació Parc Taulí doi: 10.3389/fpsyt.2021.807839 (I3PT), Sabadell, Spain, 40 Department of Psychiatry, University of Ottawa, Ottawa, ON, Canada Frontiers in Psychiatry | www.frontiersin.org 1 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity 41 Department of Mental Health, The Ottawa Hospital, Ottawa, ON, Canada, 42 Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, ON, Canada, 43 Department of Pediatrics, Yonsei University College of Medicine, Seoul, South Korea Background: In mental health, comorbidities are the norm rather than the exception. However, current meta-analytic methods for summarizing the neural correlates of mental disorders do not consider comorbidities, reducing them to a source of noise and bias rather than benefitting from their valuable information. Objectives: We describe and validate a novel neuroimaging meta-analytic approach that focuses on comorbidities. In addition, we present the protocol for a meta-analysis of all major mental disorders and their comorbidities. Methods: The novel approach consists of a modification of Seed-based d Mapping—with Permutation of Subject Images (SDM-PSI) in which the linear models have no intercept. As in previous SDM meta-analyses, the dependent variable is the brain anatomical difference between patients and controls in a voxel. However, there is no primary disorder, and the independent variables are the percentages of patients with each disorder and each pair of potentially comorbid disorders. We use simulations to validate and provide an example of this novel approach, which correctly disentangled the abnormalities associated with each disorder and comorbidity. We then describe a protocol for conducting the new meta-analysis of all major mental disorders and their comorbidities. Specifically, we will include all voxel-based morphometry (VBM) studies of mental disorders for which a meta-analysis has already been published, including at least 10 studies. We will use the novel approach to analyze all included studies in two separate single linear models, one for children/adolescents and one for adults. Discussion: The novel approach is a valid method to focus on comorbidities. The meta-analysis will yield a comprehensive atlas of the neuroanatomy of all major mental disorders and their comorbidities, which we hope might help develop potential diagnostic and therapeutic tools. Keywords: meta-analysis, magnetic resonance imaging (MRI), seed-based d mapping (SDM), gray matter (GM), mental disorder, comorbidity, medication INTRODUCTION disorder (OCD) neuroanatomical findings, where although the abnormalities of the corticostriatal-thalamocortical circuits have Authors have reported potential brain anatomical abnormalities been consistently reported (7–9), recent evidence has been for different mental disorders since the 1980s (1). At present, accumulating to other regions outside these circuits with less the neuroscientific community has enough data (thousands of agreement among meta- and mega-analysis reports (7, 10, 11). studies) to create an atlas of these abnormalities, but this is Moreover, the exact direction of increases and decreases in not yet a reality due to the heterogeneity in the findings across GM volume in certain areas, such as the orbitofrontal cortex studies investigating the same disorder. For instance, a meta- (OFC), has been unclear (12). For example, some reports show analysis of structural brain alterations of social anxiety disorder reduced GM in bilateral (13) or right (14) OFC, whereas (SAD) found that studies presented contradictory findings, there are also studies reporting increased bilateral (15) or left such as increases and decreases in gray matter (GM) volume (16) OFC. in the hippocampus and other brain regions (2). Similarly, The found heterogeneity may be partly related to the use whereas several meta-analyses had reported significant SAD- of magnetic resonance imaging (MRI) devices with varying related abnormalities in GM in the amygdala-hippocampal, field strengths (17) and head coils (18) or to the techniques prefrontal, and parietal regions (3–5), an ENIGMA study only or software used to process the images (18, 19). In addition, found a significant larger GM volume in the right putamen several clinical parameters might moderate the findings, such (6). Another example could be the case of obsessive-compulsive as the age at disease onset and the duration of disease (20, 21), Frontiers in Psychiatry | www.frontiersin.org 2 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity the current age (22), symptom heterogeneity (23), the gender of age, sex, medication, age of onset, duration of illness, and distribution (24), the medication status (25, 26), or clinical symptom severity. stage (27). For example, in a meta-analysis of attention-deficit hyperactivity disorder (ADHD) (28), we found that samples with METHODS AND ANALYSIS more medicated patients showed less decreased GM volume in the right caudate. The Novel Approach Another relevant but little explored source of heterogeneity The new approach is conceptually novel, but it involves only a may be the varying presence of comorbidities, which are the rule minor modification of the “seed-based d mapping-permutation in psychiatry (29). About half of people with mental disorders of subject images” (SDM-PSI) (www.sdmproject.com) (11, 40– have more than one comorbid mental disorder (30, 31). Some 42), an already validated and widely used brain imaging meta- authors claim that “there are no patients without comorbidity” analytic method (43–51). The main advantage of this method (32). We must acknowledge that some studies exclude individuals is that it directly tests whether there are differences between with specific comorbid mental disorders. Still, this exclusion patients and controls, rather than conducting indirect tests such is frequently limited to very few entities such as psychosis as whether peaks tend to converge in some regions more than in and substance use disorders. Similarly, we must acknowledge others (52). that some comorbidities are conceptually impossible, e.g., a It first creates maps of the lower and upper bounds of patient with bipolar disorder (BD) cannot be diagnosed with possible effect sizes for each study based on the available comorbid major depressive disorder (MDD), but again, the list statistical information and the anisotropic covariance between of impossible comorbidities is limited. Therefore, most patients adjacent voxels (53). Second, it uses maximum likelihood included in a study of a given mental disorder might also estimation techniques to impute several effect sizes maps for have other mental disorders. For example, in a meta-analysis of each study, assuming that the effect size follows a truncated neuroimaging studies in OCD, patients had comorbid depression normal distribution within the lower and upper bounds. Third, or anxiety disorders in 65% of the studies. The percentage of it fits a random-effects meta-analytic linear model separately patients with comorbid disorders reached 50% (for depression) for each voxel. Fourth, it combines the meta-analytic maps and 85% (for anxiety disorders) in some included studies (33). resulting from the different imputations using Rubin’s rules. This common presence of abnormalities from other mental Finally, it conducts a permutation test to yield threshold-free disorders might thus confound the results of case-control studies. cluster enhancement (TFCE)-based (54) familywise error rates Relevantly, previous studies have found that some brain (FWER, i.e., corrected p-values). abnormalities associated with different mental disorders are non- In this paper, we will first describe the new concept, then specific. For example, in several meta-analyses, we observed report the small methodological changes, and finally report a similar decreases of GM volume in the anterior cingulate/medial validation of the approach using simulations. frontal cortex in disorders as different as psychosis, anxiety disorders (AD), ADHD, and autism spectrum disorders (ASD) Description