[Skip to Navigation]
Sign In
Figure 1.  Consolidated Standards of Reporting Trials (CONSORT) Participant Flow Diagram
Consolidated Standards of Reporting Trials (CONSORT) Participant Flow Diagram

AP indicates antipsychotic; DUP, duration of untreated psychosis; MIPT, AP medication plus intensive psychotherapy; MRI, magnetic resonance imaging; PIPT, placebo plus intensive psychotherapy.

Figure 2.  Baseline and Short-term Longitudinal Functional Connectivity Changes in Antipsychotic (AP)–Naive Patients
Baseline and Short-term Longitudinal Functional Connectivity Changes in Antipsychotic (AP)–Naive Patients

Legend shows the Schaefer et al28 network and Tian et al29 subcortex parcellations (Scale II). Each panel contains (left to right) (1) visualization of the significant Network Based Statistic (NBS) subnetwork, with nodes colored by network and weighted by degree; (2) heatmap of the proportion of edges within the NBS component that fall within each of the canonical networks, quantified using raw (upper triangle) and normalized (lower triangle) proportions; and (3) surface renderings depicting the number of edges in the NBS subnetwork attached to each brain region (NBS degree). a Indicates anterior; AMY, amygdala; CAU, caudate nucleus; d, dorsal; DA, dorsoanterior; DMN, default mode network; DorsAttn, dorsal attention network; DP, dorsoposterior; FPN, frontoparietal network; GP, globus pallidus; HIP, hippocampus; l, lateral; Lim, cortical limbic network; m, medial; MTL, medial-temporal lobe (amygdala and hippocampus); NAc, nucleus accumbens; p, posterior; SomMot, somatomotor network; Stri, striatum; PUT, putamen; THA, thalamus; Thal, thalamus; v, ventral; VA, ventroanterior; VentAttn, ventral attention network; Vis, visual network; and VP, ventroposterior.

Figure 3.  Association Between Short-term Functional Connectivity (FC) Changes and Behavioral Outcomes
Association Between Short-term Functional Connectivity (FC) Changes and Behavioral Outcomes

Panels depict the edges where change in FC was significantly correlated (false discovery rate [FDR]–corrected) with the brain change canonical variate. Each panel contains (from left to right) (1) a visualization of the edges within the Network Based Statistic (NBS) component that correlated with the brain change variate, with nodes colored by canonical network and weighted by the number of connections to which they are attached in the NBS subnetwork; (2) a heatmap of raw (upper triangle) and normalized (lower triangle) proportions of edges in NBS component that fall within each canonical network; (3) a ridge plot of the distribution of correlations between each edge of the NBS component and the brain variate in the canonical correlation analysis (CCA), separated by network and sorted by the median value; (4) surface renderings colored by the number of edges in the NBS subnetwork attached to each node (NBS degree). Amyg indicates amygdala; DMN, default mode network; DorsAttn, dorsal attention network; FPN, frontoparietal network; Hippo, hippocampus; Lim, cortical limbic network; MTL, medial-temporal lobe (amyg and hippo); SomMot, somatomotor network; Stri, striatum; Thal, thalamus; VentAttn, ventral attention network; Vis, visual network.

Figure 4.  Longitudinal Functional Connectivity (FC) Changes Due to Antipsychotic (AP) Treatment
Longitudinal Functional Connectivity (FC) Changes Due to Antipsychotic (AP) Treatment

Each panel shows changes from baseline to 3 months and contains (from left to right) (1) a visualization of the significant Network Based Statistic (NBS) subnetwork, with nodes colored by network and weighted by degree; (2) a heatmap of the proportion of edges within the NBS component that fall within each of the canonical networks, quantified using raw (upper triangle) and normalized (lower triangle) proportions; and (3) surface renderings depicted the number of edges in the NBS subnetwork attached to each brain region (NBS degree). DMN indicates default mode network; DorsAttn, dorsal attention network; FPN, frontoparietal network; Lim, cortical limbic network; MTL, medial-temporal lobe (amygdala and hippocampus); SomMot, somatomotor network; Stri, striatum; Thal, thalamus; VentAttn, ventral attention network; Vis, visual network.

Table.  Participant Characteristics
Participant Characteristics
Supplement 3.

eMethods 1. Additional Details

eMethods 2. Trial Safety Procedures

eMethods 3. Antipsychotic and Concomitant Medication Details

eTable 1. Cumulative Antipsychotic Exposure (olanzapine equivalents, mg) in Placebo Group (PIPT) and Medication Group (MIPT)

eTable 2. Percentage of Each Treatment Group Included in Analyses Who Received Nonantipsychotic Psychotropic Medication During the Primary Treatment Period

eFigure 1. Histogram of Cumulative Dose (olanzapine equivalents, mg) Exposure in the Placebo and Medication Groups Across the Three Time Points

eMethods 4. MRI Acquisition Details

eMethods 5. Image Processing and Quality Control

eMethods 6. Further Details on Statistical Analysis

eMethods 7. Demographic and Sample Characteristics

eAppendix 1. Antipsychotic-Naïve Results (Baseline)

eAppendix 2. Antipsychotic-Naïve Results (Baseline to 3 mo)

eAppendix 3. Antipsychotic-Related Results (Baseline to 3 mo)

eAppendix 4. Long-Term Changes in Antipsychotic-Naïve Patients Compared to Healthy Controls (Baseline to 12 mo)

eAppendix 5. Long-Term Antipsychotic-Related Changes (Baseline to 12 mo)

eAppendix 6. Association Between Long-Term Changes and Symptoms and Functioning

eFigure 2. Canonical Correlation Analysis Between Short-Term Functional Connectivity Changes and Behavioral Outcomes

eFigure 3. Longer-Term Effects in Antipsychotic-Naïve Patients (A, B) and Longer-Term Effects Related to Antipsychotic Medication (C,D)

eFigure 4. Antipsychotic-Naïve and Antipsychotic-Treated Results at a Network Level (FWE P < .05) at Primary Thresholds of P < .01 and P < .001

eFigure 5. Baseline to 12 Months Antipsychotic-Naïve and Antipsychotic-Related Effects After Removing the 9 Subjects in the PIPT Group Who Were Exposed to Antipsychotics

