Research JAMA Psychiatry | Original Investigation Association of Adverse Outcomes With Emotion Processing and Its Neural Substrate in Individuals at Clinical High Risk for Psychosis Gemma Modinos, PhD; Matthew J. Kempton, PhD; Stefania Tognin, PhD; Maria Calem, PhD; Lilla Porffy, MSc; Mathilde Antoniades, PhD; Ava Mason, MSc; Matilda Azis, PhD; Paul Allen, PhD; Barnaby Nelson, PhD; Patrick McGorry, MD, PhD; Christos Pantelis, MD; Anita Riecher-Rössler, MD; Stefan Borgwardt, MD; Rodrigo Bressan, MD, PhD; Neus Barrantes-Vidal, PhD; Marie-Odile Krebs, MD, PhD; Merete Nordentoft, MD, PhD; Birte Glenthøj, MD, DMSc; Stephan Ruhrmann, MD; Gabriele Sachs, MD; Bart Rutten, MD, PhD; Jim van Os, MD, PhD; Lieuwe de Haan, MD, PhD; Eva Velthorst, PhD; Mark van der Gaag, PhD; Lucia R. Valmaggia, PhD; Philip McGuire, MD, PhD; for the EU-GEI High Risk Study Group Supplemental content IMPORTANCE The development of adverse clinical outcomes in patients with psychosis has been associated with behavioral and neuroanatomical deficits related to emotion processing. However, the association between alterations in brain regions subserving emotion processing and clinical outcomes remains unclear. OBJECTIVE To examine the association between alterations in emotion processing and regional gray matter volumes in individuals at clinical high risk (CHR) for psychosis, and the association with subsequent clinical outcomes. DESIGN, SETTING, AND PARTICIPANTS This naturalistic case-control study with clinical follow-up at 12 months was conducted from July 1, 2010, to August 31, 2016, and collected data from 9 psychosis early detection centers (Amsterdam, Basel, Cologne, Copenhagen, London, Melbourne, Paris, The Hague, and Vienna). Participants (213 individuals at CHR and 52 healthy controls) were enrolled in the European Network of National Schizophrenia Networks Studying Gene-Environment Interactions (EU-GEI) project. Data were analyzed from October 1, 2018, to April 24, 2019. MAIN MEASURES AND OUTCOMES Emotion recognition was assessed with the Degraded Facial Affect Recognition Task. Three-Tesla magnetic resonance imaging scans were acquired from all participants, and gray matter volume was measured in regions of interest (medial prefrontal cortex, amygdala, hippocampus, and insula). Clinical outcomes at 12 months were evaluated for transition to psychosis using the Comprehensive Assessment of At-Risk Mental States criteria, and the level of overall functioning was measured through the Global Assessment of Functioning [GAF] scale. RESULTS A total of 213 individuals at CHR (105 women [49.3%]; mean [SD] age, 22.9 [4.7] years) and 52 healthy controls (25 women [48.1%]; mean [SD] age, 23.3 [4.0] years) were included in the study at baseline. At the follow-up within 2 years of baseline, 44 individuals at CHR (20.7%) had developed psychosis and 169 (79.3%) had not. Of the individuals at CHR reinterviewed with the GAF, 39 (30.0%) showed good overall functioning (GAF score, ⱖ65), whereas 91 (70.0%) had poor overall functioning (GAF score, <65). Within the CHR sample, better anger recognition at baseline was associated with worse functional outcome (odds ratio [OR], 0.88; 95% CI, 0.78-0.99; P = .03). In individuals at CHR with a good functional outcome, positive associations were found between anger recognition and hippocampal volume (ze = 3.91; familywise error [FWE] P = .02) and between fear recognition and medial prefrontal cortex volume (z = 3.60; FWE P = .02), compared with participants with a poor outcome. The onset of psychosis was not associated with baseline emotion recognition performance (neutral OR, 0.93; 95% CI, 0.79-1.09; P = .37; happy OR, 1.03; 95% CI, 0.84-1.25; P = .81; fear OR, 0.98; 95% CI, 0.85-1.13; P = .77; anger OR, 1.00; 95% CI, 0.89-1.12; P = .96). No difference was Author Affiliations: Author observed in the association between performance and regional gray matter volumes in affiliations are listed at the end of this individuals at CHR who developed or did not develop psychosis (FWE P < .05). article. Group Information: The members of CONCLUSIONS AND RELEVANCE In this study, poor functional outcome in individuals at CHR the EU-GEI High Risk Study Group was found to be associated with baseline abnormalities in recognizing negative emotion. This appear at the end of the article. finding has potential implications for the stratification of individuals at CHR and suggests that Corresponding Author: Gemma interventions that target socioemotional processing may improve functional outcomes. Modinos, PhD, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, SE5 8AF JAMA Psychiatry. 2020;77(2):190-200. doi:10.1001/jamapsychiatry.2019.3501 London, United Kingdom Published online November 13, 2019. (gemma.modinos@kcl.ac.uk). 190 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 07/30/2020 Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis Original Investigation Research P sychotic disorders are associated with socioemotional dysfunction, which manifests as emotion perception Key Points and expression deficits and heightened emotional Question Is altered emotion recognition associated with adverse responsivity1 and represents a relatively poor prognosis.2-4 clinical and functional outcomes in people at clinical high risk Neuroimaging studies in established schizophrenia indicate that for psychosis? socioemotional dysfunction is associated with functional5-13 and Findings In this case-control study of 213 individuals at clinical structural14-16 alterations within a corticolimbic circuit that in- high risk for psychosis and 52 healthy participants, abnormalities cludes the medial prefrontal cortex (MPFC), amygdala, hippo- in the recognition of negative emotion at baseline were associated campus, and insula, consistent with postmortem evidence im- with neuroanatomical alterations in the medial prefrontal cortex plicating these regions in the disorder.17 Preclinical studies in and hippocampus and with a low level of functioning at animal models of psychosis suggest that targeting corticolim- a 12-month follow-up. bic dysregulation during the premorbid phase may prevent the Meaning This study found that, in people with high risk for emergence of schizophrenia-like features in adulthood.18,19 developing psychosis, functional outcomes are associated with Human participants at clinical high risk (CHR) for psychosis also the degree to which their emotion processing is altered. show deficits in emotion processing20-22 and hyperactivation within corticolimbic regions during emotional tasks23-25 that are qualitatively similar to those seen in patients with schizophre- Inclusion and Exclusion Criteria and Baseline Measures nia. However, the association between alterations in brain Whether participants met CHR criteria was assessed with regions subserving emotion processing and clinical outcomes the Comprehensive Assessment of At-Risk Mental States in individuals at CHR for psychosis remains unclear. (CAARMS).28 Exclusion criteria were past or present diagno- The primary aim of the present study was to examine the sis of psychotic disorders, past or present neurological disor- association between emotion processing, gray matter vol- der, substance abuse or dependence according to DSM-IV ume (GMV) in brain areas implicated in emotion processing, criteria, contraindication to scanning, or estimated IQ lower and clinical and functional outcomes in individuals at CHR for than 60. Healthy controls could not meet the criteria for CHR psychosis. We assessed emotion recognition and regional GMVs or have a reported personal or (first-degree) family history of in individuals at CHR and healthy controls, and we evaluated a psychiatric or neurological disorder. Individuals who met CHR clinical and functional outcomes in the CHR sample after 12 criteria and were being treated with antipsychotic medica- months of follow-up. We tested the hypothesis that in indi- tion were not excluded as long as this medication had not been viduals at CHR, adverse clinical outcomes (the subsequent prescribed for a psychotic episode. Data on age, sex, race/ onset of psychosis or a poor level of functioning) would be as- ethnicity, and years of education were obtained from the Medi- sociated with abnormalities in emotion recognition (happy, cal Research Council Sociodemographic Schedule.29 At base- angry, fearful, and neutral faces) and decreased GMV in cor- line, trained raters assessed participants using the CAARMS28 ticolimbic areas (MPFC, amygdala, hippocampus, and insula) and the Global Assessment of Functioning (GAF) scale (score at baseline.15,26 range: 0-100, with the highest score indicating superior func- tioning and no symptoms).30 Interrater reliability was as- sessed with mandatory rating of online CAARMS and GAF train- ing videos (eMethods and eTable 2 in the Supplement). Methods Estimated IQ was identified using the