Commonly, the primary (random-effects meta-analytic) linear (33–37). Similarly, an ENIGMA study reported a high similarity model of a meta-analysis is just a (weighted) mean: of brain structural abnormalities between MDD, BD, OCD, and schizophrenia (38, 39). We fully acknowledge that some Yi = β + εi of these non-specific abnormalities may be transdiagnostic, i.e., associated with two or more mental disorders. However, where Yi is the effect size of the ith study, β is the meta-analytic there is also the possibility that some others are related to the effect size, and εi is the residual for the ith study. It is important to confounding effects of comorbidities. remember that this model is conducted separately for each voxel. This study aims to describe and validate a novel neuroimaging In some previous meta-analyses, we attempted to control for meta-analytic approach that focuses on comorbidities and comorbidity via meta-regression (covariation) by the percentage presents the protocol for a meta-analysis of all major mental of patients with a comorbid disorder: disorders and their potential comorbidities. We exemplify this protocol for voxel-based morphometry (VBM) studies Yi = βA + βAB−A Xi,AB + εi investigating GM volume differences between patients with mental disorders and healthy controls. However, it could be where the intercept βA is the effect size for patients with only similarly applied to any neuroimaging modality compatible disorder A, the coefficient βAB−A is the difference between with SDM (e.g., functional MRI or diffusion tensor imaging). patients with both disorders and patients with only the disorder Specifically, we will meta-analyze brain abnormalities from A, and X i,AB is the proportion of patients with both disorders in different disorders with different comorbidities, with a single the ith study. meta-linear model (though separately for children/adolescents However, these attempts had two relevant limitations. First, and adults). This analysis will yield an MRI-based atlas that βAB−A (the difference between patients with both disorder A and dissects the specific brain anatomical abnormalities of each disorder B and patients with only disorder A) mixed the effects mental disorder and comorbidity. In complementary analyses of disorder B and the effects of the comorbidity AB. Thus, we and depending on data availability reported in the studies, we could not know which part of βAB−A was shared by any patient will explore the potentially confounding or moderating effects with disorder B (with or without disorder A) and which part Frontiers in Psychiatry | www.frontiersin.org 3 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity of βAB−A was specific to those patients with both disorder A and disorder B. For instance, imagine that patients with only disorder A show decreased amygdala, patients with only disorder B show decreased cerebellum, and patients with both disorder A and disorder B show decreased amygdala, cerebellum, and prefrontal cortex. The decrease in the cerebellum is shared by any patient with disorder B, and the reduction in the prefrontal cortex is specific to those with both disorder A and disorder B, but βAB−A would mix both abnormalities. Second, we could only include studies on disorder A in which a variable proportion of patients had disorder B. Conversely, we could not include studies on disorder B in which a variable proportion of patients had the disorder A, with the subsequent loss in precision. Here, we propose to use the following model to study two disorders A and B: Yi = βA XiA + βB XiB + βAB XiAB + εi where βA is the effect size for patients with only disorder A, XiA FIGURE 1 | Rules for creating the gray matter maps of a simulated patient. is the proportion of patients with disorder A in the ith study, After creating a map using white noise, we added or subtracted 0.5 in the βB is the effect size for patients with only disorder B, XiB is the colored brain regions of interest (ROI) depending on the disorders he/she had. proportion of patients with disorder B in the ith study, βAB is the effect size of the comorbidity AB, and XiAB is the proportion of patients with both disorders in the ith study. This model overcomes the limitations of the previous attempts For example, imagine that disorder A was OCD, disorder B was because it separates the effects of disorder B and the effects of MDD, and disorder C referred to anxiety disorders. Studies on the comorbidity AB and can accept both studies on disorder OCD would be coded as XiA = 1, XiAB = [proportion of patients A and studies on disorder B because it treats all disorders with comorbid MDD], and XiAC = [proportion of patients with equally. Indeed, we can extend the model to as many disorders comorbid anxiety disorders]. Studies on MDD would be coded and comorbidities as wished. Still, considering the complexity as XiB = 1, XiAB = [proportion of patients with comorbid of the analysis and the likely poor reporting of co-occurring OCD], and XiBC = [proportion of patients with comorbid anxiety comorbidities, we will only consider pairs of comorbid mental disorders]. Finally, studies on anxiety disorders would be coded disorders that are possible (e.g., anxiety and MDD) and have been as XiC = 1, XiAC = [proportion of patients with comorbid studied by at least ten studies. OCD], and XiBC = [proportion of patients with comorbid MDD]. Although studies seldom report the proportion of patients with Validation multiple comorbidities (e.g., in studies on OCD, the proportion To validate the novel approach, we simulated 64 studies on of patients with both comorbid MDD and anxiety disorders), our disorder A, 64 on disorder B, and 64 on disorder C, with new approach could easily account for them if reported. Meta- varying levels of comorbidity, and then meta-analyzed them analysts should add a regressor for each reported combination of using the novel approach. The reason to simulate 64 studies for comorbid mental disorders. each disorder is that we simulated eight levels of comorbidity For comparison purposes, we also conducted three meta- (including no comorbidity) for each of the two comorbid regressions using the previous SDM-PSI approach (again, with disorders, e.g., for disorder A, we simulated eight levels of default values and defining statistical significance as TFCE comorbid disorder B X 8 levels of comorbid disorder C. permutation-based FWER < 0.05): Specifically, we first simulated for each study that a varying proportion of the patients had one of the other disorders or both. Yi = βA + βAB−A XiAB + βAC−A XiAC We then created the subjects’ gray matter maps as white noise Yi = βB + βAB−B XiAB + βBC−B XiBC following a standard normal distribution. Still, for each patient, we added or subtracted 0.5 in four brain regions depending Yi = βC + βAC−C XiAC + βBC−C XiBC on the disorders he/she had and the rules in Figure 1. We thus created abnormalities with a medium effect size (Cohen’s d Results of the Validation around 0.5). Finally, we conducted the voxelwise t-test between The novel SDM-PSI approach detected all the simulated patients and controls to derive the study t-map. abnormalities with the correct effect size and did not yield any To meta-analyze the studies’ t-maps with the novel SDM-PSI falsely positive findings (Table 1, left column; and Figure 2). approach (with default values and defining statistical significance The previous SDM-PSI reported a list of findings very similar as TFCE permutation-based FWER < 0.05), we modeled: to the simulated factual data (Table 1, right columns; and Figure 3). However, the results showed the two main limitations Yi = βA XiA + βB XiB + βC XiC + βAB XiAB + βAC XiAC + βBC XiBC we had expected. First, they mixed the abnormalities of the Frontiers in Psychiatry | www.frontiersin.org 4 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity TABLE 1 | Results of the novel approach validation. Simulated Novel SDM-PSI Previous SDM-PSI approach factual data approach Main disorder is A Main disorder is B Main disorder is C Disorder A ROI 1, g +0.5 βA = ROI 