eFigure 6. Baseline to 3 Months Antipsychotic-Naïve and Antipsychotic-Related Effects After Adjusting for Nonantipsychotic Psychotropic Medication Use

eReferences

1.
Rubinov  M, Bullmore  E.  Fledgling pathoconnectomics of psychiatric disorders.   Trends Cogn Sci. 2013;17(12):641-647. doi:10.1016/j.tics.2013.10.007 PubMedGoogle ScholarCrossref
2.
Fornito  A, Zalesky  A, Pantelis  C, Bullmore  ET.  Schizophrenia, neuroimaging and connectomics.   Neuroimage. 2012;62(4):2296-2314. doi:10.1016/j.neuroimage.2011.12.090 PubMedGoogle ScholarCrossref
3.
Fornito  A, Harrison  BJ, Goodby  E,  et al.  Functional dysconnectivity of corticostriatal circuitry as a risk phenotype for psychosis.   JAMA Psychiatry. 2013;70(11):1143-1151. doi:10.1001/jamapsychiatry.2013.1976 PubMedGoogle ScholarCrossref
4.
Anticevic  A, Hu  X, Xiao  Y,  et al.  Early-course unmedicated schizophrenia patients exhibit elevated prefrontal connectivity associated with longitudinal change.   J Neurosci. 2015;35(1):267-286. doi:10.1523/JNEUROSCI.2310-14.2015 PubMedGoogle ScholarCrossref
5.
Baker  JT, Dillon  DG, Patrick  LM,  et al.  Functional connectomics of affective and psychotic pathology.   Proc Natl Acad Sci U S A. 2019;116(18):9050-9059. doi:10.1073/pnas.1820780116 PubMedGoogle ScholarCrossref
6.
Baker  JT, Holmes  AJ, Masters  GA,  et al.  Disruption of cortical association networks in schizophrenia and psychotic bipolar disorder.   JAMA Psychiatry. 2014;71(2):109-118. doi:10.1001/jamapsychiatry.2013.3469 PubMedGoogle ScholarCrossref
7.
Rapoport  JL, Giedd  JN, Gogtay  N.  Neurodevelopmental model of schizophrenia: update 2012.   Mol Psychiatry. 2012;17(12):1228-1238. doi:10.1038/mp.2012.23 PubMedGoogle ScholarCrossref
8.
Howes  OD, Kapur  S.  The dopamine hypothesis of schizophrenia: version III–the final common pathway.   Schizophr Bull. 2009;35(3):549-562. doi:10.1093/schbul/sbp006 PubMedGoogle ScholarCrossref
9.
McCutcheon  RA, Jauhar  S, Pepper  F,  et al.  The topography of striatal dopamine and symptoms in psychosis: an integrative positron emission tomography and magnetic resonance imaging study.   Biol Psychiatry Cogn Neurosci Neuroimaging. 2020;5(11):1040-1051. doi:10.1016/j.bpsc.2020.04.004 PubMedGoogle Scholar
10.
Howes  OD, Murray  RM.  Schizophrenia: an integrated sociodevelopmental-cognitive model.   Lancet. 2014;383(9929):1677-1687. doi:10.1016/S0140-6736(13)62036-X PubMedGoogle ScholarCrossref
11.
González-Vivas  C, Soldevila-Matías  P, Sparano  O,  et al.  Longitudinal studies of functional magnetic resonance imaging in first-episode psychosis: a systematic review.   Eur Psychiatry. 2019;59:60-69. doi:10.1016/j.eurpsy.2019.04.009 PubMedGoogle ScholarCrossref
12.
Duan  X, Hu  M, Huang  X,  et al.  Effects of risperidone monotherapy on the default-mode network in antipsychotic-naïve first-episode schizophrenia: posteromedial cortex heterogeneity and relationship with the symptom improvements.   Schizophr Res. 2020;218:201-208. doi:10.1016/j.schres.2020.01.001 PubMedGoogle ScholarCrossref
13.
Kraguljac  NV, White  DM, Hadley  N,  et al.  Aberrant hippocampal connectivity in unmedicated patients with schizophrenia and effects of antipsychotic medication: a longitudinal resting state functional MRI study.   Schizophr Bull. 2016;42(4):1046-1055. doi:10.1093/schbul/sbv228 PubMedGoogle ScholarCrossref
14.
O’Neill  A, Mechelli  A, Bhattacharyya  S.  Dysconnectivity of large-scale functional networks in early psychosis: a meta-analysis.   Schizophr Bull. 2019;45(3):579-590. doi:10.1093/schbul/sby094 PubMedGoogle ScholarCrossref
15.
Anticevic  A, Haut  K, Murray  JD,  et al.  Association of thalamic dysconnectivity and conversion to psychosis in youth and young adults at elevated clinical risk.   JAMA Psychiatry. 2015;72(9):882-891. doi:10.1001/jamapsychiatry.2015.0566 PubMedGoogle ScholarCrossref
16.
Avram  M, Brandl  F, Bäuml  J, Sorg  C.  Cortico-thalamic hypo- and hyperconnectivity extend consistently to basal ganglia in schizophrenia.   Neuropsychopharmacology. 2018;43(11):2239-2248. doi:10.1038/s41386-018-0059-z PubMedGoogle ScholarCrossref
17.
Guo  W, Liu  F, Chen  J,  et al.  Hyperactivity of the default-mode network in first-episode, drug-naive schizophrenia at rest revealed by family-based case-control and traditional case-control designs.   Medicine (Baltimore). 2017;96(13):e6223. doi:10.1097/MD.0000000000006223 PubMedGoogle Scholar
18.
Sarpal  DK, Robinson  DG, Lencz  T,  et al.  Antipsychotic treatment and functional connectivity of the striatum in first-episode schizophrenia.   JAMA Psychiatry. 2015;72(1):5-13. doi:10.1001/jamapsychiatry.2014.1734 PubMedGoogle ScholarCrossref
19.
Hu  ML, Zong  XF, Zheng  JJ,  et al.  Short-term effects of risperidone monotherapy on spontaneous brain activity in first-episode treatment-naïve schizophrenia patients: a longitudinal fMRI study.   Sci Rep. 2016;6:34287. doi:10.1038/srep34287 PubMedGoogle ScholarCrossref
20.
Bergé  D, Carmona  S, Salgado  P, Rovira  M, Bulbena  A, Vilarroya  O.  Limbic activity in antipsychotic naïve first-episode psychotic subjects during facial emotion discrimination.   Eur Arch Psychiatry Clin Neurosci. 2014;264(4):271-283. doi:10.1007/s00406-013-0465-5 PubMedGoogle ScholarCrossref
21.