shortened Wechsler Adult Participants Intelligence Scale,31 cannabis use (yes or no) was assessed with In this case-control study, baseline neuroimaging and the modified Cannabis Experiences Questionnaire,32 and emotion processing data and clinical and functional out- tobacco (cigarettes per day) and alcohol (drinks per week) use comes were collected from 9 psychosis early detection cen- were recorded with the Composite International Diagnostic ters (Amsterdam, the Netherlands; Basel, Switzerland; Interview.33 Cologne, Germany; Copenhagen, Denmark; London, United Kingdom; Melbourne, Australia; Paris, France; The Hague, Emotion Recognition Assessment, MRI, the Netherlands; and Vienna, Austria) from July 1, 2010, and Clinical Follow-up to August 31, 2016, as part of the European Network A computerized version of the Degraded Facial Affect Recog- of National Schizophrenia Networks Studying Gene- nition (DFAR) Task34 was used to assess emotion recognition Environment Interactions (EU-GEI) project.27 The present performance. The task included photographs of 4 different study included the subset of individuals from the total actors (2 men and 2 women) portraying 4 different emotions: EU-GEI sample (345 help-seeking individuals at CHR and angry, happy, fearful, and neutral. The task comprised 64 trials, 66 healthy controls) for whom both neuroimaging and with 16 presentations of each of the 4 emotion categories, emotional processing data were available; eTable 1 in the shown at 100% and 75% intensity to increase task difficulty.35,36 Supplement shows basic characteristics of individuals in When a face was displayed on the computer screen, a partici- and out of the study). Ethical approval for this study was pant indicated its emotional expression by pressing a button. obtained from the local research ethics committees at each A participant’s DFAR task accuracy was computed on the ba- of the 9 sites. All participants provided written informed sis of the total number of neutral, happy, fearful, and angry consent. emotions correctly recognized; higher DFAR scores indicated jamapsychiatry.com (Reprinted) JAMA Psychiatry February 2020 Volume 77, Number 2 191 Downloaded From: https://jamanetwork.com/ on 07/30/2020 Research Original Investigation Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis better performance. The Benton Facial Recognition Test of familywise error (FWE) P < .05,43 using a prespecified bilat- (BFRT)37 was used to control for the possibility that impaired eral mask. The mask was derived from the WFU_Pitckatlas tool- facial affect recognition was secondary to a deficit in general box in SPM12 and comprised a network implicated in emotion facial recognition.35,36 Details on BFRT performance are shown (MPFC, amygdala, hippocampus, and insula). The MPFC and in the eResults, eTable 3, and eFigure 1 in the Supplement. amygdala were chosen because of their central roles in emo- At baseline, 3-T magnetic resonance imaging (MRI) scans tion processing44-46 and because emotion-processing abnor- were collected from all participants and preprocessed using malities in schizophrenia have been associated with volumet- voxel-based morphometry 38 implemented on statistical ric alterations in these regions.10,12,15,24,47 The hippocampus was parametric mapping software (SPM12; GNU General Public selected because of its key role in the onset of psychosis in pre- License). The eMethods and eFigure 2 in the Supplement pro- clinical models48 and volumetric decreases in this region in vide details on MRI acquisition, quality assessment, and pre- individuals at CHR who transition to psychosis.49 The insula was processing. At 12 months, the level of overall functioning was chosen because of its involvement in emotion processing50,51 and assessed with the GAF scale.30 Changes in GAF scores over time its role in facial emotion recognition in individuals at CHR.21 were also analyzed (eResults and eFigure 3 in the Supple- Potential confounding effects of antipsychotic or antide- ment). Transition or nontransition to psychosis within a 2-year pressant medication (yes or no), substance use (tobacco, can- period after baseline was identified using the CAARMS psycho- nabis, or alcohol), or baseline levels of CAARMS anxiety or de- sis threshold criteria (eTable 4 in the Supplement), with diag- pression symptom severity on regions showing statistically nosis confirmed by the Structured Clinical Interview for DSM-IV significant DFAR-GMV interactions were assessed in SPSS Axis I Disorders,39 administered by a researcher trained in its use. (antidepressants, substances, and CAARMS anxiety or depres- sion) or SPM (antipsychotic drugs) (eResults in the Supple- Statistical Analysis ment). Because our hypotheses involved the association be- Demographic and Clinical Data tween DFAR-GMV interactions with clinical outcomes, only Analyses of demographic and clinical data were performed in participants for whom these data were available were in- SPSS, version 25 (IBM Corp). The association of group with these cluded in the present study. Group differences in GMV are cur- measures was examined using 2-sample, unpaired, 2-tailed rently under analysis (M. J. Kempton, PhD, unpublished data, t tests or χ2 tests. Effect sizes are expressed as odds ratios (OR) 2019). Analysis of DFAR task performance and GMV by site or and considered statistically significant at 2-sided P < .05. Data scanner is shown in eTable 6 in the Supplement. Sensitivity were analyzed from October 1, 2018, to April 24, 2019. analyses data are reported in the eResults, eTables 7 and 8, and eFigures 4 and 5 in the Supplement. DFAR Data and Data Integration Binary logistic regression examined the associations between DFAR task performance and case-control status at baseline, ad- justed for age, sex, IQ, site, and BFRT score.36 To find the asso- Results ciations between DFAR performance and clinical outcomes, we Demographic and Clinical Data dichotomized the CHR sample according to transition vs non- At baseline, the participants who had MRI and DFAR data and transition to psychosis40 and in terms of good (GAF score ≥65) therefore were included in the study were 213 help-seeking in- vs poor (GAF score <65) overall functioning at follow-up.30 A GAF dividuals at CHR for psychosis (105 women [49.3%]; mean [SD] score of 65 was chosen for consistency with the score in previ- age, 22.9 [4.7] years) and 52 healthy controls (25 women ous neuroimaging studies in CHR.41,42 Binary logistic regres- [48.1%]; mean [SD] age, 23.3 [4.0] years) (Table 1). Of the 213 sion analyses were performed (transition vs nontransition; good individuals at CHR, 193 (90.6%) were naive to antipsychotic vs poor functioning) with the same covariates. After preprocess- medication, and the remaining 20 (9.4%) were receiving low ing, segmented, normalized, and smoothed GMV images were doses of antipsychotics (<1.5 mg haloperidol chlorpromazine analyzed in a group using SPM12 to find the associations with equivalents per day). The CHR and healthy control groups did DFAR task performance. Individual DFAR task accuracy values not differ statistically significantly in age (t = 0.596; P = .55), were entered as regressors in separate voxelwise analyses of vari- sex (χ2 = 0.025; P = .88), race/ethnicity (χ2 = 9.023; P = .11), ance to examine the interactions between group status (healthy BFRT score (t = –0.005; P = .99), cigarettes (t = –1.913; P = .06), controls vs CHR; transition vs nontransition; good vs poor func- alcohol (t = –0.304; P = .76), or cannabis (χ2 = 0.403; P = .53) tioning) and DFAR task performance on each emotion cat- use. However, individuals at CHR had fewer years of educa- egory, covarying for age, sex, IQ, scanner, and BFRT score. For tion (t = 3.639; P < .001) and lower IQ (t = 5.051; P < .001). the imaging analysis, the variable scanner instead of site was used. Although the DFAR task was administered at each site, par- Clinical Outcomes and Facial Emotional Processing ticipants from Amsterdam and The Hague were scanned in After 12 months, 39 of the 130 individuals (30.0%) at CHR re- Amsterdam, a site that changed scanners during the EU-GEI proj- interviewed with the GAF at follow-up had good overall func- ect (eTable 5 in the Supplement). The analyses of variance also tioning (CHR-GO), whereas 91 (70.0%) had poor overall func- used proportional scaling of the total intracranial volume to ad- tioning (CHR-PO). No significant differences at baseline were just for global effects. An initial height threshold of uncor- observed in any clinical or demographic measures between rected P < .001 was used to then apply small volume correction these subgroups (Table 1). Within the 2 years after baseline, 44 for region-of-interest analyses at a voxelwise height threshold individuals at CHR (20.7%) had developed a first episode of 192 JAMA Psychiatry February 2020 Volume 