1, g +0.5 βA = ROI 1, g +0.5 βAB−B = ROI 1, +0.5 βAC −C = ROI 1, g +0.5 ROI 2, g +0.4 ROI 4, +0.5 ROI 4, g +0.2 Comorbidity AB ROI 4, g +0.5 βAB = ROI 4, g +0.5 βAB−A = ROI 4, g +0.5 (Not studied) ROI 2, g +0.4 ROI 3, g +0.5 Disorder B ROI 2, g +0.5 βB = ROI 2, g +0.5 βB = ROI 2, g +0.7 βBC −C = ROI 2, g +0.4 ROI 3, g +0.5 ROI 3, g +0.5 ROI 3, g +0.3 ROI 3, g +0.5 Comorbidity BC (None) βBC = (None) (Not studied) βBC−B = ROI 2, g +0.5 ROI 3, g −0.5 Disorder C ROI 2, g +0.5 βC = ROI 2, g +0.5 βAC−A = ROI 2, g +0.5 βC = ROI 2, g +0.7 ROI 3, g −0.5 ROI 3, g −0.5 ROI 3, g −0.5 ROI 3, g −0.3 Comorbidity AC (None) βAC = (None) (Not studied) (See Disorder A) “g”, average Hedges’ g of the voxels within the cluster of statistical significance; ROI, region of interest. FIGURE 2 | Regions showing statistically significant effects using the novel SDM-PSI approach. ROI, region of interest. comorbid disorder and the comorbidity. For instance, in the confusing in this example because we would still not know meta-regression using studies on disorder A, the coefficient whether abnormality in the region of interest (ROI) 4 is due βAB−A mixes the anomalies simulated for disorder B and to disorder A or comorbidity AB. The second limitation was comorbidity AB. We acknowledge that this limitation could a slight loss of accuracy, as shown by that some Hedges’ g are be potentially overcome by looking at the meta-regression slightly different from 0.5, and there are a few falsely positive using studies on disorder B. However, this strategy would be results. Frontiers in Psychiatry | www.frontiersin.org 5 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity FIGURE 3 | Regions showing statistically significant effects using the previous SDM-PSI approach (separate results for meta-analysis of studies on disorder A, meta-analysis of studies on disorder B, and meta-analysis of studies on disorder C). ROI, region of interest. Therefore, while the novel SDM-PSI approach does not Design invalidate the previous version, it better disentangles the specific Meta-regression of case-control VBM studies of GM volume abnormalities of comorbid disorders. abnormalities in all major mental disorders. The dependent Results were similar when we created the simulated data with variable will be the brain anatomical differences between patients double error, thus expecting Cohen’s d around 0.25 (Table 2). and controls in a voxel. The independent variables will be the That said, one can expect poorer estimations with smaller effect percentages of patients with each mental disorder and each pair sizes or in meta-analyses with few studies. Thus, we would not of potentially comorbid mental disorders. recommend the new approach when the number of studies for each regressor is too small for the expected effect sizes. Systematic Search With few exceptions (see below), we will include all whole- brain VBM studies in any mental disorder listed in the mental, Protocol for the -Meta-Analysis behavioral, or neurodevelopmental disorders classification of We pre-registered this protocol to ICD-11 (International Classification of Diseases 11th Revision). PROSPERO (CRD42021245098). Note that we will use ICD-11 to select the major disorders to Frontiers in Psychiatry | www.frontiersin.org 6 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity TABLE 2 | Results of the novel approach validation after creating the simulated data with double error. Simulated Novel SDM-PSI Previous SDM-PSI approach factual data approach Main disorder is A Main disorder is B Main disorder is C Disorder A ROI 1, g +0.25 βA = ROI 1, g +0.25 βA = ROI 1, g +0.26 βAB −B = ROI 1, +0.23 βAC−C = ROI 1, g ROI 2, g +0.23 ROI 4, +0.27 +0.25 ROI 4, g +0.11 Comorbidity AB ROI 4, g +0.25 βAB = ROI 4, g +0.25 βAB −A = ROI 4, (not studied) g +0.23 ROI 2, g +0.22 ROI 3, g +0.24 Disorder B ROI 2, g +0.25 βB = ROI 2, g +0.23 βB = ROI 2, g +0.35 βBC −C = ROI 2, ROI 3, g +0.25 ROI 3, g +0.26 ROI 3, g +0.15 g +0.29 ROI 3, g +0.25 Comorbidity BC (None) βBC = (None) (Not studied) βBC−B = ROI 2, g +0.25 ROI 3, g −0.26 Disorder C ROI 2, g +0.25 βC = ROI 2, g +0.24 βAC −A = ROI 2, βC = ROI 2, g +0.37 ROI 3, g −0.25 ROI 3, g −0.24 g +0.20 ROI 3, g −0.15 ROI 3, g −0.31 Comorbidity AC (None) βAC = (None) (not studied) (see Disorder A) “g”, average Hedges’ g of the voxels within the cluster of statistical significance; ROI, region of interest. investigate. Still, as we clarify later, we will include studies using meta-analysis may have included studies with fewer than 10 any standard clinical assessment beyond ICD (e.g., DSM). Our participants per group while we will exclude them). Suppose search will have two steps. during our search, a new meta-analysis is published after we have included a meta-analysis for the same mental disorder. In First Step: Search and Inclusion of Meta-Analyses that case, we will include both analyses (but we will include the In the first step, we will search for the most recent SDM duplicated studies only once. meta-analysis (if any) in the PubMed and Scopus databases for each mental disorder listed in ICD-11 classification (excluding Second Step: Search and Inclusion of nicotine use disorder, substance-induced specific disorders, Individual Studies neurocognitive disorders, and mental or behavioral disorders In the second step, we will search in PubMed and Scopus associated with pregnancy, childbirth, or the puerperium). The databases for the studies published since the search date of keywords will be the mental disorder (e.g., “major depression,” the selected meta-analysis. The keywords used in this search “anxiety disorders,” “bipolar disorders,” etc.) AND (“meta- will be [Title/Abstract]: (selected mental disorder) and (“VBM” analysis”) AND (“voxel-based morphometry” OR “VBM” OR OR “morphometry” OR “voxel-based” OR “voxelwise” OR “gray “gray matter” OR “grey matter”). We will first screen all the matter” OR “grey matter”). We will first screen all the results by results by the title/abstract and afterward by full-text reading. the title/abstract and then by full-text reading. The inclusion criterion will be meta-analyses of studies Inclusion criteria will be: (1) studies reporting whole-brain that employed VBM to investigate whole-brain GM volume regional GM volume differences between individuals with differences between patients with the above disorders and healthy the included mental disorders, diagnosed by standard clinical controls. The exclusion criterion will be meta-analyses from assessments (DSM or ICD), and matched healthy controls; (2) which we can include <10 studies even after adding new studies employing VBM to conduct the comparisons, (3) reporting the as described in the second step. We will select the most recent peaks of the clusters of statistically significant voxels or null meta-analysis conducted with SDM if more than one meta- findings, or availability of statistical parametric map; (4) using a analysis meets our inclusion/exclusion criteria. Suppose the constant statistical threshold throughout the whole gray matter; inclusion/exclusion criteria of a meta-analysis led to the exclusion (5) published as peer-reviewed original articles in English in of studies that we would include according to our second-step indexed journals. Exclusion criteria will be: (1) sample size study inclusion/exclusion criteria (see next). In that case, we will smaller than 10 participants in either the patient or the control look for these potentially includable studies (e.g., a meta-analysis group; (2) no case-control comparisons; (3) disorders’ subtypes may have excluded studies in children/adolescents while we with a known organic origin (e.g., pediatric autoimmune will include them). Conversely, suppose the inclusion/exclusion neuropsychiatric disorders associated with streptococci); (4) criteria of a meta-analysis led to the inclusion of studies that we coordinates of the peaks of the clusters cannot be obtained after would exclude according to our second-step inclusion/exclusion contacting the authors (unless maps are available, in which case criteria. In