Reske  M, Kellermann  T, Habel  U,  et al.  Stability of emotional dysfunctions? a long-term fMRI study in first-episode schizophrenia.   J Psychiatr Res. 2007;41(11):918-927. doi:10.1016/j.jpsychires.2007.02.009 PubMedGoogle ScholarCrossref
22.
Li  F, Lui  S, Yao  L,  et al.  Longitudinal changes in resting-state cerebral activity in patients with first-episode schizophrenia: a 1-year follow-up functional MR imaging study.   Radiology. 2016;279(3):867-875. doi:10.1148/radiol.2015151334 PubMedGoogle ScholarCrossref
23.
Chopra  S, Fornito  A, Francey  SM,  et al.  Differentiating the effect of antipsychotic medication and illness on brain volume reductions in first-episode psychosis: a longitudinal, randomised, triple-blind, placebo-controlled MRI study.   Neuropsychopharmacology. Published online February 26, 2021. doi:10.1038/s41386-021-00980-0PubMedGoogle Scholar
24.
Francey  SM, O’Donoghue B, Nelson B,  et al. Psychosocial intervention with or without antipsychotic medication for first-episode psychosis: a randomized noninferiority clinical trial. SBO. Published online March 20, 2020:sgaa015. doi:10.1093/schizbullopen/sgaa015
25.
Schaefer  A, Kong  R, Gordon  EM,  et al.  Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI.   Cereb Cortex. 2018;28(9):3095-3114. doi:10.1093/cercor/bhx179 PubMedGoogle ScholarCrossref
26.
Tian  Y, Margulies  DS, Breakspear  M, Zalesky  A.  Topographic organization of the human subcortex unveiled with functional connectivity gradients.   Nat Neurosci. 2020;23(11):1421-1432. doi:10.1038/s41593-020-00711-6 PubMedGoogle ScholarCrossref
27.
Guillaume  B, Hua  X, Thompson  PM, Waldorp  L, Nichols  TE; Alzheimer’s Disease Neuroimaging Initiative.  Fast and accurate modelling of longitudinal and repeated measures neuroimaging data.   Neuroimage. 2014;94:287-302. doi:10.1016/j.neuroimage.2014.03.029 PubMedGoogle ScholarCrossref
28.
Zalesky  A, Fornito  A, Bullmore  ET.  Network-based statistic: identifying differences in brain networks.   Neuroimage. 2010;53(4):1197-1207. doi:10.1016/j.neuroimage.2010.06.041 PubMedGoogle ScholarCrossref
29.
Alberton  BAV, Nichols  TE, Gamba  HR, Winkler  AM.  Multiple testing correction over contrasts for brain imaging.   Neuroimage. 2020;216:116760. doi:10.1016/j.neuroimage.2020.116760 PubMedGoogle Scholar
30.
Winkler  AM, Renaud  O, Smith  SM, Nichols  TE.  Permutation inference for canonical correlation analysis.   Neuroimage. 2020;220:117065. doi:10.1016/j.neuroimage.2020.117065 PubMedGoogle Scholar
31.
Stephan  KE, Friston  KJ, Frith  CD.  Dysconnection in schizophrenia: from abnormal synaptic plasticity to failures of self-monitoring.   Schizophr Bull. 2009;35(3):509-527. doi:10.1093/schbul/sbn176 PubMedGoogle ScholarCrossref
32.
Pettersson-Yeo  W, Allen  P, Benetti  S, McGuire  P, Mechelli  A.  Dysconnectivity in schizophrenia: where are we now?   Neurosci Biobehav Rev. 2011;35(5):1110-1124. doi:10.1016/j.neubiorev.2010.11.004 PubMedGoogle ScholarCrossref
33.
Nelson  B, Fornito  A, Harrison  BJ,  et al.  A disturbed sense of self in the psychosis prodrome: linking phenomenology and neurobiology.   Neurosci Biobehav Rev. 2009;33(6):807-817. doi:10.1016/j.neubiorev.2009.01.002 PubMedGoogle ScholarCrossref
34.
Nelson  B, Whitford  TJ, Lavoie  S, Sass  LA.  What are the neurocognitive correlates of basic self-disturbance in schizophrenia? integrating phenomenology and neurocognition: part 2 (aberrant salience).   Schizophr Res. 2014;152(1):20-27. doi:10.1016/j.schres.2013.06.033 PubMedGoogle ScholarCrossref
35.
Catani  M, Dell’acqua  F, Thiebaut de Schotten  M.  A revised limbic system model for memory, emotion and behaviour.   Neurosci Biobehav Rev. 2013;37(8):1724-1737. doi:10.1016/j.neubiorev.2013.07.001 PubMedGoogle ScholarCrossref
36.
Kapur  S.  Psychosis as a state of aberrant salience: a framework linking biology, phenomenology, and pharmacology in schizophrenia.   Am J Psychiatry. 2003;160(1):13-23. doi:10.1176/appi.ajp.160.1.13 PubMedGoogle ScholarCrossref
37.
Kee  KS, Green  MF, Mintz  J, Brekke  JS.  Is emotion processing a predictor of functional outcome in schizophrenia?   Schizophr Bull. 2003;29(3):487-497. doi:10.1093/oxfordjournals.schbul.a007021 PubMedGoogle ScholarCrossref
38.
Yung  AR, McGorry  PD.  The prodromal phase of first-episode psychosis: past and current conceptualizations.   Schizophr Bull. 1996;22(2):353-370. doi:10.1093/schbul/22.2.353 PubMedGoogle ScholarCrossref
39.
Blessing  EM, Murty  VP, Zeng  B, Wang  J, Davachi  L, Goff  DC.  Anterior hippocampal-cortical functional connectivity distinguishes antipsychotic naïve first-episode psychosis patients from controls and may predict response to second-generation antipsychotic treatment.   Schizophr Bull. 2020;46(3):680-689. doi:10.1093/schbul/sbz076 PubMedGoogle ScholarCrossref
40.
Nelson  EA, Kraguljac  NV, Maximo  JO,  et al.  Hippocampal dysconnectivity and altered glutamatergic modulation of the default mode network: a combined resting-state connectivity and magnetic resonance spectroscopy study in schizophrenia.   Biol Psychiatry Cogn Neurosci Neuroimaging. 2020;S2451-9022(20)30108-7. doi:10.1016/j.bpsc.2020.04.014 PubMedGoogle Scholar
41.
Xia  CH, Ma  Z, Ciric  R,  et al.  Linked dimensions of psychopathology and connectivity in functional brain networks.   Nat Commun. 2018;9(1):3003. doi:10.1038/s41467-018-05317-y PubMedGoogle ScholarCrossref