77, Number 2 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 07/30/2020 Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis Original Investigation Research Table 1. Baseline Demographic, Clinical, and Medication Characteristics of Participants HC CHR CHR-NT CHR-T CHR-GO CHR-PO Group Group Group Group Group Group Measure (n = 52) (n = 213) P Value (n = 169) (n = 44) P Value (n = 39) (n = 91) P Value Age, mean (SD), y 23.3 (4.0) 22.9 (4.7) .55 23.0 (4.7) 22.6 (4.7) .59 23.5 (4.7) 23.1 (5.0) .66 Sex, No. Male 27 108 83 25 20 50 .88 .36 .70 Female 25 105 86 19 19 41 Years of education, mean (SD) 16.3 (2.9) 14.6 (3.1) <.001 14.7 (3.1) 14.3 (3.0) .53 15.5 (2.8) 15.0 (3.2) .42 Race/ethnicity (% white), % 65.4 72.8 .11 72.8 72.7 .98 71.8 76.9 .41 CAARMS score, mean (SD) Positive 0.7 (1.6) 9.9 (4.2) <.001 9.8 (4.4) 10.6 (3.6) .24 9.7 (4.5) 10.3 (4.1) .46 Negative 0.8 (1.7) 7.2 (3.4) <.001 7.1 (3.5) 7.3 (3.4) .72 7.3 (3.5) 7.6 (3.2) .70 Anxiety 0.6 (1.1) 3.1 (1.6) <.001 3.1 (1.6) 3.2 (1.6) .84 3.1 (1.6) 3.2 (1.5) .86 Depression 0.4 (0.9) 3.4 (1.3) <.001 3.4 (1.3) 3.5 (1.4) .62 3.3 (1.1) 3.5 (1.2) .41 Baseline GAF score, mean (SD) 87.2 (9.1) 54.0 (10.0) <.001 53.7 (9.6) 55.3 (11.6) .41 53.8 (8.7) 54.2 (11.1) .86 Antipsychotic drugs, No. Total 52 167 131 36 31 72 No 52 147 NA 122 25 <.001 30 62 .12 Yes 0 20 9 11 1 10 Antidepressants, No. Total 52 167 131 36 31 72 No 52 102 NA 79 23 .70 22 42 .23 Yes 0 65 52 13 9 30 Psychological treatment, No.a Total 51 193 153 40 33 85 No 47 117 <.001 87 30 .04 20 49 .77 Yes 4 76 66 10 13 36 Total intracranial volume, 1 493 654.5 1 493 871.5 .99 1 498 994.0 1 474 196.6 .42 1543 165.2 1 492 014.2 .15 mean (SD), mm3 (178 968.1) (180 458.9) (178 703.8) (187 846.8) (169 397.6) (192 273.4) BFRT score, mean (SD) 22.3 (2.3) 22.3 (2.2) .99 22.2 (2.2) 23.1 (1.8) .01 22.3 (2.1) 22.3 (2.3) .99 Abbreviations: BFRT, Benton Facial Recognition Test; CAARMS, Comprehensive 0-100, with the highest score indicating superior functioning and Assessment of At-Risk Mental States; CHR, clinical high risk; CHR-GO, clinical no symptoms); HC, healthy controls; MRI, magnetic resonance imaging; high risk with good overall functioning (GAF ⱖ65); CHR-NT, clinical high NA, not applicable. risk–nontransitioned; CHR-PO, clinical high risk with poor overall functioning a Psychological treatment included counseling sessions, cognitive behavioral (GAF <65); CHR-T, clinical high risk–transitioned; DFAR, Degraded Facial therapy, family therapy, psychoeducation, or other form of therapy. Affect Recognition; GAF, Global Assessment of Functioning scale (score range: Table 2. Group Differences in Facial Emotion Recognitiona HC Group (n = 52) CHR-GO Group (n = 39) CHR-NT Group (n = 169) vs CHR Group (n = 213) vs CHR-PO Group (n = 91) vs CHR-T Group (n = 44) DFAR Task OR (95% CI) P Value OR (95% CI) P Value OR (95% CI) P Value Neutral 1.02 (0.86-1.21) .82 1.03 (0.85-1.24) .77 0.93 (0.79-1.09) .37 Happy 1.01 (0.83-1.23) .91 0.96 (0.76-1.21) .70 1.03 (0.84-1.25) .81 Fear 0.89 (0.77-1.02) .10 1.13 (0.96-1.32) .13 0.98 (0.85-1.13) .77 Anger 1.08 (0.96-1.22) .22 0.88 (0.78-0.99) .03 1.00 (0.89-1.12) .96 Abbreviations: CHR, clinical high risk; CHR-GO, clinical high risk with good GAF, Global Assessment of Functioning scale; HC, healthy controls; overall functioning (GAF ⱖ65); CHR-NT, clinical high risk–nontransitioned; OR, odds ratio. CHR-PO, clinical high risk with poor overall functioning (GAF <65); a Adjusted for age, sex, IQ, site, and general facial recognition. CHR-T, clinical high risk–transitioned; DFAR, Degraded Facial Affect Recognition; psychosis or transitioned (CHR-T). The mean (SD) time to tran- tribution of individuals in the CHR-NT or CHR-T groups among sition in the CHR-T group was 296.3 (257.6) days. Among in- the CHR-GO and CHR-PO follow-up groups is shown in eFig- dividuals at CHR, 169 (79.3%) did not develop psychosis within ure 6 in the Supplement. this period, or nontransitioned (CHR-NT). At baseline, the only At baseline, DFAR task accuracy did not differ for any emo- significant difference in clinical or demographic measures be- tion between the healthy control and the total CHR groups tween the CHR-NT and the CHR-T subgroups was that the independent of outcomes (Table 2 and Figure 1A). However, CHR-T group had a higher BFRT score (t = –2.470; P = .01) and within the CHR sample, anger recognition at baseline was included more individuals receiving low doses of antipsy- significantly associated with the level of functioning at chotic medications (χ2 = 15.028; P < .001) (Table 1). The dis- 12-month follow-up; anger recognition was abnormal in the jamapsychiatry.com (Reprinted) JAMA Psychiatry February 2020 Volume 77, Number 2 193 Downloaded From: https://jamanetwork.com/ on 07/30/2020 Research Original Investigation Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis Figure 1. Group Differences in Emotion Recognition DFAR and GMV Data Integration At baseline, a significant group × DFAR happy × GMV inter- A Baseline case control action was observed in the left MPFC (x, y, z = –12, 54, 0; 20 z = 4.01; FWE P = .03). In healthy controls, a negative corre- lation between the recognition of happiness and MPFC vol- ume was found, which was absent in the CHR sample 15 (Figure 2A). Similarly, a significant group × GMV interaction Mean Accuracy in the MPFC for DFAR anger was observed (x, y, z = 0, 60, 18; 10 z = 3.83; FWE P = .049), reflecting a positive correlation be- tween the recognition of anger and GMV in the MPFC among 5 healthy controls that was absent in individuals at CHR (Figure 2B). No other significant interactions with neutral or 0 fearful emotion were found. Neutral Happiness Fear Anger Neutral Happiness Fear Anger HC Group CHR Group Functional Outcome and Transition to Psychosis Subdivision of the CHR sample according to level of function- B Overall functioning outcome ing at follow-up revealed that participants in the CHR-GO group 20 P =.03 showed a positive association between anger recognition and left hippocampal volume (x, y, z = −32, −40, −3; z = 3.91; FWE 15 P = .02) and between fear recognition and left MPFC volume Mean Accuracy (x, y, z = −12, 38, –9; z = 3.60; FWE P = .02), compared with par- 10 ticipants in the CHR-PO group (Figure 3). No other significant group interactions with neutral or happy emotion were ob- 5 served. No significant group × DFAR × GMV interactions based on transition vs nontransition outcomes were found. Analy- 0 sis of the potential confounders on all DFAR × GMV interac- Neutral Happiness Fear Anger Neutral Happiness Fear Anger tion data (at baseline and follow-up) rendered the results largely CHR-GO Group CHR-PO Group unchanged (eResults; eTable 10; and eFigures 7, 8, and 9 in the Supplement). C Psychosis transition outcome The onset of psychosis was not associated with baseline 20 emotion recognition performance (neutral OR, 0.93; 95% CI, 0.79-1.09; P = .37; happy OR, 1.03; 95% CI, 0.84-1.25; P = .81; fear 15 OR, 0.98; 95% CI, 0.85-1.13; P = .77; anger OR, 1.00; 95% CI, 0.89- Mean Accuracy 1.12; P = .96). No difference was observed in the association be- 10 tween performance and regional GMVs in individuals at CHR who developed or did not develop psychosis (FWE P < .05). 5 0 Neutral Happiness Fear Anger Neutral Happiness Fear Anger Discussion CHR-NT Group CHR-T Group To our knowledge, this case-control study is the first to as- sess the association between alterations in emotion process- A, The healthy control (HC) group comprised 52 participants, and the clinical high risk (CHR) group comprised 213 participants. B, The CHR with ing and clinical and functional outcomes in individuals at CHR good overall functioning (CHR-GO) group comprised 39 participants, and the for psychosis. We studied a large sample of participants at CHR, CHR with poor overall functioning (CHR-PO) group comprised 91 participants. most of whom were not taking antipsychotic drugs. The main C, The CHR–nontransitioned (CHR-NT) group comprised 169 outcomes, finding was that, in individuals at CHR, a poor functional out- and the CHR–transitioned (CHR-T) group comprised 44 outcomes. The group differences were adjusted for age, sex, IQ, site, and general facial come was associated with baseline abnormalities in both the recognition. The horizontal line in each box represents the median; recognition of angry emotion and in GMV in brain regions im- top and bottom box borders, 75th and 25th percentiles, respectively; plicated in the processing of anger and fear. More specifi- whiskers, 90th and 10th percentiles; white circles, out values; and black circles, far out values. cally, compared with individuals at CHR with poor overall func- tioning, individuals at CHR with a good functional outcome showed a statistically significant positive association be- tween anger recognition and hippocampal volume as well as CHR-PO group compared with the CHR-GO group (OR, 