that case, we would exclude these studies (e.g., a we will not need the coordinates); (5) ROI or small volume Frontiers in Psychiatry | www.frontiersin.org 7 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity correction (SVC) analyses; (6) ANOVA analysis without whole- exposure for case-control studies. The “selection of the study brain t-test post hoc analyses; (7) case reports, conferences groups” evaluates the adequate definition of case and control, abstracts, editorials, non-scientific letters, and research protocols; as well as the representativity of the cases (e.g., selection of all (8) duplicated datasets (we will only include the largest sample eligible cases with the outcome of interest over a defined period, size); (9) studies that only analyze the correlation of GM volume in a defined catchment area or hospital, etc.), and the controls with any other measure or that use GM volume features to predict (community controls or hospital controls). The “comparability of diagnosis unless they specify an additional VBM case-control the groups” evaluates if researchers matched cases and controls comparison in the abstract. Special cases will be: (1) longitudinal and/or adjusted for confounders (e.g., age, sex, handedness). studies: we will only include the baseline comparison; (2) Here, statements of no differences between groups or non- studies reporting different subgroup analyses: we will include statistically significant differences are insufficient for establishing the combined analysis of all subgroups if available. Otherwise, comparability. Finally, we will not evaluate the “ascertainment we will include them as different studies if they use different of the exposure for case-control studies” because both groups control groups and provide demographic and clinical data for underwent a structural MRI in our studies. both subgroups separately. If they share the control group, we will We will also assess how much demographic or clinical data divide the control sample size between the number of subgroups. each independent study reports. Two researchers will conduct the systematic search independently, and we will resolve any discrepancies by Imputation of Missing Comorbidity Data consensus with a third researcher. Not all studies report the percentage of patients with specific comorbid disorders. For instance, in the meta-analysis of OCD Data Collection mentioned earlier, 12% of studies had not excluded comorbid For each study separately, we will extract the sample sizes, MDD but did not report how many patients had this diagnosis. demographic and clinical data, methodologic details, and the To impute these unreported data, we will assume they are missing original statistical parametric map (when available) or the at random. In other words, the proportion of patients with coordinates and t-values (or equivalent statistics when available) no information about comorbid MDD should follow a similar of the peaks of the clusters of statistically significant voxels (or distribution than in studies reporting this information. null findings). The proportion of patients with a comorbid disorder likely Demographic data will include age distribution (mean and follows a zero-inflated distribution. For example, the percentage standard deviation) and percentage of males and females. of patients with MDD might follow some statistical distribution, Clinical data will consist of the percentages of patients with but this distribution probably has excess zeroes due to the studies different mental disorders, the percentage of patients receiving that excluded patients with MDD. However, as fitting zero- each medication group (antipsychotics, antidepressants, inflated distributions with the small data available would be anxiolytics -other than hypnotic-, mood stabilizers, and unfeasible, we will use a more straightforward, distribution-free stimulants), the severity of the primary disorder assessed by approach. Specifically, the imputation will consist of assigning standard measures [e.g., Hamilton Depression Rating Scale to each study not reporting the proportion of patients with (HDRS) (55)], the age of onset or illness duration of the primary comorbid MDD, the proportion from another random OCD disorder (mean and standard deviation), and the different study. Thus, for example, we may estimate that the missing subtypes for the primary disorder as reported in the included proportion in a given study is the same as in the study by van meta-analyses (e.g., inattentive, hyperactive, or combined type den Heuvel et al. (57), or the same as in the study by Pujol et al. for ADHD, type I or II for BD, etc.). (58), or the same as in any other random study (including studies Methodological details will include the pre-processing that excluded MDD). analysis software (e.g., FSL, SPM) and their version, stereotactic We will repeat these imputations 50 times. We want to remark space (e.g., MNI, Talairach space, or MNI coordinates converted that his number is commonly considered more than adequate for to Talairach using the old Brett transform), and the statistical multiple imputation (Rubin recommended 3 to 10 imputations significance threshold (e.g., FWER < 0.05). For studies reporting (59). Additionally, we have checked that the histogram of the peaks obtained using two or more whole-brain statistical imputed proportions of patients with MDD is similar to the significance levels (e.g., uncorrected p < 0.001 and corrected histogram of known proportions of patients with MDD after only FWER p < 0.05), we will include all peaks obtained using the ten imputations and nearly identical after twenty imputations. less conservative threshold. We will also record the information These imputations will be conducted separately for each required for the quality assessment (see below). comorbid disorder. Thus, for instance, in studies with Given the magnitude of data, we will store them in a database OCD, we will impute comorbid MDD and comorbid anxiety and a well-organized file system with automatic daily backups. disorders separately. Quality Assessment Statistical Analyses We will use the Newcastle-Ottawa Scale (NOS) for case-control We will carry out the data pre-processing and the statistical studies to assess each study’s quality (56). The NOS assesses three analysis with the SDM-PSI 6.21 software (https://www. characteristics of the studies: the selection of the study groups, sdmproject.com/) (11, 40–42). We will conduct two independent the comparability of the groups, and the ascertainment of the analyses, one for adults and one for children/adolescents. Frontiers in Psychiatry | www.frontiersin.org 8 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity The pre-processing of statistical parametric maps is the Conventionally, I 2 values above 50% are interpreted as an straightforward conversion into images of effect sizes. For studies indication of significant heterogeneity (61). with only peak information available, the pre-processing consists of estimating the 3D images of the lower and higher bounds of DISCUSSION potential effect sizes. The software will later impute the effect sizes multiple times within these bounds. This paper first presents and validates a neuroimaging meta- We will include all major mental disorders in one single linear analytic approach that focuses on comorbidities in mental model described earlier. Then, we will estimate the effects related disorders. Then, using simulations, we show that the new method to each disorder and comorbidity by testing different contrasts may detect all GM volume differences with the correct effect within the model. The steps will be those of standard SDM-PSI size and without falsely positive findings. Finally, we describe the (11, 40–42) unless otherwise specified: protocol for a meta-analysis of all major mental disorders and their comorbidities, separately for adult and pediatric groups. We 1. Estimation of the 3D images of maximum likely effect sizes for will also assess the potentially confounding effects of medication, each model coefficient. age