42.
Dandash  O, Fornito  A, Lee  J,  et al.  Altered striatal functional connectivity in subjects with an at-risk mental state for psychosis.   Schizophr Bull. 2014;40(4):904-913. doi:10.1093/schbul/sbt093 PubMedGoogle ScholarCrossref
43.
Li  A, Zalesky  A, Yue  W,  et al.  A neuroimaging biomarker for striatal dysfunction in schizophrenia.   Nat Med. 2020;26(4):558-565. doi:10.1038/s41591-020-0793-8 PubMedGoogle ScholarCrossref
44.
Sabaroedin  K, Tiego  J, Parkes  L,  et al.  Functional connectivity of corticostriatal circuitry and psychosis-like experiences in the general community.   Biol Psychiatry. 2019;86(1):16-24. doi:10.1016/j.biopsych.2019.02.013 PubMedGoogle ScholarCrossref
45.
O’Donoghue  B, Francey  SM, Nelson  B,  et al.  Staged treatment and acceptability guidelines in early psychosis study (STAGES): a randomized placebo controlled trial of intensive psychosocial treatment plus or minus antipsychotic medication for first-episode psychosis with low-risk of self-harm or aggression: study protocol and baseline characteristics of participants.   Early Interv Psychiatry. 2019;13(4):953-960. doi:10.1111/eip.12716 PubMedGoogle ScholarCrossref
46.
Mason  L, Peters  ER, Dima  D, Williams  SC, Kumari  V.  Cognitive behavioral therapy normalizes functional connectivity for social threat in psychosis.   Schizophr Bull. 2016;42(3):684-692. doi:10.1093/schbul/sbv153 PubMedGoogle ScholarCrossref
47.
Pantelis  C, Velakoulis  D, McGorry  PD,  et al.  Neuroanatomical abnormalities before and after onset of psychosis: a cross-sectional and longitudinal MRI comparison.   Lancet. 2003;361(9354):281-288. doi:10.1016/S0140-6736(03)12323-9 PubMedGoogle ScholarCrossref
48.
Grace  AA.  Dysregulation of the dopamine system in the pathophysiology of schizophrenia and depression.   Nat Rev Neurosci. 2016;17(8):524-532. doi:10.1038/nrn.2016.57 PubMedGoogle ScholarCrossref
49.
Röder  CH, Hoogendam  JM, van der Veen  FM.  FMRI, antipsychotics and schizophrenia: influence of different antipsychotics on BOLD-signal.   Curr Pharm Des. 2010;16(18):2012-2025. doi:10.2174/138161210791293088 PubMedGoogle ScholarCrossref
50.
van Dellen  E, Börner  C, Schutte  M,  et al.  Functional brain networks in the schizophrenia spectrum and bipolar disorder with psychosis.   NPJ Schizophr. 2020;6(1):22. doi:10.1038/s41537-020-00111-6 PubMedGoogle ScholarCrossref
51.
Hwang  K, Bertolero  MA, Liu  WB, D’Esposito  M.  The human thalamus is an integrative hub for functional brain networks.   J Neurosci. 2017;37(23):5594-5607. doi:10.1523/JNEUROSCI.0067-17.2017 PubMedGoogle ScholarCrossref
52.
Creese  I, Burt  DR, Snyder  SH.  Dopamine receptor binding predicts clinical and pharmacological potencies of antischizophrenic drugs.   Science. 1976;192(4238):481-483. doi:10.1126/science.3854 PubMedGoogle ScholarCrossref
53.
Carlsson  A, Waters  N, Carlsson  ML.  Neurotransmitter interactions in schizophrenia–therapeutic implications.   Biol Psychiatry. 1999;46(10):1388-1395. doi:10.1016/S0006-3223(99)00117-1 PubMedGoogle ScholarCrossref
54.
NICE. Psychosis and schizophrenia in adults: prevention and management. Accessed January 15, 2021. https://www.nice.org.uk/guidance/cg178/chapter/1-Recommendations
55.
Johnstone  EC, Crow  TJ, Frith  CD, Owens  DG.  The Northwick Park “functional” psychosis study: diagnosis and treatment response.   Lancet. 1988;2(8603):119-125. doi:10.1016/S0140-6736(88)90682-4 PubMedGoogle ScholarCrossref
56.
Legge  SE, Jones  HJ, Kendall  KM,  et al.  Association of genetic liability to psychotic experiences with neuropsychotic disorders and traits.   JAMA Psychiatry. 2019;76(12):1256-1265. doi:10.1001/jamapsychiatry.2019.2508 PubMedGoogle ScholarCrossref
57.
Torrent  C, Reinares  M, Martinez-Arán  A,  et al; PEPS group.  Affective versus non-affective first episode psychoses: a longitudinal study.   J Affect Disord. 2018;238:297-304. doi:10.1016/j.jad.2018.06.005 PubMedGoogle ScholarCrossref
58.
Jauhar  S, Krishnadas  R, Nour  MM, Cunningham-Owens  D, Johnstone  EC, Lawrie  SM.  Is there a symptomatic distinction between the affective psychoses and schizophrenia? a machine learning approach.   Schizophr Res. 2018;202:241-247. doi:10.1016/j.schres.2018.06.070 PubMedGoogle ScholarCrossref
59.
Craddock  N, Owen  MJ.  The Kraepelinian dichotomy—going, going... but still not gone.   Br J Psychiatry. 2010;196(2):92-95. doi:10.1192/bjp.bp.109.073429 PubMedGoogle ScholarCrossref
60.
Brockington  IF, Kendell  RE, Wainwright  S, Hillier  VF, Walker  J.  The distinction between the affective psychoses and schizophrenia.   Br J Psychiatry. 1979;135:243-248. doi:10.1192/bjp.135.3.243 PubMedGoogle ScholarCrossref
61.
Leucht  S, Kane  JM, Kissling  W, Hamann  J, Etschel  E, Engel  R.  Clinical implications of Brief Psychiatric Rating Scale scores.   Br J Psychiatry. 2005;187:366-371. doi:10.1192/bjp.187.4.366 PubMedGoogle ScholarCrossref
62.
Henry  LP, Amminger  GP, Harris  MG,  et al.  The EPPIC follow-up study of first-episode psychosis: longer-term clinical and functional outcome 7 years after index admission.   J Clin Psychiatry. 2010;71(6):716-728. doi:10.4088/JCP.08m04846yel PubMedGoogle ScholarCrossref
Original Investigation
June 23, 2021