0.88; between fear recognition and MPFC volume. No associations 95% CI, 0.78-0.99; P = .03; Table 2 and Figure 1B). No signifi- were found between alterations in emotional processing or in cant associations were observed with CHR-NT and CHR-T out- brain regions implicated in emotional processing and the sub- comes (Table 2 and Figure 1C). See eTable 9 in the Supplement sequent onset of psychosis in individuals at CHR. However, in for analysis of potential confounders. healthy controls at baseline, emotion recognition (eg, happy 194 JAMA Psychiatry February 2020 Volume 77, Number 2 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 07/30/2020 Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis Original Investigation Research Figure 2. Baseline Associations Between Emotion Recognition (DFAR), Gray Matter Volume (GMV), and Group Status A Baseline group × GMV × DFAR happy 80 HC CHR 70 GMV in MPFC 60 50 0 5 10 15 20 DFAR Happy Score B Baseline group × GMV × DFAR anger 65 60 GMV in MPFC 55 50 45 40 0 5 10 15 20 DFAR Anger Score The healthy control (HC; blue) group comprised 52 participants, and the clinical DFAR, degraded facial affect recognition. Orange marks are the statistical high risk (CHR; orange) group comprised 213 participants. Baseline associations parametric maps of the interactions between GMV, DFAR performance, and were adjusted for age, sex, IQ, scanner, and general facial recognition group overlaid on a standard T1 template in MRICron software (NITRC). (familywise error P < .05). MPFC indicates medial prefrontal cortex; or angry) was significantly associated with MPFC volume, and episode psychosis.58 In CHR samples, poor functional out- these associations were absent in the CHR sample. These find- comes have been associated with increased hippocampal ings were observed after adjusting for age, sex, scanner, IQ, glutamate levels, 4 2 inc reased resting hippoc ampal and BFRT score and remained largely unchanged after remov- perfusion, 5 9 and altered hippoc ampal activation. 4 1 ing participants at CHR taking low doses of antipsychotic medi- Decreased MPFC volume has been associated with altered cations, as well as when examining potential effects of anti- emotional processing in patients with schizophrenia com- depressants or substance use. Overall, these findings suggest pared with healthy controls.15,16,26 In terms of specific emo- that emotional dysfunction and related brain regions may play tions, smaller prefrontal volumes have been associated with a role in the development of adverse functional outcomes in worse recognition of angry emotion in a large sample of individuals at increased risk for psychosis. patients with schizophrenia, a finding consistent with ours Consistent with the main hypothesis is the finding in the CHR-PO group, which found anger as the only DFAR that, in individuals at CHR, a low level of overall functioning task impairment in patients after adjusting for age, sex, esti- at follow-up was associated with baseline alterations in mated IQ, and BFRT score.26 Furthermore, a recent study the recognition of angry emotion and decreased anger- and using machine learning in individuals at CHR found that fear-related hippocampal and MPFC volume. Aberrant baseline MFPC and temporo-parieto-occipital volume reduc- emotion recognition is a robust factor in poor social52 and tions were factors in socio-occupational impairments at functional 53 outcomes in patients with schizophrenia. follow-up.60 By linking abnormalities in emotion recognition The hippocampus plays a key role in the pathophysiological and emotion-related brain regions to adverse functional out- process of schizophrenia,54 but the clinical and functional comes in individuals at CHR, the present study provides fur- importance of changes in this region is still unclear.55 Abnor- ther support for the notion that the pathophysiological pro- mal hippocampal volume has been associated with lower cess of CHR states for psychosis involves emotion-related global functioning in both schizophrenia 56,57 and first- regions.14 These findings may inform the development of jamapsychiatry.com (Reprinted) JAMA Psychiatry February 2020 Volume 77, Number 2 195 Downloaded From: https://jamanetwork.com/ on 07/30/2020 Research Original Investigation Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis Figure 3. Associations Between 12-Month Functional Outcomes, Gray Matter Volume (GMV), and Degraded Facial Affect Recognition (DFAR) A Functional outcome × GMV × DFAR anger B Functional outcome × GMV × DFAR fear 70 60 CHR-GO CHR-PO 55 GMV in Hippocampus 60 GMV in MPFC 50 50 45 40 40 0 5 10 15 20 0 5 10 15 20 DFAR Anger Score DFAR Fear Score The clinical high risk with good overall functioning (CHR-GO; blue) group error P < .05). Blue marks are the statistical parametric maps of the interactions comprised 39 participants, and the clinical high risk with poor overall between GMV, DFAR performance, and group overlaid on a standard T1 functioning (CHR-PO; orange) group comprised 91 participants. Associations template in MRICron software (NITRC). adjusted for age, sex, IQ, scanner, and general facial recognition (familywise new treatments for individuals at CHR, most of whom have comes may be more similar. Results of the present study sug- deficits in social and role functioning that persist after first gest that altered emotion processing may predispose individu- presentation.61,62 als at CHR to poor functional outcomes through interactions The subsequent onset of psychosis was not associated with with prefronto-hippocampal anatomy. abnormalities in either emotion recognition or emotion- Another finding was that the DFAR-GMV analysis revealed related brain regions. This finding was unexpected given that significant differences between the healthy control group and impaired facial emotion recognition has been reported in the total CHR group at baseline independent of outcomes. patients with established schizophrenia 63,6 4 and first- Within healthy controls, MPFC volume was negatively episode psychosis,35,65,66 and a previous study in at-risk youths associated with happy emotion recognition, whereas anger reported that lower baseline emotion recognition was associ- recognition was positively associated with MPFC volume (and ated with psychosis transition.67 Nevertheless, these find- insula volume after removing participants taking antipsy- ings are in line with those from another large study of indi- chotic medications); these associations were absent in the CHR viduals at CHR, which found no association between baseline sample. Broadly, these findings align with reports that GMV in emotion recognition and the subsequent onset of psychosis,68 a topographically similar MPFC region was associated with defi- and with recent meta-analytic evidence that social cognitive cits in social cognitive and emotional tasks in patients with deficits are not associated with the subsequent transition to established schizophrenia.15,16,26 This ventral portion of the psychosis.69 Divergent results may reflect differences in the MPFC is involved in monitoring internal affective states and mean age of the participants and in sample size of the respec- regulates the influence of those states on behavior.44 The di- tive studies (23 years in the present study [n = 213], 16 years vergent directionality of the observed DFAR-GMV correlations in Allott et al67 [n = 37], and 20 years in Addington et al68 in healthy controls (negative for happiness, and positive for [n = 172]). Another potential factor is limited statistical power. anger) is interesting and merits further investigation. It may Even with relatively large CHR samples, the size of the CHR-T relate to different requirements for prefrontal involvement subgroup may be small because only some participants will de- as a function of emotional valence and its associated velop psychosis. This small size may be less of a problem when emotion regulation requirement or goal. 70 Because no outcome is defined by level of functioning, as the numbers of further significant associations were observed with this participants in subgroups with good and poor functional out- a priori region-of-interest analysis or in the complementary 196 JAMA Psychiatry February 2020 Volume 77, Number 2 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 07/30/2020 Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis Original Investigation Research whole-brain analysis, these results indicate that the promi- into account current symptom severity and level of nent role of the MPFC in emotion recognition may be compro- functioning. 