of onset or illness duration and the symptom severity 2. Multiple imputation of the study 3D images of effect size of the primary disorder, and the moderator effects of sex on adding spatially realistic noise to the expected effect size GM volume. according to the estimated distribution within the bounds. We broadly expect some findings according to previous Following SDM-PSI default parameters, we will conduct this literature, though a significant part of these findings might process 50 times, resulting in 50 imputed datasets covering the change due to the improvements of the new approach. For imputations’ uncertainty. example, for chronic schizophrenia, previous meta-analyses have 3. Separately for each imputation dataset, random-effects meta- detected reduced GM volume in the bilateral insula/ superior linear model. The dependent variable will be the effect size of temporal gyrus, dorsal, and rostral anterior cingulate cortex the voxel. The independent variables will be the percentages (ACC) / medial frontal gyrus, and the thalamus (62, 63). of patients with each mental disorder and the percentages Similarly, for the first episode of psychosis, we expect a reduced of patients with each pair of potentially comorbid disorders GM volume in the right dorsal ACC and the right posterior (as far as they involve at least ten studies). In case that at insula/superior temporal gyrus (35, 62, 63). In OCD, previous least studies reported the percentage of patients with more meta-analyses have detected both increased GM volumes, mainly than three or more comorbid disorders (e.g., OCD, MDD, located in subcortical regions (e.g., bilateral putamen, left and anxiety), we will also include these percentages as an cerebellum, and left hippocampus), and decreased GM volumes, independent variable in the model. located primarily on prefrontal and cingulate areas (e.g., bilateral 4. Using Rubin’s rules, combination of the meta-analytic 3D ACC/ventromedial prefrontal cortex, bilateral inferior frontal images of effect size from the different imputation datasets. gyrus) (9, 33, 34). For BD and MDD, previous meta-analyses To assess the statistical significance, the software converts have detected a commonly reduced GM volume in the medial the 3D image of z-values into a 3D image of threshold- prefrontal system and ACC, regions strongly implicated in free cluster enhancement (TFCE) statistics and finds the mood regulation (64). However, smaller hippocampus and p-value of the TFCE statistics using a Freedman-Lane- parahippocampal gyrus volumes have been more reported in based permutation test (60). We will consider statistically MDD (26, 37, 65–69). For ADHD, previous meta-analyses have significant those voxels with family-wise error-rate (FWER, reported a reduced GM volume in ventromedial orbitofrontal i.e., corrected p-values) <0.05. For comprehensive reporting, cortex/ventromedial prefrontal cortex/rostral ACC, and the right we will also publish supplementary results using significance basal ganglia/anterior and posterior insula (34, 70). For ASD, thresholds of FWER < 0.01, uncorrected p < 0.001, and previous meta-analyses have reported reduced GM volume in uncorrected p < 0.005. dorsal ACC/dorsomedial prefrontal cortex, left cerebellum, and In the complementary analyses, we will add additional increased GM volume in the left middle superior anterior lobe independent variables to the linear model to explore potential and middle frontal gyrus (33, 70). For anxiety disorders, previous interactions with mean sample age and percentage of males and meta-analyses have reported a reduced GM volume in the right control the medication’s possible confounding effects. Depending ventral ACC and inferior frontal gyrus (71). These areas have on our final dataset, we will try to assess the effects of the age been primarily reported in panic disorder, together with the of onset or illness duration and the severity of the primary prefrontal cortex (72, 73). However, some of these findings may disorder, as reported in the original studies. Furthermore, we will have been influenced by comorbidity. Thus, we need to conduct perform a subgroup analysis excluding those disorders for which the new meta-analysis to know which results remain, which do we cannot safely collect whether they are comorbidities for other not, and which had not been detected due to the confounding disorders (e.g., we expect that we will not have information on effects of comorbidities. comorbid personality disorders in many studies). We hope that this improved atlas of the anatomical We will use the I 2 statistic to quantify heterogeneity localization of the brain abnormalities associated with each and conduct meta-regression by the standard error [similar mental disorder will help improve our understanding of their to an Egger-test (41)] to detect potential publication bias. physio-pathological processes. In addition, we hope that this atlas Frontiers in Psychiatry | www.frontiersin.org 9 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity could be the base for developing MRI-based diagnostic tools that drawback may be the heterogeneity across studies. Considering help earlier diagnoses and, therefore, more targeted treatments. comorbidity, age, sex, and medication, we aim to explain For instance, when complex psychotic symptoms hamper the the heterogeneity more than previous meta-analyses, but we assessment of other symptoms needed for the diagnosis, the anticipate that there will still be unexplained heterogeneity. A “opinion” of an MRI-based diagnostic tool could provide timely significant source of heterogeneity may be due to differences extra information to establish an affective vs. non-affective in the MRI equipment (e.g., varying field strength or head diagnosis and thus a more focused treatment earlier. Or similarly, coils) and acquisition parameters (76) and the VBM processing in depressed individuals at risk of manic shift, the “opinion” of an method employed by the different studies, such as software and MRI-based diagnostic tool may help the clinician better evaluate version, normalization, statistical correction, or the size of the the probability of bipolar vs. unipolar disorder and thus design smoothing kernel (19). Another relevant source of heterogeneity a more personalized preventive strategy. Indeed, we have found may be different subject-specific artifacts such as head motion elsewhere that diagnostic labels are among the variables that best (77), body mass index (78), drops in signal-to-noise ratio due to predict the future recurrence of an episodic disorder. susceptibility artifacts, the symptom severity of the disorders, or Better knowledge about the disorder-specific abnormalities the different phases present in some disorders (e.g., the various could also increase the efficacy of therapies to modify specific episodes in BD) (79, 80). Also, we will only study those mental brain regions’ activity [e.g., deep brain stimulation or non- disorders for which a meta-analysis has already been published invasive brain stimulations such as repetitive transcranial and examined by at least ten studies. Last but not least, we magnetic stimulation (74)]. Improved knowledge about the must highlight that even when, for simplicity, we talk about GM spatial distribution of these abnormalities may help localize the volume abnormalities, we should more appropriately refer to brain targets better. Indeed, previous studies have already shown differences in T1-MRI signal, given that the acquired MRI data how the efficacy of such therapies depends on the exact position are not a direct measure of brain structure (81). of the brain target (75). We acknowledge that the novel approach has several limitations. The first relates to the debatable nosology of current AUTHOR CONTRIBUTIONS mental disorders, based on clinical consensus rather than known LF, AA-E, and JR conceived the study, modified the SDM- biological underpinnings. We know, for example, that major PSI software, and validated the novel approach. All authors psychiatric disorders share