Functional Connectivity in Antipsychotic-Treated and Antipsychotic-Naive Patients With First-Episode Psychosis and Low Risk of Self-harm or Aggression: A Secondary Analysis of a Randomized Clinical Trial

Author Affiliations
  • 1Turner Institute for Brain and Mental Health, Monash University School of Psychological Sciences, Clayton, Victoria, Australia
  • 2Monash Biomedical Imaging, Monash University, Clayton, Victoria, Australia
  • 3Orygen, Parkville, Victoria, Australia
  • 4Centre for Youth Mental Health, The University of Melbourne, Melbourne, Victoria, Australia
  • 5Melbourne Neuropsychiatry Centre, Department of Psychiatry, University of Melbourne & Melbourne Health, Melbourne, Victoria, Australia
  • 6Melbourne School of Population and Global Health, The University of Melbourne, Melbourne, Victoria, Australia
  • 7Department of Social Work, Monash University, Caulfield, Victoria, Australia
  • 8University of Birmingham School of Psychology, Edgbaston, United Kingdom
JAMA Psychiatry. 2021;78(9):994-1004. doi:10.1001/jamapsychiatry.2021.1422
Key Points

Question  Are there different patterns of brain connectivity in antipsychotic-treated and antipsychotic-naive patients with psychosis, and how do these patterns evolve over time?

Findings  In this secondary analysis of a triple-blind, longitudinal, placebo-controlled, randomized clinical trial in antipsychotic-naive patients, patients showed widespread brain dysconnectivity at baseline relative to healthy controls. Patients treated with therapy and placebo or with therapy and antipsychotics showed evidence of circuit-specific and treatment-specific normalization of connectivity over time.

Meaning  In this study, some prominent baseline connectivity differences in first-episode psychosis normalized with both psychosocial therapy and placebo early in the illness course, with antipsychotics exerting circuit-specific effects.

Abstract

Importance  Altered functional connectivity (FC) is a common finding in resting-state functional magnetic resonance imaging (rs-fMRI) studies of people with psychosis, yet how FC disturbances evolve in the early stages of illness, and how antipsychotic treatment influences these disturbances, remains unknown.

Objective  To investigate longitudinal FC changes in antipsychotic-naive and antipsychotic-treated patients with first-episode psychosis (FEP).

Design, Setting, and Participants  This secondary analysis of a triple-blind, randomized clinical trial was conducted over a 5-year recruitment period between April 2008 and December 2016 with 59 antipsychotic-naive patients with FEP receiving either a second-generation antipsychotic or a placebo pill over a treatment period of 6 months. Participants were required to have low suicidality and aggression, to have a duration of untreated psychosis of less than 6 months, and to be living in stable accommodations with social support. Both FEP groups received intensive psychosocial therapy. A healthy control group was also recruited. Participants completed rs-fMRI scans at baseline, 3 months, and 12 months. Data were analyzed from May 2019 to August 2020.

Interventions  Resting-state functional MRI was used to probe brain FC. Patients received either a second-generation antipsychotic or a matched placebo tablet. Both patient groups received a manualized psychosocial intervention.

Main Outcomes and Measures  The primary outcomes of this analysis were to investigate (1) FC differences between patients and controls at baseline; (2) FC changes in medicated and unmedicated patients between baseline and 3 months; and (3) associations between longitudinal FC changes and clinical outcomes. An additional aim was to investigate long-term FC changes at 12 months after baseline. These outcomes were not preregistered.

Results  Data were analyzed for 59 patients (antipsychotic medication plus psychosocial treatment: 28 [47.5%]; mean [SD] age, 19.5 [3.0] years; 15 men [53.6%]; placebo plus psychosocial treatment: 31 [52.5%]; mean [SD] age, 18.8 [2.7]; 16 men [51.6%]) and 27 control individuals (mean [SD] age, 21.9 [1.9] years). At baseline, patients showed widespread functional dysconnectivity compared with controls, with reductions predominantly affecting interactions between the default mode network, limbic systems, and the rest of the brain. From baseline to 3 months, patients receiving placebo showed increased FC principally within the same systems; some of these changes correlated with improved clinical outcomes (canonical correlation analysis R = 0.901; familywise error–corrected P = .005). Antipsychotic exposure was associated with increased FC primarily between the thalamus and the rest of the brain.

Conclusions and Relevance  In this secondary analysis of a clinical trial, antipsychotic-naive patients with FEP showed widespread functional dysconnectivity at baseline, followed by an early normalization of default mode network and cortical limbic dysfunction in patients receiving placebo and psychosocial intervention. Antipsychotic exposure was associated with FC changes concentrated on thalamocortical networks.