73 Nevertheless, complementary analyses of mised in the CHR state. functional outcomes adjusted for psychosis transition Contrary to our expectations, we found no significant outcomes left the results of DFAR task performance and associations between amygdala volumes and DFAR perfor- DFAR-GMV associations unchanged, suggesting that GAF mance between the baseline groups or in association with clini- scores captured a unique component of outcome beyond cal or functional outcomes. The amygdala plays a key role in psychosis alone. Although combining multicenter data sets emotional processing,46 and evidence for abnormal amyg- increases sensitivity, the application of voxel-based mor- dala reactivity has been provided by several functional MRI phometry to large-scale investigations pooling neuroimaging studies that used emotion processing tasks in patients with data across sites has some potential limitations. We used the schizophrenia.10,11 However, because the altered amygdala re- scanner as a covariate to mitigate against the introduction of sponse in schizophrenia is primarily evident during implicit between-center sources of variability to the data related to, emotion paradigms,12 the lack of association in this study for example, imaging hardware, because of evidence that this may be attributed to the use of an explicit test (DFAR task) and approach can suppress scanner effects even when the ratio of may align with previous negative findings with the DFAR task cases to controls was unbalanced across sites.74 However, by being associated with amygdala volumes in a large sample recently developed methods such as ComBat (a popular of patients with schizophrenia.26 Future studies examining method in genomics for combatting batch effects when com- implicit and explicit emotion processing in CHR samples are bining batches of gene expression microarray data) appear to needed to expand on this possibility. be successful at harmonizing cortical thickness measure- ments obtained from multiple sites75 and should be consid- Limitations ered in future large-scale collaborative imaging studies. The present study has limitations. It assessed functional outcomes in the CHR sample at 12 months and transition or nontransition outcomes within 2 years from baseline. Although most transitions to psychosis seem to occur in this Conclusions period,71 a median duration of the prodromal phase of 64 This case-control study found that poor functional outcome months has been reported 72 ; during a longer follow-up in individuals at CHR of psychosis was associated with base- period, additional transitions may have been detected, line abnormalities in the recognition of angry emotion and with which could have altered the results. Although a 20% abnormal associations between anger and fear emotion rec- transition rate provides reasonable power, and transition or ognition and between hippocampal and MPFC volumes. These nontransition outcomes were recorded for all participants findings have potential implications for the stratification of at CHR at follow-up, the sample size may still be limited individuals at CHR according to subsequent outcomes and sug- to detect an effect associated with transition. The GAF gest that functional outcomes might be improved by interven- scale was used to index global functioning, which takes tions that target socioemotional processing. ARTICLE INFORMATION Centre Glostrup, Copenhagen, Denmark (Pantelis, Clinical Intervention and Neuropsychiatric Accepted for Publication: September 1, 2019. Glenthøj); Center for Clinical Intervention and Schizophrenia Research, Mental Health Center Neuropsychiatric Schizophrenia Research, Glostrup, Mental Health Services in the Capital Published Online: November 13, 2019. University of Copenhagen, Mental Health Centre Region of Copenhagen, University of Copenhagen, doi:10.1001/jamapsychiatry.2019.3501 Glostrup, Copenhagen, Denmark (Pantelis, Copenhagen, Denmark (Nordentoft); Department Open Access: This is an open access article Glenthøj); Department of Clinical Medicine, Faculty of Psychiatry and Psychotherapy, University of distributed under the terms of the CC-BY License. of Health and Medical Sciences, University of Cologne, Cologne, Germany (Ruhrmann); © 2019 Modinos G et al. JAMA Psychiatry. Copenhagen, Copenhagen, Denmark (Pantelis); Department of Psychiatry and Psychotherapy, Author Affiliations: Department of Psychosis University of Basel Psychiatric Hospital, Basel, Medical University of Vienna, Vienna, Austria Studies, Institute of Psychiatry, Psychology & Switzerland (Riecher-Rössler, Borgwardt); LiNC— (Sachs); School for Mental Health and Neuroscience, King's College London, London, Lab Interdisciplinar Neurociências Clínicas, Depto Neuroscience, Department of Psychiatry and United Kingdom (Modinos, Kempton, Tognin, Psiquiatria, Escola Paulista de Medicina, Neuropsychology, Faculty of Health, Medicine and Calem, Porffy, Antoniades, Mason, Azis, Allen, Universidade Federal de São Paulo—UNIFESP, Life Sciences, Maastricht University, Maastricht, the van Os, McGuire); Department of Neuroimaging, São Paulo, Brazil (Bressan); Departament de Netherlands (Rutten); University Medical Centre Institute of Psychiatry, Psychology & Neuroscience, Psicologia Clínica i de la Salut (Universitat Utrecht Brain Center, Department of Psychiatry, King's College London, London, United Kingdom Autònoma de Barcelona), Fundació Sanitària Sant Utrecht University Medical Centre, Utrecht, the (Modinos); National Institute for Health Research, Pere Claver (Spain), Spanish Mental Health Netherlands (van Os); Early Psychosis Department, Biomedical Research Centre, London, United Research Network (CIBERSAM), Barcelona, Spain Amsterdam UMC, Amsterdam, the Netherlands Kingdom (Kempton, McGuire); Department of (Barrantes-Vidal); University of Paris, GHU-Paris, (de Haan, Velthorst); Department of Psychiatry, Psychology, University of Roehampton, London, Sainte-Anne, C’JAAD, Hospitalo-Universitaire Icahn School of Medicine at Mount Sinai, New York, United Kingdom (Allen); Orygen, The National Department SHU, Inserm U1266, Institut de New York (Velthorst); Amsterdam Public Mental Centre of Excellence in Youth Mental Health, Psychiatrie (CNRS 3557), Paris, France (Krebs); Health Research Institute, Department of Clinical University of Melbourne, Melbourne, Victoria, Mental Health Center Copenhagen, Center for Psychology, Faculty of Behavioural and Movement Australia (Nelson, McGorry); Centre for Youth Clinical Intervention and Neuropsychiatric Sciences, Vrije Universiteit Amsterdam, Mental Health, University of Melbourne, Schizophrenia Research, Mental Health Center Amsterdam, the Netherlands (van der Gaag); Melbourne, Victoria, Australia (Nelson, McGorry); Glostrup, Mental Health Services in the Capital Parnassia Psychiatric Institute, Department of Center for Neuropsychiatric Schizophrenia Region of Copenhagen, University of Copenhagen, Psychosis Research, The Hague, the Netherlands Research, University of Copenhagen, Mental Health Copenhagen, Denmark (Nordentoft); Center for (van der Gaag); Institute of Psychiatry, Psychology jamapsychiatry.com (Reprinted) JAMA Psychiatry February 2020 Volume 77, Number 2 197 Downloaded From: https://jamanetwork.com/ on 07/30/2020 Research Original Investigation Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis & Neuroscience, Department of Psychology, King's grants from Medical Research Council and the of Clinical Psychology and EMGO+ Institute for College London, London, United Kingdom Lundbeck Foundation during the conduct of the Health and Care Research, and Parnassia Psychiatric (Valmaggia); South London and Maudsley National study; and receiving grants from The Lundbeck Institute, Department of Psychosis Research; Eva Health Service Foundation Trust, London, Foundation and H. Lundbeck A/S outside the Velthorst, AMC, Academic Psychiatric Centre, United Kingdom (McGuire). submitted work. Dr Ruhrmann reported receiving Department Early Psychosis, and Icahn School of Author Contributions: Dr Modinos had full access grants from the European Community during the Medicine at Mount Sinai, Department of Psychiatry; to all of the data in the study and takes conduct of the study and personal fees from Tamar C. Kraan, AMC, Academic Psychiatric Centre, responsibility for the integrity of the data and the Boehringer Ingelheim outside the submitted work. Department Early Psychosis; Daniella S. van Dam, accuracy of the data analysis. Dr Sachs reported receiving grants from European AMC, Academic Psychiatric Centre, Department Concept and design: Modinos, Kempton, Mason, Community Seventh Framework Program during Early Psychosis; Nadine Burger, Parnassia McGorry, Krebs, Rutten, van Os, de Haan, the conduct of the study and personal fees from Psychiatric Institute, Department of Psychosis Valmaggia, McGuire. Lundbeck and Janssen-Cilag outside the submitted Research; Barnaby Nelson, Centre for Youth Mental Acquisition, analysis, or interpretation of data: work. Dr Rutten reported receiving grants from Health, University of Melbourne; Patrick McGorry, Modinos, Kempton, Tognin, Calem, Porffy, European Union