some genetic risk factors, and there participated in the redaction of the manuscript, including minor are high percentages of comorbidity and diagnostic change. or major modifications of the protocol. However, this does not mean that there are no disorder-specific brain correlates. As noted above, diagnostic labels are among the best predictors of future outcomes, highlighting their clinical relevance. The second relates to the commonly poor reporting of FUNDING some comorbid disorders in the literature and the subthreshold This work was supported by several grants from the Instituto disorder-specific symptoms that we will not consider due to the de Salud Carlos III - Subdirección General de Evaluación y complexity of the analysis and the expected large amount of Fomento de la Investigación: Research Project Grant (PI19/00394 missing data. For instance, studies may check for schizophrenia to JR), Miguel Servet Research Contract (CPII19/0009 to JR), but not for personality disorders. For this reason, we will conduct and the PFIS contract (FI20/00047 to LF). NV thanks the subgroup analyses excluding the disorders under-reported as BITRECS project, which has received funding from the European comorbidities. A third limitation is that, as stated earlier, some Union’s Horizon 2020 research and innovation programme studies may not report the proportion of patients with specific under the Marie Skłodowska-Curie Grant Agreement No 754550 comorbidities. Thus, we will have to use multiple imputation. We and from La Caixa Foundation (ID 100010434), under the will use a simplistic imputation algorithm without considering agreement LCF/PR/GN18/50310006. whether the proportion of patients with a given comorbid disorder may depend on the age or symptom severity of the sample. We preferred this simple algorithm because we anticipate ACKNOWLEDGMENTS that we would not be able to collect the necessary data for robustly using more complex imputation algorithms. Another We thank all the participants of the studies that we will include limitation is that, as in any other meta-analysis, a potential in the meta-analysis. REFERENCES neurofunctional model. Neurosci Biobehav Rev. (2014) 47:260–80. doi: 10.1016/j.neubiorev.2014.08.003 1. DeLisi LE, Goldin LR, Hamovit JR, Maxwell ME, Kurtz D, Nurnberger 3. Brühl AB, Hänggi J, Baur V, Rufer M, Delsignore A, Weidt S, JI, et al. Cerebral ventricular enlargement as a possible genetic marker for et al. Increased cortical thickness in a frontoparietal network schizophrenia. Psychopharmacol Bull. (1985) 21:365–7. in social anxiety disorder: increased cortical thickness in 2. Brühl AB, Delsignore A, Komossa K, Weidt S. Neuroimaging in SAD. Hum Brain Mapp. (2014) 35:2966–77. doi: 10.1002/hbm. social anxiety disorder—a meta-analytic review resulting in a new 22378 Frontiers in Psychiatry | www.frontiersin.org 10 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity 4. Irle E. Reduced amygdalar and hippocampal size in adults with coordinate-based meta-analysis. Neurosci Biobehav Rev. (2021) 123:83–103. generalized social phobia. J Psychiatry Neurosci. (2010) 35:126–31. doi: 10.1016/j.neubiorev.2021.01.010 doi: 10.1503/jpn.090041 22. Schmaal L, Hibar DP, Sämann PG, Hall GB, Baune BT, Jahanshad N, et al. 5. Irle E, Barke A, Lange C, Ruhleder M. Parietal abnormalities are related to Cortical abnormalities in adults and adolescents with major depression avoidance in social anxiety disorder: a study using voxel-based morphometry based on brain scans from 20 cohorts worldwide in the ENIGMA Major and manual volumetry. Psychiatry Res Neuroimaging. (2014) 224:175–83. Depressive Disorder Working Group. Mol Psychiatry. (2017) 22:900–9. doi: 10.1016/j.pscychresns.2014.08.013 doi: 10.1038/mp.2016.60 6. Bas-Hoogendam JM, van Steenbergen H, Nienke Pannekoek J, Fouche J-P, 23. Koutsouleris N, Davatzikos C, Borgwardt S, Gaser C, Bottlender R, Lochner C, Hattingh CJ, et al. Voxel-based morphometry multi-center mega- Frodl T, et al. Accelerated brain aging in schizophrenia and beyond: a analysis of brain structure in social anxiety disorder. Neuroimage Clin. (2017) neuroanatomical marker of psychiatric disorders. Schizophr Bull. (2014) 16:678–88. doi: 10.1016/j.nicl.2017.08.001 40:1140–53. doi: 10.1093/schbul/sbt142 7. Rotge J-Y, Langbour N, Guehl D, Bioulac B, Jaafari N, Allard M, et al. 24. Bora E, Fornito A, Yücel M, Pantelis C. The effects of gender on grey matter Gray matter alterations in obsessive–compulsive disorder: an anatomic abnormalities in major psychoses: a comparative voxelwise meta-analysis likelihood estimation meta-analysis. Neuropsychopharmacol. (2010) 35:686– of schizophrenia and bipolar disorder. Psychol Med. (2012) 42:295–307. 91. doi: 10.1038/npp.2009.175 doi: 10.1017/S0033291711001450 8. Tang W, Zhu Q, Gong X, Zhu C, Wang Y, Chen S. Cortico-striato-thalamo- 25. Torres US, Portela-Oliveira E, Borgwardt S, Busatto GF. Structural brain cortical circuit abnormalities in obsessive-compulsive disorder: a voxel-based changes associated with antipsychotic treatment in schizophrenia as revealed morphometric and fMRI study of the whole brain. Behav Brain Res. (2016) by voxel-based morphometric MRI: an activation likelihood estimation meta- 313:17–22. doi: 10.1016/j.bbr.2016.07.004 analysis. BMC Psychiatry. (2013) 13:342. doi: 10.1186/1471-244X-13-342 9. Picó-Pérez M, Moreira PS, de Melo Ferreira V, Radua J, Mataix- 26. Zhao Y-J, Du M-Y, Huang X-Q, Lui S, Chen Z-Q, Liu J, et al. Cols D, Sousa N, et al. Modality-specific overlaps in brain structure Brain grey matter abnormalities in medication-free patients with major and function in obsessive-compulsive disorder: multimodal meta-analysis depressive disorder: a meta-analysis. Psychol Med. (2014) 44:2927–37. of case-control MRI studies. Neurosci Biobehav Rev. (2020) 112:83–94. doi: 10.1017/S0033291714000518 doi: 10.1016/j.neubiorev.2020.01.033 27. Fusar-Poli P, Salazar de Pablo G, Correll CU, Meyer-Lindenberg A, 10. de Wit SJ, Alonso P. Multicenter voxel-based morphometry mega-analysis Millan MJ, Borgwardt S, et al. Prevention of psychosis: advances in of structural brain scans in obsessive-compulsive disorder. Am J Psychiatry. detection, prognosis, and intervention. JAMA Psychiatry. (2020) 77:755. (2014) 171:340–9. doi: 10.1176/appi.ajp.2013.13040574 doi: 10.1001/jamapsychiatry.2019.4779 11. Radua J, Mataix-Cols D. Voxel-wise meta-analysis of grey matter changes 28. Nakao T, Radua J, Rubia K, Mataix-Cols D. Gray matter volume in obsessive–compulsive disorder. Br J Psychiatry. (2009) 195:393–402. abnormalities in ADHD: voxel-based meta-analysis exploring the doi: 10.1192/bjp.bp.108.055046 effects of age and stimulant medication. AJP. (2011) 168:1154–63. 12. Eng GK, Sim K, Chen S-HA. Meta-analytic investigations of structural grey doi: 10.1176/appi.ajp.2011.11020281 matter, executive domain-related functional activations, and white matter 29. Fusar-Poli P, Solmi M, Brondino N, Davies C, Chae C, Politi P, et al. diffusivity in obsessive compulsive disorder: an integrative review. Neurosci Transdiagnostic psychiatry: a systematic review. World Psychiatry. (2019) Biobehav Rev. (2015) 52:233–57. doi: 10.1016/j.neubiorev.2015.03.002 18:192–207. doi: 10.1002/wps.20631 13. Hoexter MQ, de Souza Duran FL, D’Alcante CC, Dougherty DD, Shavitt RG, 30. Kessler RC, Chiu WT, Demler O, Walters EE. Prevalence, Severity, Lopes AC, et al. Gray matter volumes in obsessive-compulsive disorder before and Comorbidity of Twelve-month DSM-IV disorders in the national and after fluoxetine or cognitive-behavior therapy: a randomized clinical trial. comorbidity survey replication (NCS-R). Arch Gen Psychiatry. (2005) 62:617– Neuropsychopharmacology. (2012) 37:734–45. doi: 10.1038/npp.2011.250 27. doi: 10.1001/archpsyc.62.6.617 14. Togao O, Yoshiura T, Nakao T, Nabeyama M, Sanematsu H, Nakagawa A, 31. Caspi A, Houts RM, Ambler A, Danese A, Elliott ML, Hariri A, et al. et al. Regional gray and white matter volume abnormalities in obsessive– Longitudinal assessment of mental health disorders and comorbidities across compulsive disorder: a voxel-based