Trial Registration  ACTRN12607000608460

Introduction

Psychosis is thought to emerge from disrupted communication between large-scale functional brain networks.1,2 Resting-state functional magnetic resonance imaging (rs-fMRI) has been widely used to map altered functional connectivity (FC) in patients across different illness stages and within specific corticosubcortical3,4 and corticocortical networks.5,6 Numerous causes of altered FC in patients with psychosis have been proposed, including abnormal neurodevelopment,7 dopamine dysfunction,8,9 traumatic early-life experiences,10 and the neuromodulatory effect of antipsychotic medication.11 In particular, widespread and early treatment of patients with first-episode psychosis (FEP) with antipsychotics has made it difficult to disentangle the effects of medication from other factors, such as illness progression or nonpharmacological treatments, on FC.

Cross-sectional studies have shown widespread hypoconnectivity in antipsychotic-naive patients compared with healthy controls, with decreased connectivity between subcortical, frontal, and temporal regions.12-14 Hyperconnectivity has also been reported between the thalamus and primary sensory cortices15 and within the default mode network.15-17 These results suggest that not all FC abnormalities can be explained by medication. Nonetheless, longitudinal studies provide evidence of FC normalization after antipsychotic exposure,11 particularly within fronto-striatal-thalamic circuits12,18,19 and corticolimbic18,20,21 and corticocortical systems,4,22 which is sometimes correlated with symptom improvement.4,18

To our knowledge, no prior study has compared longitudinal FC changes in patients who were medicated and nonmedicated, making it impossible to disentangle FC changes attributable to antipsychotic medication vs other factors, such as the natural course of the illness or adjunct interventions. This comparison can be done only through a randomized placebo-controlled study of antipsychotic-naive patients. We recently used such a design to show, using a noninferiority design, that the placebo group had comparable clinical and functional outcomes to the medicated group23 and that patients who were medicated and unmedicated show different trajectories of gray matter volume, with antipsychotics normalizing basal ganglia volume in early illness stages.23

Here, we report a secondary analysis of FC in this cohort. Patients with FEP were scanned at baseline while antipsychotic-naive and then again at 3 and 12 months after randomization to receive treatment with either antipsychotic medication plus intensive psychosocial treatment (MIPT) or placebo plus intensive psychosocial treatment (PIPT). Our primary aims were to (1) identify FC differences between patients and controls at baseline; (2) compare FC changes over time in antipsychotic-treated patients and antipsychotic-naive patients during the initial stages of psychotic illness; and (3) examine whether any observed FC changes were associated with changes in symptoms or functional outcomes. We additionally investigated longer-term FC changes at the 12-month follow-up, after a period of time in which some participants in the placebo group had been exposed to antipsychotics.

Methods
Study Design

This secondary analysis of a randomized clinical trial was conducted over a 5-year recruitment period between April 2008 and December 2016 at the Early Psychosis Prevention and Intervention Centre, which is part of Orygen Youth Health in Melbourne, Australia. Patients were randomized to 1 of 2 groups: 1 group was given MIPT, and the other was given PIPT. A third healthy control group who received no intervention was also recruited. For both patient groups, the controlled treatment period spanned 6 months. Clinical assessments and MRIs were conducted at baseline, 3 months, and a final follow-up of 12 months. The preregistered primary and secondary clinical trial outcome measures were the Social and Occupational Functioning Assessment Scale (SOFAS) and the Brief Psychiatric Rating Scale (BPRS) total scores, respectively. The trial was registered with the Australian New Zealand Clinical Trials Registry in November 2007 and received ethics approval from the Melbourne Health Human Research and Ethics committee. We report here on a secondary analysis of FC measures, which was not part of the preregistered protocol. All participants gave written informed consent after having the study fully explained to them; parental consent was also obtained for participants younger than 18 years. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline. Further research and safety information can be found in the trial protocol in Supplement 1, the MRI protocol in Supplement 2, and eMethods 1 to 3, eTables 1 and 2, and eFigure 1 in Supplement 3.24

Participants

Patients were aged 15 to 25 years and were experiencing a first episode of psychosis, defined as fulfilling the structured clinical interview for DSM-IV criteria for a psychotic spectrum disorder. Additional inclusion criteria to minimize risk included (1) duration of untreated psychosis of less than 6 months, (2) living in stable accommodations, (3) low risk to self or others, and (4) no or minimal previous exposure to antipsychotic medication. Healthy control participants were aged between 18 and 25 years and were psychiatrically, neurologically, and medically healthy (Figure 1 contains a detailed participant flow diagram).

MRI Acquisition and Processing

Whole-brain T2*-weighted echo-planar images and anatomical T1-weighted scans were acquired for each participant using a 3-T Siemens Trio Tim scanner (Siemens Healthineers). Of a total of 202 rs-fMRI data sets, 193 scans survived our quality control and standard image processing procedures. To generate whole-brain FC matrices, we parceled each individual’s normalized scans into 300 cortical25 and 32 subcortical regions.26 Further details can be found in eMethods 4 and 5 in Supplement 3.

Statistical Analysis

Nonparametric mixed-effects marginal models27 were used to analyze brain-wide FC changes across the 3 groups (MIPT, PIPT, and controls) and 3 time points (baseline, 3 months, and 12 months). The Network Based Statistic28 (NBS) was used to perform familywise error-corrected (FWE) inference at the level of connected components of edges, with the primary component-forming threshold, τ, set to P < .05. Further statistical details and results for τ = 0.01 and τ = 0.001 are reported in eMethods 6 and 7, eAppendixes 1 to 3, and eFigure 4 in Supplement 3.

We evaluated NBS results using a Bonferroni-corrected threshold of FWE P < .016, adjusted for 3 key contrasts29: (1) a contrast assessing baseline differences between healthy controls and patients; (2) a contrast isolating differential FC changes over time in the PIPT group (excluding 3-month scans for 4 patients in the PIPT group who were exposed to antipsychotics), compared with the healthy control group; and (3) a contrast isolating the specific effects of antipsychotic treatment by examining differential FC changes over time in the MIPT group compared with both the PIPT and healthy control groups. We repeated the same procedure for the secondary analysis of longer-term antipsychotic-naive and antipsychotic-related effects after the trial treatment period, this time including the 12-month time point (eMethods 6 and 7 in Supplement 3). We used nonparametric canonical correlation analysis30 to investigate associations between FC changes (ΔFC) within any identified NBS subnetworks and changes in the 2 preregistered24 trial outcome measures (SOFAS and BPRS) across all patients (details in eMethods 6 and 7 in Supplement 3). All statistical analyses were conducted in R Studio, version 1.3.1073 (R Studio Team). All P values were 2-sided, and data were analyzed from May 2019 to August 2020.