during the conduct of the study. Centre for Youth Mental Health, University of Antoniades, Azis, Allen, Nelson, Pantelis, No other disclosures were reported. Melbourne; G Paul Amminger, Centre for Youth Riecher-Rössler, Borgwardt, Bressan, Funding/Support: The European Network of Mental Health, University of Melbourne; Christos Barrantes-Vidal, Nordentoft, Glenthøj, Ruhrmann, National Schizophrenia Networks Studying Pantelis, Center for Neuropsychiatric Schizophrenia Sachs, Rutten, van Os, de Haan, Velthorst, Gene-Environment Interactions (EU-GEI) Project Research (CNSR) and Center for Clinical van der Gaag, Valmaggia, McGuire. is funded by grant agreement Intervention and Neuropsychiatric Schizophrenia Drafting of the manuscript: Modinos, Tognin, HEALTH-F2-2010-241909 (Project EU-GEI) from Research (CINS), University of Copenhagen, and Mason, Allen. the European Community Seventh Framework University of Copenhagen, Faculty of Health and Critical revision of the manuscript for important Programme. Additional financial support was Medical Sciences, Department of Clinical Medicine; intellectual content: Modinos, Kempton, Tognin, obtained from the Institut National de la Santé et de Athena Politis, Centre for Youth Mental Health, Calem, Porffy, Antoniades, Azis, Nelson, McGorry, la Recherche Médicale (recurrent funding and University of Melbourne; Joanne Goodall, Centre Pantelis, Riecher-Rössler, Borgwardt, Bressan, fellowships) and by Fondation Pierre Deniker. The for Youth Mental Health, University of Melbourne; Barrantes-Vidal, Krebs, Nordentoft, Glenthøj, study received grant 08-MNP-007 from the French Anita Riecher-Rössler, University of Basel Ruhrmann, Sachs, Rutten, van Os, de Haan, government Agence Nationale de la Recherche and Psychiatric Hospital; Stefan Borgwardt, University Velthorst, van der Gaag, Valmaggia, McGuire. grant AOM-07-118 (Influence of Cannabis of Basel Psychiatric Hospital; Charlotte Rapp, Statistical analysis: Modinos, Kempton. Psychopathological Outcome in At-risk Mental University of Basel Psychiatric Hospital; Sarah Ittig, Obtained funding: McGorry, Pantelis, State [ICAAR study]) from the French Health University of Basel Psychiatric Hospital; Erich Riecher-Rössler, Borgwardt, Glenthøj, Ruhrmann, Ministry Programme Hospitalier de Recherche Studerus, University of Basel Psychiatric Hospital; Rutten, van Os, van der Gaag, Valmaggia, McGuire. Clinique. The Sainte-Anne Hospital Center Renata Smieskova, University of Basel Psychiatric Administrative, technical, or material support: promoted the study. Dr Kempton was supported by Hospital; Rodrigo Bressan, LiNC—Lab Modinos, Kempton, Tognin, Calem, Porffy, Mason, a Medical Research Council Fellowship grant MR/ Interdisciplinar Neurociências Clínicas, Depto Azis, Pantelis, Glenthøj, Sachs, van Os, J008915/1. Dr Pantelis was supported by Australia's Psiquiatria, Escola Paulista de Medicina, van der Gaag, McGuire. National Health and Medical Research Council Universidade Federal de São Paulo—UNIFESP; Supervision: Modinos, Allen, Nelson, Pantelis, Senior Principal Research Fellowship (ID: 628386 & Ary Gadelha, LiNC, Depto Psiquiatria, Escola Borgwardt, Bressan, Barrantes-Vidal, Krebs, 1105825) and by grant R246-2016-3237 from the Paulista de Medicina, UNIFESP; Elisa Brietzke, Nordentoft, de Haan, Velthorst, Valmaggia, Lundbeck Foundation. Dr Barrantes-Vidal was Depto Psiquiatria, Escola Paulista de Medicina, McGuire. supported by the Ministerio de Ciencia, Innovación UNIFESP; Graccielle Asevedo, LiNC, Depto Other—coordination data collection Amsterdam e Universidades (PSI2017-87512-C2-1-R), and the Psiquiatria, Escola Paulista de Medicina, UNIFESP; site/IRR training: Velthorst. Generalitat de Catalunya (2017SGR1612 and ICREA Elson Asevedo, LiNC, Depto Psiquiatria, Escola Academia Award). Dr Modinos was supported by a Paulista de Medicina, UNIFESP; Andre Zugman, Conflict of Interest Disclosures: Dr McGorry LiNC, Depto Psiquiatria, Escola Paulista de reported holding 2 patents issued (AU Sir Henry Dale Fellowship #202397/Z/16/Z, jointly funded by The Wellcome Trust and the Royal Medicina, UNIFESP; Neus Barrantes-Vidal, 2015203289; 9884034) and 2 pending Departament de Psicologia Clínica i de la Salut (CA2773031; 15/844444). Dr Pantelis reported Society. (Universitat Autònoma de Barcelona), Fundació receiving grants from Australian National Health & Role of the Funder/Sponsor: The funders had Sanitària Sant Pere Claver, Spanish Mental Health Medical Research Council during the conduct of the no role in the design and conduct of the study; Research Network (CIBERSAM); Tecelli study; grants from Lundbeck Foundation, personal collection, management, analysis, and Domínguez-Martínez, CONACYT-Dirección de fees from Lundbeck and Australia Pty Ltd; and interpretation of the data; preparation, review, or Investigaciones Epidemiológicas y Psicosociales, personal fees from Lundbeck and Australia Pty Ltd approval of the manuscript; and decision to submit Instituto Nacional de Psiquiatría Ramón de la outside the submitted work. Dr Riecher-Rössler the manuscript for publication. Fuente Muñiz; Manel Monsonet, Departament de reported receiving personal fees from Lundbeck EU-GEI High Risk Study Group: Philip McGuire, Psicologia Clínica i de la Salut (Universitat and personal fees from Angelini Pharma outside the Department of Psychosis Studies, Institute of Autònoma de Barcelona); Lídia Hinojosa, submitted work. Dr Bressan reported receiving Psychiatry, Psychology & Neuroscience, King's Departament de Psicologia Clínica i de la Salut grants from Fundação de Amparo à Pesquisa do College London; Lucia R. Valmaggia, Department of (Universitat Autònoma de Barcelona); Anna Estado de São Paulo, The Brazilian National Council Psychology, Institute of Psychiatry, Psychology & Racioppi, Departament de Psicologia Clínica i de la for Scientific and Technological Development, Neuroscience, King's College London; Matthew J. Salut (Universitat Autònoma de Barcelona); European Research Council, and Medical Research Kempton, Department of Psychosis Studies, Thomas R. Kwapil, Department of Psychology, Council UK during the conduct of the study; Institute of Psychiatry, Psychology & Neuroscience, University of Illinois at Urbana-Champaign; personal fees and nonfinancial support from King's College London; Maria Calem, Department of Mathilde Kazes, University of Paris, GHU-Paris, Janssen; personal fees from Pfizer; and personal Psychosis Studies, Institute of Psychiatry, Sainte-Anne, C’JAAD, Hospitalo-Universitaire fees from Sanofi-Aventis outside the submitted Psychology & Neuroscience, King's College London; department SHU; Claire Daban, University of Paris, work. Dr Krebs reported receiving grants PHRC Stefania Tognin, Department of Psychosis Studies, GHU-Paris, Sainte-Anne, C’JAAD, 07-118 (ICAAR study) from the French Health Institute of Psychiatry, Psychology & Neuroscience, Hospitalo-Universitaire department SHU; Julie Ministry during the conduct of the study; King's College London; Gemma Modinos, Bourgin, University of Paris, GHU-Paris, participating on the boards of Roche and Janssen; Department of Psychosis Studies, Institute of Sainte-Anne, C’JAAD, Hospitalo-Universitaire and receiving financial support from Janssen, Psychiatry, Psychology & Neuroscience, King's department SHU; Olivier Gay, University of Paris, Otsuka Lundbeck Alliance, and EIsai for conference College London; Lieuwe de Haan, AMC, Academic GHU-Paris, Sainte-Anne, C’JAAD, or dissemination initiatives. Dr Glenthøj reported Psychiatric Centre, Department Early Psychosis; Hospitalo-Universitaire department SHU; Célia being the leader of Lundbeck Foundation Centre of Mark van der Gaag, VU University, Faculty of Mam-Lam-Fook, University of Paris, GHU-Paris, Excellence for Clinical Intervention and Behavioural and Movement Sciences, Department Sainte-Anne, C’JAAD, Hospitalo-Universitaire Neuropsychiatric Schizophrenia Research; receiving 198 JAMA Psychiatry February 2020 Volume 77, Number 2 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 07/30/2020 Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis Original Investigation Research department SHU; Marie-Odile Krebs, University of 6. Holt DJ, Kunkel L, Weiss AP, et al. Increased 21. Kohler CG, Richard JA, Brensinger CM, et al. Paris, GHU-Paris, Sainte-Anne, C’JAAD, medial temporal lobe activation during the passive Facial emotion perception differs in young persons Hospitalo-Universitaire department SHU; Dorte viewing of emotional and neutral facial expressions at genetic and clinical high-risk for psychosis. Nordholm, Mental Health Center Copenhagen and in schizophrenia. Schizophr Res. 2006;82(2-3): Psychiatry Res. 2014;216(2):206-212. doi:10.1016/ CINS, Mental Health Center Glostrup, Mental Health 153-162. doi:10.1016/j.schres.2005.09.021 j.psychres.2014.01.023 Services in the Capital Region of Copenhagen, 7. Hall J, Whalley HC, McKirdy JW, et al. 22. Barbato M, Liu L, Cadenhead KS, et al. Theory University of Copenhagen; Lasse Randers, Mental Overactivation of fear systems to neutral faces in of mind, emotion recognition and social perception Health Center Copenhagen and CINS, Mental schizophrenia. Biol Psychiatry. 