morphometry study. Psychiatry Res 4 decades among participants in the dunedin birth cohort study. JAMA Netw Neuroimaging. (2010) 184:29–37. doi: 10.1016/j.pscychresns.2010.06.011 Open. (2020) 3:e203221. doi: 10.1001/jamanetworkopen.2020.3221 15. Szeszko PR, Christian C, MacMaster F, Lencz T, Mirza Y, Taormina SP, 32. Perugi G, Barbuti M. There are no patients without et al. Gray matter structural alterations in psychotropic drug-naive pediatric comorbidity. Eur Neuropsychopharmacol. (2021) 50:104–6. obsessive-compulsive disorder: An optimized voxel-based morphometry doi: 10.1016/j.euroneuro.2021.05.002 study. AJP. (2008) 165:1299–307. doi: 10.1176/appi.ajp.2008.08010033 33. Carlisi CO, Norman LJ, Lukito SS, Radua J, Mataix-Cols D, 16. Christian CJ, Lencz T, Robinson DG, Burdick KE, Ashtari M, Malhotra AK, Rubia K. Comparative multimodal meta-analysis of structural and et al. Gray matter structural alterations in obsessive–compulsive disorder: functional brain abnormalities in autism spectrum disorder and relationship to neuropsychological functions. Psychiatry Res Neuroimaging. obsessive-compulsive disorder. Biol Psychiatry. (2017) 82:83–102. (2008) 164:123–31. doi: 10.1016/j.pscychresns.2008.03.005 doi: 10.1016/j.biopsych.2016.10.006 17. Pardoe H, Pell GS, Abbott DF, Berg AT, Jackson GD. Multi-site 34. Norman LJ, Carlisi C, Lukito S, Hart H, Mataix-Cols D, Radua J, voxel-based morphometry: methods and a feasibility demonstration et al. Structural and functional brain abnormalities in attention- with childhood absence epilepsy. Neuroimage. (2008) 42:611–6. deficit/hyperactivity disorder and obsessive-compulsive disorder: doi: 10.1016/j.neuroimage.2008.05.007 a comparative meta-analysis. JAMA Psychiatry. (2016) 73:815. 18. Focke NK, Helms G, Kaspar S, Diederich C, Tóth V, Dechent P, et al. Multi- doi: 10.1001/jamapsychiatry.2016.0700 site voxel-based morphometry — not quite there yet. Neuroimage. (2011) 35. Radua J, Borgwardt S, Crescini A, Mataix-Cols D, Meyer-Lindenberg 56:1164–70. doi: 10.1016/j.neuroimage.2011.02.029 A, McGuire PK, et al. Multimodal meta-analysis of structural and 19. Henley SMD, Ridgway GR, Scahill RI, Klöppel S, Tabrizi SJ, Fox NC, functional brain changes in first episode psychosis and the effects et al. Pitfalls in the use of voxel-based morphometry as a biomarker: of antipsychotic medication. Neurosci Biobehav Rev. (2012) 36:2325–33. examples from Huntington disease. AJNR Am J Neuroradiol. (2010) 31:711–9. doi: 10.1016/j.neubiorev.2012.07.012 doi: 10.3174/ajnr.A1939 36. Radua J, van den Heuvel OA, Surguladze S, Mataix-Cols D. Meta-analytical 20. Truong W, Minuzzi L, Soares CN, Frey BN, Evans AC, MacQueen comparison of voxel-based morphometry studies in obsessive-compulsive GM, et al. Changes in cortical thickness across the lifespan in major disorder vs other anxiety disorders. Arch Gen Psychiatry. (2010) 67:701–11. depressive disorder. Psychiatry Res Neuroimaging. (2013) 214:204–11. doi: 10.1001/archgenpsychiatry.2010.70 doi: 10.1016/j.pscychresns.2013.09.003 37. Wise T, Radua J, Via E, Cardoner N, Abe O, Adams TM, et al. Common and 21. Liloia D, Brasso C, Cauda F, Mancuso L, Nani A, Manuello J, et al. distinct patterns of grey-matter volume alteration in major depression and Updating and characterizing neuroanatomical markers in high-risk subjects, bipolar disorder: evidence from voxel-based meta-analysis. Mol Psychiatry. recently diagnosed and chronic patients with schizophrenia: a revised (2017) 22:1455–63. doi: 10.1038/mp.2016.72 Frontiers in Psychiatry | www.frontiersin.org 11 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity 38. Opel N, Goltermann J, Hermesdorf M, Berger K, Baune BT, Dannlowski 56. Wells G, Shea B, O’Connell J. The Newcastle-Ottawa Scale (NOS) for Assessing U. Cross-disorder analysis of brain structural abnormalities in six major The Quality of Nonrandomised Studies in Meta-analyses. Ottawa Health psychiatric disorders: a secondary analysis of mega- and meta-analytical Research Institute Web site (2014). findings from the ENIGMA consortium. Biol Psychiatry. (2020) 88:678–86. 57. van den Heuvel OA, Remijnse PL, Mataix-Cols D, Vrenken H, Groenewegen doi: 10.1016/j.biopsych.2020.04.027 HJ, Uylings HBM, et al. The major symptom dimensions of obsessive- 39. Arnone D, Cavanagh J, Gerber D, Lawrie SM, Ebmeier KP, McIntosh compulsive disorder are mediated by partially distinct neural systems. Brain. AM. Magnetic resonance imaging studies in bipolar disorder and (2008) 132:853–68. doi: 10.1093/brain/awn267 schizophrenia: meta-analysis. Br J Psychiatry. (2009) 195:194–201. 58. Pujol J, Soriano-Mas C, Alonso P, Cardoner N, Menchón JM, Deus J, et al. doi: 10.1192/bjp.bp.108.059717 Mapping structural brain alterations in obsessive-compulsive disorder. Arch 40. Radua J, Mataix-Cols D, Phillips ML, El-Hage W, Kronhaus DM, Cardoner N, Gen Psychiatry. (2004) 61:720. doi: 10.1001/archpsyc.61.7.720 et al. A new meta-analytic method for neuroimaging studies that combines 59. Rubin DB. Multiple Imputation for Nonresponse in Surveys. New York: John reported peak coordinates and statistical parametric maps. Eur Psychiatry. Wiley & Sons Inc. (1987). doi: 10.1002/9780470316696 (2012) 27:605–11. doi: 10.1016/j.eurpsy.2011.04.001 60. Winkler AM, Ridgway GR, Webster MA, Smith SM, Nichols TE. Permutation 41. Albajes-Eizagirre A, Solanes A, Vieta E, Radua J. Voxel-based meta-analysis inference for the general linear model. NeuroImage. (2014) 92:381–97. via permutation of subject images (PSI): theory and implementation for SDM. doi: 10.1016/j.neuroimage.2014.01.060 Neuroimage. (2019) 186:174–84. doi: 10.1016/j.neuroimage.2018.10.077 61. Ioannidis JPA, Patsopoulos NA, Evangelou E. Uncertainty in 42. Albajes-Eizagirre A, Solanes A, Fullana MA, Ioannidis JPA, Fusar-Poli P, heterogeneity estimates in meta-analyses. BMJ. (2007) 335:914–6. Torrent C, et al. Meta-analysis of voxel-based neuroimaging studies using doi: 10.1136/bmj.39343.408449.80 Seed-based d Mapping with Permutation of Subject Images (SDM-PSI). J Vis 62. Bora E, Fornito A, Radua J, Walterfang M, Seal M, Wood SJ, et al. Exp. (2019) 153. doi: 10.3791/59841 Neuroanatomical abnormalities in schizophrenia: a multimodal voxelwise 43. Webler RD, Berg H, Fhong K, Tuominen L, Holt DJ, Morey RA, meta-analysis and meta-regression analysis. Schizophr Res. (2011) 127:46–57. et al. The neurobiology of human fear generalization: meta-analysis doi: 10.1016/j.schres.2010.12.020 and working neural model. Neurosci Biobehav Rev. (2021) 128:421–36. 63. Ellison-Wright I, Glahn DC, Laird AR, Thelen SM, Bullmore E. The doi: 10.1016/j.neubiorev.2021.06.035 anatomy of first-episode and chronic schizophrenia: an anatomical 44. Pezzoli S, Sánchez-Valle R, Solanes A, Kempton MJ, Bandmann O, Shin likelihood estimation meta-analysis. AJP. (2008) 165:1015–23. JI, et al. Neuroanatomical and cognitive correlates of visual hallucinations doi: 10.1176/appi.ajp.2008.07101562 in Parkinson’s disease and dementia with Lewy bodies: voxel-based 64. Rive MM, van Rooijen G, Veltman DJ, Phillips ML, Schene AH, Ruhé HG. morphometry and neuropsychological meta-analysis. Neurosci Biobehav Rev. Neural correlates of dysfunctional emotion regulation in major depressive (2021) 128:367–82. doi: 10.1016/j.neubiorev.2021.06.030 disorder. A systematic review of neuroimaging studies. Neurosci Biobehav Rev. 45. Del Fabro L, Schmidt A, Fortea L, Delvecchio G, D’Agostino A, Radua J, et al. (2013) 37:2529–53. doi: 10.1016/j.neubiorev.2013.07.018 Functional brain network dysfunctions in subjects at high-risk for psychosis: a 65. Kempton MJ, Salvador Z, Munafò MR, Geddes JR, Simmons A, Frangou S, meta-analysis of resting-state functional connectivity. Neurosci Biobehav Rev. et al. Structural neuroimaging studies in major depressive disorder: meta- (2021) 128:90–101. doi: 10.1016/j.neubiorev.2021.06.020 analysis and comparison with bipolar disorder. Arch Gen Psychiatry. (2011) 46. Jauhar S, Fortea L, Solanes A, Albajes-Eizagirre A, McKenna PJ, Radua J. Brain 68:675. doi: 10.1001/archgenpsychiatry.2011.60 activations associated with anticipation and delivery of monetary reward: 66. Arnone D, McIntosh AM, Ebmeier KP, Munafò MR, Anderson IM. a systematic review and meta-analysis of fMRI studies. PLoS ONE. (2021) Magnetic resonance imaging studies in unipolar depression: systematic review 16:e0255292. doi: 10.1371/journal.pone.0255292 and meta-regression analyses. Eur Neuropsychopharmacol. (2012) 22:1–16. 