Results
Antipsychotic-Naive Effects
Baseline

A total of 59 patients (antipsychotic medication plus psychosocial treatment: 28 [47.5%]; mean [SD] age, 19.5 [3.0] years; 15 men [53.6%]; placebo plus psychosocial treatment: 31 [52.5%]; mean [SD] age, 18.8 [2.7] years; 16 men [51.6%]) and 27 controls (mean [SD] age, 21.9 [1.9] years; 17 women [63.0%]) completed a baseline scan; 39 patients (66.1%) and 21 controls (77.8%) completed a 3-month follow-up scan; and 35 patients (59.3%) and 21 controls (77.8%) completed a 12-month follow-up scan. Further information on demographic characteristics of the MIPT, PIPT, and control groups are presented in the Table and described in eMethods 6 and 7 in Supplement 3.

We identified a single NBS component showing widespread FC differences in patients compared with controls, comprising 4087 edges linking all 316 regions (FWE P = .02). Within this network, 2195 edges (53.7%) showed lower FC, and 1892 edges (46.3%) showed higher FC in patients (Figure 2A and B).

Using raw proportions, connections associated with reduced FC in patients were predominantly concentrated in the default mode network (DMN) (Figure 2A). Using normalized proportions, which emphasize network involvement after accounting for differences in network size (eMethods 6 and 7 in Supplement 3), limbic and medial-temporal lobe areas were also strongly implicated. At a regional level, the left superior temporal pole, inferior frontal and parietal cortices, and the right postcentral gyrus, ventroanterior thalamus, and posterior caudate were among the areas most strongly implicated in the network of FC reductions (Figure 2A). Edges in which FC was higher in patients showed a more homogeneous distribution across different networks (Figure 2B). Strongly implicated regions included the right posterior caudate, dorsoanterior and ventroanterior thalamus, left frontal eye field, left cuneus, right anterior caudate, and superior lateral occipital cortex.

Baseline to 3 Months

We identified a single widespread NBS component showing an altered FC trajectory in the PIPT group compared with controls, comprising 4128 edges linking all 316 regions (FWE P = .007). Within this network, 1913 edges (46.3%) showed decreasing FC and 2215 edges (53.7%) showed increasing FC in patients over time (Figure 2C and D). The edges showing declining FC over time in patients in the PIPT group had a diffuse distribution across different systems, with some evidence for preferential involvement of FC between the default mode and somatomotor systems (Figure 2C). The right posterior and anterior middle temporal gyrus, dorsoanterior thalamus, bilateral precuneus and posterior cingulate cortex, and left precentral gyrus were among the areas linked to many of the FC reductions over time (Figure 2C).

Edges showing increased FC in patients in the PIPT group over time were strongly concentrated in the DMN, using both raw and normalized counts. Normalized counts additionally implicated the cortical limbic system. Both of these networks showed reduced FC in patients compared with controls at baseline. Regions showing many FC increases over time included the right polar and lateral occipital cortices; bilateral postcentral, precentral, and lingual gyri; and the left lateral occipital and parietal cortices (Figure 2D).

Associations With Clinical Outcomes

A canonical correlation analysis including 12 principal components each explaining greater than 2% of variance in ΔFC identified a single significant canonical variate linking FC changes to the primary trial outcome measures (eFigure 2 in Supplement 3) (FWE P = .005; R = 0.901). The canonical loadings indicate that this variate was strongly associated with reducing SOFAS over time (r = −0.99) and moderately associated with increasing BPRS (r = 0.49). A total of 179 edges showed significant correlations with the ΔFC variate (false discovery rate, threshold P < .05; only the edges in the network that survived P < .05 were retained), such that 59 edges correlated positively (0.32 < r < 0.66) (Figure 3A) and 120 edges correlated negatively (−0.41 < r < −0.65) (Figure 3B). Positively correlated edges (Figure 3A) largely implicated links between sensorimotor/visual and association and subcortical systems, particularly the striatum. Regions with the largest number of positively correlated edges were the right nucleus accumbens core, left posterior cingulate, and posterior caudate. Negatively correlated edges (Figure 3B) strongly implicated links between somatomotor and visual networks. Regions with the largest number of negatively correlated edges included bilateral precentral and postcentral gyrus, occipital pole, and lingual gyrus. Thus, across both PIPT and MIPT patients, worse clinical outcome was associated with increased FC between striatal and somatomotor regions, coupled with decreased FC between sensory and motor systems.

Antipsychotic-Related Changes
Baseline to 3 Months

We identified a single NBS component showing an altered FC trajectory in the MIPT group compared with the PIPT and control groups, comprising 634 edges and including all 316 regions (FWE P < .001) (Figure 4 and eFigure 6 in the Supplement). This network indicates that antipsychotic exposure is generally associated with more FC increases (446 edges) than decreases (188 edges) over time. The subnetwork showing increased FC over time strongly implicates the thalamus and its connections to all other networks, particularly when considering normalized counts (Figure 4A). Regionally, the right lingual gyrus, bilateral occipital pole, left superior frontal gyrus, and right precuneus cortex featured prominently in this subnetwork. Edges showing medication-related FC decreases are more diffusely spread across the networks (Figure 4B). Regionally, decreases predominantly involve the left posterior hippocampus and left inferior frontal gyrus. We found no significant associations between medication-related ΔFC and change in primary outcome measures at 3 months.

Longer-term Illness-Related and Antipsychotic-Related Changes (12-Month Follow-up)

Long-term changes in FC in antipsychotic-naive patients were circumscribed to a relatively small subset of edges showing large effects concentrated predominantly in the default mode and somatomotor networks (eAppendixes 4-6 and eFigure 3 in Supplement 3). Long-term medication-related changes paralleled the 3-month results (eFigures 3 and 5 in Supplement 3), with more FC increases than decreases over time and a preferential involvement of the DMN and cortical limbic network. Notably, these networks showed lower FC in patients at baseline (Figure 2A).