2008;64(1):70-73. in individuals at clinical high risk for psychosis: Health Center Glostrup, Mental Health Services in doi:10.1016/j.biopsych.2007.12.014 findings from the NAPLS-2 cohort. Schizophr Res the Capital Region of Copenhagen, University of 8. Surguladze S, Russell T, Kucharska-Pietura K, Cogn. 2015;2(3):133-139. doi:10.1016/j.scog.2015.04. Copenhagen; Kristine Krakauer, Mental Health et al. A reversal of the normal pattern of 004 Center Copenhagen and CINS, Mental Health parahippocampal response to neutral and fearful 23. Seiferth NY, Pauly K, Habel U, et al. Increased Center Glostrup, Mental Health Services in the faces is associated with reality distortion in neural response related to neutral faces in Capital Region of Copenhagen, University of schizophrenia. Biol Psychiatry. 2006;60(5):423-431. individuals at risk for psychosis. Neuroimage. 2008; Copenhagen; Louise Birkedal Glenthøj, Department doi:10.1016/j.biopsych.2005.11.021 40(1):289-297. doi:10.1016/j.neuroimage.2007.11. of Psychosis Studies, Institute of Psychiatry, 9. Taylor SF, Phan KL, Britton JC, Liberzon I. Neural 020 Psychology & Neuroscience, King's College London, response to emotional salience in schizophrenia. 24. Modinos G, Tseng HH, Falkenberg I, Samson C, and CNSR and CINS, University of Copenhagen; Neuropsychopharmacology. 2005;30(5):984-995. McGuire P, Allen P. Neural correlates of aberrant Birte Glenthøj, CNSR and CINS, University of doi:10.1038/sj.npp.1300679 emotional salience predict psychotic symptoms and Copenhagen; Merete Nordentoft, Mental Health 10. Anticevic A, Van Snellenberg JX, Cohen RE, global functioning in high-risk and first-episode Center Copenhagen and CINS, Mental Health Repovs G, Dowd EC, Barch DM. Amygdala psychosis. Soc Cogn Affect Neurosci. 2015;10(10): Center Glostrup, Mental Health Services in the recruitment in schizophrenia in response to 1429-1436. doi:10.1093/scan/nsv035 Capital Region of Copenhagen, University of aversive emotional material: a meta-analysis of 25. Lee SY, Bang M, Kim KR, et al. Impaired facial Copenhagen; Stephan Ruhrmann, Department neuroimaging studies. Schizophr Bull. 2012;38(3): emotion recognition in individuals at ultra-high risk of Psychiatry and Psychotherapy, University of 608-621. doi:10.1093/schbul/sbq131 for psychosis and with first-episode schizophrenia, Cologne; Dominika Gebhard, Department 11. Li HJ, Chan RC, Gong QY, et al. Facial emotion and their associations with neurocognitive deficits of Psychiatry and Psychotherapy, University of processing in patients with schizophrenia and their and self-reported schizotypy. Schizophr Res. 2015; Cologne; Julia Arnhold, Psyberlin; Joachim non-psychotic siblings: a functional magnetic 165(1):60-65. doi:10.1016/j.schres.2015.03.026 Klosterkötter, Department of Psychiatry and resonance imaging study. Schizophr Res. 2012;134 26. Maat A, van Haren NEM, Bartholomeusz CF, Psychotherapy, University of Cologne; Gabriele (2-3):143-150. doi:10.1016/j.schres.2011.10.019 Kahn RS, Cahn W. Emotion recognition and theory Sachs, Department of Psychiatry and 12. Taylor SF, Kang J, Brege IS, Tso IF, Hosanagar A, of mind are related to gray matter volume of the Psychotherapy, Medical University of Vienna; Johnson TD. Meta-analysis of functional prefrontal cortex in schizophrenia. Eur Iris Lasser, Department of Psychiatry and neuroimaging studies of emotion perception and Neuropsychopharmacol. 2016;26(2):255-264. Psychotherapy, Medical University of Vienna; experience in schizophrenia. Biol Psychiatry. 2012; doi:10.1016/j.euroneuro.2015.12.013 Bernadette Winklbaur, Department of Psychiatry 71(2):136-145. doi:10.1016/j.biopsych.2011.09.007 27. van Os J, Rutten BP, Myin-Germeys I, et al; and Psychotherapy, Medical University of Vienna; European Network of National Networks studying 13. Seiferth NY, Pauly K, Kellermann T, et al. Philippe A Delespaul, Department of Psychiatry and Neuronal correlates of facial emotion discrimination Gene-Environment Interactions in Schizophrenia Neuropsychology, School for Mental Health and in early onset schizophrenia. (EU-GEI). Identifying gene-environment Neuroscience, and Mondriaan Mental Health Trust; Neuropsychopharmacology. 2009;34(2):477-487. interactions in schizophrenia: contemporary Bart Rutten, Department of Psychiatry and doi:10.1038/npp.2008.93 challenges for integrated, large-scale investigations. Neuropsychology, School for Mental Health and Schizophr Bull. 2014;40(4):729-736. doi:10.1093/ 14. Aleman A, Kahn RS. Strange feelings: do schbul/sbu069 Neuroscience; Jim van Os, Department of Psychosis amygdala abnormalities dysregulate the emotional Studies, Institute of Psychiatry, Psychology & brain in schizophrenia? Prog Neurobiol. 2005;77(5): 28. Yung AR, Yuen HP, McGorry PD, et al. Mapping Neuroscience, King's College London, and 283-298. doi:10.1016/j.pneurobio.2005.11.005 the onset of psychosis: the Comprehensive Department of Psychiatry, UMC Utrecht Brain Assessment of At-Risk Mental States. Aust N Z J 15. Hooker CI, Bruce L, Lincoln SH, Fisher M, Psychiatry. 2005;39(11-12):964-971. doi:10.1080/ Center, Utrecht University Medical Centre. Vinogradov S. Theory of mind skills are related to j.1440-1614.2005.01714.x gray matter volume in the ventromedial prefrontal REFERENCES 29. Mallett R. Sociodemographic Schedule. London, cortex in schizophrenia. Biol Psychiatry. 2011;70 1. Kring AM, Caponigro JM. Emotion in (12):1169-1178. doi:10.1016/j.biopsych.2011.07.027 UK: Section of Social Psychiatry, Institute of schizophrenia: where feeling meets thinking. Curr Psychiatry; 1997. 16. Yamada M, Hirao K, Namiki C, et al. Social Dir Psychol Sci. 2010;19(4):255-259. doi:10.1177/ 30. Hall RC. Global assessment of functioning: cognition and frontal lobe pathology in 0963721410377599 a modified scale. Psychosomatics. 1995;36(3):267- schizophrenia: a voxel-based morphometric study. 2. Häfner H, Maurer K, Löffler W, an der Heiden W, Neuroimage. 2007;35(1):292-298. doi:10.1016/ 275. doi:10.1016/S0033-3182(95)71666-8 Hambrecht M, Schultze-Lutter F. Modeling the early j.neuroimage.2006.10.046 31. Velthorst E, Levine SZ, Henquet C, et al. To cut a course of schizophrenia. Schizophr Bull. 2003;29 short test even shorter: reliability and validity of a 17. Benes FM. Amygdalocortical circuitry in (2):325-340. doi:10.1093/oxfordjournals.schbul. brief assessment of intellectual ability in schizophrenia: from circuits to molecules. a007008 schizophrenia–a control-case family study. Cogn Neuropsychopharmacology. 2010;35(1):239-257. 3. Tso IF, Grove TB, Taylor SF. Emotional experience doi:10.1038/npp.2009.116 Neuropsychiatry. 2013;18(6):574-593. doi:10.1080/ predicts social adjustment independent of 13546805.2012.731390 18. Du Y, Grace AA. Peripubertal diazepam neurocognition and social cognition in 32. Barkus E, Lewis S. Schizotypy and administration prevents the emergence of schizophrenia. Schizophr Res. 2010;122(1-3):156-163. psychosis-like experiences from recreational dopamine system hyperresponsivity in the MAM doi:10.1016/j.schres.2009.12.007 cannabis in a non-clinical sample. Psychol Med. developmental disruption model of schizophrenia. 4. Nikolaides A, Miess S, Auvera I, Müller R, Neuropsychopharmacology. 2013;38(10):1881-1888. 2008;38(9):1267-1276. doi:10.1017/ Klosterkötter J, Ruhrmann S. Restricted attention to doi:10.1038/npp.2013.101 S0033291707002619 social cues in schizophrenia patients. Eur Arch 33. Robins LN, Wing J, Wittchen HU, et al. The 19. Du Y, Grace AA. Amygdala hyperactivity in Psychiatry Clin Neurosci. 2016;266(7):649-661. composite international diagnostic interview: an MAM model of schizophrenia is normalized by doi:10.1007/s00406-016-0705-6 epidemiologic instrument suitable for use in peripubertal diazepam administration. 5. Holt DJ, Weiss AP, Rauch SL, et al. Sustained Neuropsychopharmacology. 2016;41(10):2455-2462. conjunction with different diagnostic systems and activation of the hippocampus in response to doi:10.1038/npp.2016.42 in different cultures. Arch Gen Psychiatry. 1988;45 fearful faces in schizophrenia. Biol Psychiatry. (12):1069-1077. doi:10.1001/archpsyc.1988. 