47. Fortea A, Batalla A, Radua J, van Eijndhoven P, Baeza I, Albajes-Eizagirre doi: 10.1016/j.euroneuro.2011.05.003 A, et al. Cortical gray matter reduction precedes transition to psychosis in 67. Arnone D, Job D, Selvaraj S, Abe O, Amico F, Cheng Y, et al. individuals at clinical high-risk for psychosis: a voxel-based meta-analysis. Computational meta-analysis of statistical parametric maps in major Schizophr Res. (2021) 232:98–106. doi: 10.1016/j.schres.2021.05.008 depression: computational meta-analysis of parametric maps in major 48. Gharehgazlou A, Freitas C, Ameis SH, Taylor MJ, Lerch JP, Radua J, et al. depression. Hum Brain Mapp. (2016) 37:1393–404. doi: 10.1002/hbm.23108 Cortical gyrification morphology in individuals with ASD and ADHD across 68. Selvaraj S, Arnone D, Job D, Stanfield A, Farrow TF, Nugent AC, et al. the lifespan: a systematic review and meta-analysis. Cerebral Cortex. (2021) Grey matter differences in bipolar disorder: a meta-analysis of voxel-based 31:2653–69. doi: 10.1093/cercor/bhaa381 morphometry studies: grey matter differences in bipolar disorder. Bipolar 49. Sheng L, Zhao P, Ma H, Radua J, Yi Z, Shi Y, et al. Cortical thickness in Disord. (2012) 14:135–45. doi: 10.1111/j.1399-5618.2012.01000.x Parkinson’s disease: a coordinate-based meta-analysis. Aging. (2021) 13:4007– 69. Pezzoli S, Emsell L, Yip SW, Dima D, Giannakopoulos P, Zarei M, 23. doi: 10.18632/aging.202368 et al. Meta-analysis of regional white matter volume in bipolar disorder 50. Soldevila-Matías P, Schoretsanitis G, Tordesillas-Gutierrez D, Cuesta MJ, de with replication in an independent sample using coordinates, T-maps, Filippis R, Ayesa-Arriola R, et al. Neuroimaging correlates of insight in non- and individual MRI data. Neurosci Biobehav Rev. (2018) 84:162–70. affective psychosis: a systematic review and meta-analysis. Rev Psiquiatr Salud doi: 10.1016/j.neubiorev.2017.11.005 Mental. (in press). doi: 10.1016/j.rpsm.2021.07.001 70. Lukito S, Norman L, Carlisi C, Radua J, Hart H, Simonoff E, et al. 51. Dugré JR, Radua J, Carignan-Allard M, Dumais A, Rubia K, Potvin S. Comparative meta-analyses of brain structural and functional abnormalities Neurofunctional abnormalities in antisocial spectrum: a meta-analysis of during cognitive control in attention-deficit/hyperactivity disorder fMRI studies on Five distinct neurocognitive research domains. Neurosci and autism spectrum disorder. Psychol Med. (2020) 50:894–919. Biobehav Rev. (2020) 119:168–83. doi: 10.1016/j.neubiorev.2020.09.013 doi: 10.1017/S0033291720000574 52. Albajes-Eizagirre A, Radua J. What do results from coordinate- 71. Shang J, Fu Y, Ren Z, Zhang T, Du M, Gong Q, et al. The common based meta-analyses tell us? NeuroImage. (2018) 176:550–3. traits of the ACC and PFC in anxiety disorders in the DSM-5: meta- doi: 10.1016/j.neuroimage.2018.04.065 analysis of voxel-based morphometry studies. PLoS ONE. (2014) 9:e93432. 53. Radua J, Rubia K, Canales-Rodríguez EJ, Pomarol-Clotet E, Fusar- doi: 10.1371/journal.pone.0093432 Poli P, Mataix-Cols D. Anisotropic kernels for coordinate-based 72. Wang X, Cheng B, Wang S, Lu F, Luo Y, Long X, et al. Distinct grey matter meta-analyses of neuroimaging studies. Front Psychiatry. (2014) 5:13. volume alterations in adult patients with panic disorder and social anxiety doi: 10.3389/fpsyt.2014.00013 disorder: a systematic review and voxel-based morphometry meta-analysis. 54. Smith S, Nichols T. Threshold-free cluster enhancement: addressing problems J Affect Disord. (2021) 281:805–23. doi: 10.1016/j.jad.2020.11.057 of smoothing, threshold dependence and localisation in cluster inference. 73. Wu Y, Zhong Y, Ma Z, Lu X, Zhang N, Fox PT, et al. Gray matter Neuroimage. (2009) 44:83–98. doi: 10.1016/j.neuroimage.2008.03.061 changes in panic disorder: a voxel-based meta-analysis and meta-analytic 55. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. connectivity modeling. Psychiatry Res Neuroimaging. (2018) 282:82–9. (1960) 23:56–62. doi: 10.1136/jnnp.23.1.56 doi: 10.1016/j.pscychresns.2018.09.009 Frontiers in Psychiatry | www.frontiersin.org 12 January 2022 | Volume 12 | Article 807839 Fortea et al. Focusing on Comorbidity 74. Padberg F, Bulubas L, Mizutani-Tiebel Y, Burkhardt G, Kranz GS, Koutsouleris Conflict of Interest: NV has received financial support for CME activities and N, et al. The intervention, the patient and the illness – personalizing non- travel funds from the following entities (unrelated to the present work): Angelini, invasive brain stimulation in psychiatry. Exp Neurol. (2021) 341:113713. Janssen-Cilag, Lundbeck, Otsuka. EVia is a co-investigator in a Janssen-Cilag, S.A., doi: 10.1016/j.expneurol.2021.113713 clinical trial with the esketamine molecule (unrelated to the present work). KR 75. Puigdemont D, Pérez-Egea R, Portella MJ, Molet J, de Diego-Adeliño received a grant from TAKEDA pharmaceutical for another project and consulting J, Gironell A, et al. Deep brain stimulation of the subcallosal cingulate fees from Lundbeck and Supernus (unrelated to the present work). gyrus: further evidence in treatment-resistant major depression. Int J Neuropsychopharmacol. (2012) 15:121–33. doi: 10.1017/S146114571100 The remaining authors declare that the research was conducted in the absence of 1088 any commercial or financial relationships that could be construed as a potential 76. Chiò A, Pagani M, Agosta F, Calvo A, Cistaro A, Filippi M. Neuroimaging conflict of interest. in amyotrophic lateral sclerosis: insights into structural and functional changes. Lancet Neurol. (2014) 13:1228–40. doi: 10.1016/S1474-4422(14)7 Publisher’s Note: All claims expressed in this article are solely those of the authors 0167-X and do not necessarily represent those of their affiliated organizations, or those of 77. Zaitsev M, Maclaren J, Herbst M. Motion artifacts in MRI: a complex problem the publisher, the editors and the reviewers. Any product that may be evaluated in with many partial solutions: motion artifacts and correction. J Magn Reson this article, or claim that may be made by its manufacturer, is not guaranteed or Imaging. (2015) 42:887–901. doi: 10.1002/jmri.24850 endorsed by the publisher. 78. McWhinney SR, Abé C, Alda M, Benedetti F, Bøen E, del Mar Bonnin C, et al. Association between body mass index and subcortical brain volumes in bipolar disorders–ENIGMA study in 2735 individuals. Mol Psychiatry. (in Copyright © 2022 Fortea, Albajes-Eizagirre, Yao, Soler, Verdolini, Hauson, Fortea, press). doi: 10.1038/s41380-021-01098-x Madero, Solanes, Wollman, Serra-Blasco, Wise, Lukito, Picó-Pérez, Carlisi, Zhang, 79. Carvalho AF, Firth J, Vieta E. Bipolar disorder. N Engl J Med. (2020) 383:58– Pan, Farré-Colomés, Arnone, Kempton, Soriano-Mas, Rubia, Norman, Fusar-Poli, 66. doi: 10.1056/NEJMra1906193 Mataix-Cols, Valentí, Via, Cardoner, Solmi, Shin, Vieta and Radua. This is an open- 80. Burdick KE, Millett CE. Cognitive heterogeneity is a key predictor access article distributed under the terms of the Creative Commons Attribution of differential functional outcome in patients with bipolar disorder. License (CC BY). The use, distribution or reproduction in other forums is permitted, Eur Neuropsychopharmacol. (2021) 53:4–6. doi: 10.1016/j.euroneuro.2021. provided the original author(s) and the copyright owner(s) are credited and that the 06.008 original publication in this journal is cited, in accordance with accepted academic 81. Weinberger DR, Radulescu E. Structural magnetic resonance imaging all over practice. No use, distribution or reproduction is permitted which does not comply again. JAMA Psychiatry. (2021) 78:11. doi: 10.1001/jamapsychiatry.2020.1941 with these terms. Frontiers in Psychiatry | www.frontiersin.org 13 January 2022 | Volume 12 | Article 807839