20. Phillips LK, Seidman LJ. Emotion processing in 01800360017003 2005;57(9):1011-1019. doi:10.1016/j.biopsych.2005. persons at risk for schizophrenia. Schizophr Bull. 01.033 34. van ’t Wout M, Aleman A, Kessels RP, Larøi F, 2008;34(5):888-903. doi:10.1093/schbul/sbn085 Kahn RS. Emotional processing in a non-clinical jamapsychiatry.com (Reprinted) JAMA Psychiatry February 2020 Volume 77, Number 2 199 Downloaded From: https://jamanetwork.com/ on 07/30/2020 Research Original Investigation Association of Adverse Outcomes With Emotion Processing in Individuals at Risk for Psychosis psychosis-prone sample. Schizophr Res. 2004;68 49. Mechelli A, Riecher-Rössler A, Meisenzahl EM, 62. Addington J, Cornblatt BA, Cadenhead KS, (2-3):271-281. doi:10.1016/j.schres.2003.09.006 et al. Neuroanatomical abnormalities that predate et al. At clinical high risk for psychosis: outcome for 35. Catalan A, Gonzalez de Artaza M, Bustamante the onset of psychosis: a multicenter study. Arch nonconverters. Am J Psychiatry. 2011;168(8):800- S, et al. Differences in facial emotion recognition Gen Psychiatry. 2011;68(5):489-495. doi:10.1001/ 805. doi:10.1176/appi.ajp.2011.10081191 between first episode psychosis, borderline archgenpsychiatry.2011.42 63. Comparelli A, Corigliano V, De Carolis A, et al. personality disorder and healthy controls. PLoS One. 50. Singer T, Critchley HD, Preuschoff K. Emotion recognition impairment is present early 2016;11(7):e0160056. doi:10.1371/journal.pone. A common role of insula in feelings, empathy and and is stable throughout the course of 0160056 uncertainty. Trends Cogn Sci. 2009;13(8):334-340. schizophrenia. Schizophr Res. 2013;143(1):65-69. 36. Fett AK, Maat A; GROUP Investigators. Social doi:10.1016/j.tics.2009.05.001 doi:10.1016/j.schres.2012.11.005 cognitive impairments and psychotic symptoms: 51. Fusar-Poli P, Placentino A, Carletti F, et al. 64. Chan RC, Li H, Cheung EF, Gong QY. Impaired what is the nature of their association? Schizophr Bull. Functional atlas of emotional faces processing: facial emotion perception in schizophrenia: 2013;39(1):77-85. doi:10.1093/schbul/sbr058 a voxel-based meta-analysis of 105 functional a meta-analysis. Psychiatry Res. 2010;178(2):381-390. 37. Levin HS. Hamsher KdS, Benton AL. A short magnetic resonance imaging studies. J Psychiatry doi:10.1016/j.psychres.2009.03.035 form of the test of facial recognition for clinical use. Neurosci. 2009;34(6):418-432. 65. Thompson A, Papas A, Bartholomeusz C, et al. J Psychol. 1975;91(2):223-228. doi:10.1080/ 52. Hooker C, Park S. Emotion processing and its Social cognition in clinical “at risk” for psychosis and 00223980.1975.9923946 relationship to social functioning in schizophrenia first episode psychosis populations. Schizophr Res. 38. Ashburner J. VBM tutorial. https://www.fil.ion. patients. Psychiatry Res. 2002;112(1):41-50. doi:10. 2012;141(2-3):204-209. doi:10.1016/j.schres.2012.08. ucl.ac.uk/~john/misc/VBMclass15.pdf. Published 1016/S0165-1781(02)00177-4 007 March 12, 2015. Accessed May 8, 2017. 53. Fett AK, Viechtbauer W, Dominguez MD, Penn 66. Edwards J, Pattison PE, Jackson HJ, Wales RJ. 39. First M, Spitzer R, Gibbon M, Williams JBW. DL, van Os J, Krabbendam L. The relationship Facial affect and affective prosody recognition in Structured Clinical Interview for DSM-IV Axis I between neurocognition and social cognition first-episode schizophrenia. Schizophr Res. 2001;48 Disorders (SCID). New York, NY: New York State with functional outcomes in schizophrenia: (2-3):235-253. doi:10.1016/S0920-9964(00) Psychiatric Institute Biometrics Research; 1995. a meta-analysis. Neurosci Biobehav Rev. 2011;35(3): 00099-2 573-588. doi:10.1016/j.neubiorev.2010.07.001 67. Allott KA, Schäfer MR, Thompson A, et al. 40. Yung AR, Phillips LJ, McGorry PD, et al. Prediction of psychosis: a step towards indicated 54. Tamminga CA, Stan AD, Wagner AD. The Emotion recognition as a predictor of transition to a prevention of schizophrenia. Br J Psychiatry Suppl. hippocampal formation in schizophrenia. Am J psychotic disorder in ultra-high risk participants. 1998;172(33):14-20. doi:10.1192/ Psychiatry. 2010;167(10):1178-1193. doi:10.1176/appi. Schizophr Res. 2014;153(1-3):25-31. doi:10.1016/ S0007125000297602 ajp.2010.09081187 j.schres.2014.01.037 41. Allen P, Chaddock CA, Egerton A, et al. 55. Heckers S, Konradi C. Hippocampal pathology 68. Addington J, Piskulic D, Perkins D, Woods SW, Functional outcome in people at high risk for in schizophrenia. Curr Top Behav Neurosci. 2010;4: Liu L, Penn DL. Affect recognition in people at psychosis predicted by thalamic glutamate levels 529-553. doi:10.1007/7854_2010_43 clinical high risk of psychosis. Schizophr Res. 2012; and prefronto-striatal activation. Schizophr Bull. 56. Anvari AA, Friedman LA, Greenstein D, 140(1-3):87-92. doi:10.1016/j.schres.2012.06.012 2015;41(2):429-439. doi:10.1093/schbul/sbu115 Gochman P, Gogtay N, Rapoport JL. Hippocampal 69. van Donkersgoed RJ, Wunderink L, Nieboer R, 42. Bossong MG, Antoniades M, Azis M, et al. volume change relates to clinical outcome in Aleman A, Pijnenborg GH. Social cognition in Association of hippocampal glutamate levels with childhood-onset schizophrenia. Psychol Med. 2015; individuals at ultra-high risk for psychosis: adverse outcomes in individuals at clinical high risk 45(12):2667-2674. doi:10.1017/S0033291715000677 a meta-analysis. PLoS One. 2015;10(10):e0141075. for psychosis. JAMA Psychiatry. 2019;76(2):199-207. 57. Andreasen NC, Nopoulos P, Magnotta V, doi:10.1371/journal.pone.0141075 doi:10.1001/jamapsychiatry.2018.3252 Pierson R, Ziebell S, Ho BC. Progressive brain 70. Kim SH, Hamann S. Neural correlates of 43. Eklund A, Nichols TE, Knutsson H. Cluster change in schizophrenia: a prospective longitudinal positive and negative emotion regulation. J Cogn failure: why fMRI inferences for spatial extent have study of first-episode schizophrenia. Biol Psychiatry. Neurosci. 2007;19(5):776-798. doi:10.1162/jocn.2007. inflated false-positive rates. Proc Natl Acad Sci U S A. 2011;70(7):672-679. doi:10.1016/j.biopsych.2011.05. 19.5.776 2016;113(28):7900-7905. doi:10.1073/pnas. 017 71. Fusar-Poli P, Bonoldi I, Yung AR, et al. Predicting 1602413113 58. Lappin JM, Morgan C, Chalavi S, et al. Bilateral psychosis: meta-analysis of transition outcomes in 44. Amodio DM, Frith CD. Meeting of minds: the hippocampal increase following first-episode individuals at high clinical risk. Arch Gen Psychiatry. medial frontal cortex and social cognition. Nat Rev psychosis is associated with good clinical, 2012;69(3):220-229. doi:10.1001/archgenpsychiatry. Neurosci. 2006;7(4):268-277. doi:10.1038/nrn1884 functional and cognitive outcomes. Psychol Med. 2011.1472 2014;44(6):1279-1291. doi:10.1017/ 72. Schultze-Lutter F, Rahman J, Ruhrmann S, et al. 45. Kober H, Barrett LF, Joseph J, Bliss-Moreau E, S0033291713001712 Lindquist K, Wager TD. Functional grouping and Duration of unspecific prodromal and clinical high cortical-subcortical interactions in emotion: 59. Allen P, Chaddock CA, Egerton A, et al. Resting risk states, and early help-seeking in first-admission a meta-analysis of neuroimaging studies. Neuroimage. hyperperfusion of the hippocampus, midbrain, and psychosis patients. Soc Psychiatry Psychiatr 2008;42(2):998-1031. doi:10.1016/j.neuroimage. basal ganglia in people at high risk for psychosis. Epidemiol. 2015;50(12):1831-1841. doi:10.1007/ 2008.03.059 Am J Psychiatry. 2016;173(4):392-399. doi:10.1176/ s00127-015-1093-3 appi.ajp.2015.15040485 73. Nelson B, Yuen HP, Wood SJ, et al. Long-term 46. Phelps EA, LeDoux JE. Contributions of the amygdala to emotion processing: from animal 60. Koutsouleris N, Kambeitz-Ilankovic L, follow-up of a group at ultra high risk (“prodromal”) models to human behavior. Neuron. 2005;48(2): Ruhrmann S, et al; PRONIA Consortium. Prediction for psychosis: the PACE 400 study. JAMA Psychiatry. 175-187. doi:10.1016/j.neuron.2005.09.025 models of functional outcomes for individuals in the 2013;70(8):793-802. doi:10.1001/jamapsychiatry. clinical high-risk state for psychosis or with 2013.1270 47. Phillips ML, Drevets WC, Rauch SL, Lane R. recent-onset depression: a multimodal, multisite Neurobiology of emotion perception II: implications 74. Takao H, Hayashi N, Ohtomo K. Effects of study machine learning analysis. JAMA Psychiatry. 2018; design in multi-scanner voxel-based morphometry for major psychiatric disorders. Biol Psychiatry. 75(11):1156-1172. doi:10.1001/jamapsychiatry.2018. 2003;54(5):515-528. doi:10.1016/S0006-3223(03) studies. Neuroimage. 2014;84:133-140. doi:10.1016/ 2165 j.neuroimage.2013.08.046 00171-9 61. Lin A, Wood SJ, Nelson B, et al. Neurocognitive 75. Fortin JP, Cullen N, Sheline YI, et al. 48. Lieberman JA, Girgis RR, Brucato G, et al. predictors of functional outcome two to 13 years Hippocampal dysfunction in the pathophysiology Harmonization of cortical thickness measurements after identification as ultra-high risk for psychosis. across scanners and sites. Neuroimage. 2018;167: of schizophrenia: a selective review and hypothesis Schizophr Res. 2011;132(1):1-7. doi:10.1016/j.schres. for early detection and intervention. Mol Psychiatry. 104-120. doi:10.1016/j.neuroimage.2017.11.024 2011.06.014 2018;23(8):1764-1772. doi:10.1038/mp.2017.249 200 JAMA Psychiatry February 2020 Volume 77, Number 2 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 07/30/2020