Discussion

This secondary analysis of a triple-blind, placebo-controlled, randomized clinical trial allowed us to disentangle longitudinal FC changes in antipsychotic-treated and antipsychotic-naive patients during the early stages of psychosis. We identified widespread FC changes in antipsychotic-naive patients within the first few months of an initial psychotic episode, involving both abnormally increasing and decreasing FC over time when compared with healthy controls, with evidence for a preferential involvement and recovery of default mode and cortical limbic systems in patients receiving psychosocial treatment and placebo. In contrast, antipsychotic-related changes predominantly involved increased FC between the thalamus and other brain networks. Across all patients, longitudinal change in FC over the first 3 months between subcortical, sensorimotor, and association networks was most strongly related to change in clinical outcome. We additionally found more circumscribed changes in antipsychotic-naive patients affecting medial parietal cortex over a 12-month period. Long-term antipsychotic effects were primarily associated with increased FC in cortical limbic and association systems.

Baseline Connectivity Differences Between Antipsychotic-Naive Patients and Controls

Our baseline analysis identified a broadly distributed subnetwork of altered FC in antipsychotic-naive patients with FEP, involving an approximately equal proportion of edges showing hyperconnectivity and hypoconnectivity relative to controls. This broad distribution of FC differences is consistent with a general disconnection hypothesis of psychotic illness,2,31,32 in which symptoms are tied to a breakdown of interregional communication. Our results suggest that, from the outset of illness, the nature of this breakdown is both complex and widespread.

Within this affected subnetwork, FC reductions are predominantly concentrated within the DMN and its interactions with other systems. We also found evidence for preferential involvement of connectivity between the medial-temporal lobe (MTL), including both the amygdala and hippocampus, and cortical limbic systems with the rest of the brain, after accounting for differences in network size. Together, these findings point to a prominent role for dysconnectivity of default mode and cortical limbic systems in FEP. Given the presumed role of the DMN in self-referential and reflective processes33,34 and the cortical limbic and MTL systems in emotional function,35 our findings align with views that psychotic symptoms may arise from aberrant salience attributed to internal representations,36 as well as evidence for prominent impairments of emotional processing and expression in people with psychotic disorders.37,38 Moreover, our findings of hypoconnectivity across the DMN align with recent studies examining antipsychotic-naive patients.12,39,40 However, DMN dysfunction is a common finding in many different psychiatric disorders and may reflect a general vulnerability to psychopathology.41 At the regional level, the striatum and thalamus were attached to a large number of affected edges, which is consistent with previous research implicating the fronto-striato-thalamic circuits in the pathogenesis of psychosis3,42-44 and our previous investigation showing volumetric decline within the pallidum in patients in the PIPT group.23

Connectivity Changes in Antipsychotic-Naive Patients

From baseline to 3 months, we also observed widespread FC changes in patients in the PIPT group relative to controls. The FC reductions were spread relatively evenly across networks, whereas FC increases were strongly concentrated on connections linking the DMN and cortical limbic system to the rest of the brain. Critically, some of these changes in FC, particularly those affecting links between sensory, subcortical, and association networks, were related to improved symptom ratings and functional outcomes over time.

The findings of increasing FC in the DMN and cortical limbic systems, together with the prominent reductions in these systems observed at baseline, suggest a normalization of FC over the first few months of illness. This recovery may be due to the placebo or psychosocial intervention.24,45 Although there is some evidence that such interventions can normalize FC across the limbic system and associated networks,46 previous research has not been able to rule out the effect of antipsychotic medication. Our findings suggest that either placebo administration or psychosocial intervention alone, or in combination, may be sufficient to normalize aberrant FC in these systems.

Our analysis of the 12-month time point found fewer and more circumscribed FC changes over time in the PIPT group relative to the control group, suggesting that the majority of FC changes occurred within the first 3 months of the study, when patients were most intensively engaged with the psychosocial therapy. However, definitively testing this hypothesis requires comparison with an additional treatment group receiving either basic support or no intervention. Such a design is difficult to justify ethically in patients who are unmedicated. Notably, our prior investigation of this cohort23 found that pallidal volume in patients receiving placebo normalized by the 12-month follow-up. In the current study, we see a normalization of altered FC in patients receiving placebo by 3 months, suggesting that normalization of function may precede structure.

The FC within the MTL was lower at baseline and did not show evidence for preferential increases over time, suggesting that reduced MTL FC may represent a core illness-related feature of psychosis that may not be modified by psychosocial treatment or placebo administration. This result aligns with patient studies47 and animal models48 suggesting a primary role for MTL areas in driving the onset of psychotic illness.

Connectivity Changes in Antipsychotic-Treated Patients

Several studies have found that lowered FC at baseline is either normalized or partly normalized after antipsychotic treatment11,18 (although see also Röder et al49 and van Dellen et al50). Accordingly, we show that antipsychotic exposure largely increases FC and preferentially affects connections between the thalamus and the rest of the brain. The thalamus is a globally connected hub that relays multimodal information between diverse functional networks.51 The therapeutic efficacy of antipsychotics is primarily mediated by their antagonism of D2 receptors in the striatum,52 and dopamine dysregulation in psychosis is thought to disrupt striato-thalamic filtering of sensory and limbic information to the cortex.53 Our findings suggest that antipsychotics may partially remediate communication between the thalamus, striatum, and cortex, thereby normalizing information flow within widespread cortico-subcortical systems.

At 12 months, we found evidence for a prolonged increase of FC, mainly in the DMN and cortical limbic systems, in patients who are medicated. An early effect of antipsychotics on thalamocortical signaling may therefore be followed by a more sustained influence on the DMN and cortical limbic FC, which showed prominent illness-related differences at baseline and a partial recovery, correlated with symptom outcome, in the PIPT group. Normalizing communication between limbic/paralimbic systems and the rest of the brain may therefore be a primary therapeutic target in FEP.

On the Generalizability of Our Findings

Our sample included patients with affective and nonaffective psychosis, given that antipsychotics are used in the treatment of both.54,55 Clinical and biological evidence supporting their nosological separation has been mixed,56-60 and our analysis presents a transdiagnostic investigation of the psychosis spectrum. Future work may seek to evaluate differences between distinct illness subtypes.

Our use of a placebo-controlled design required strict safety and ethical guidelines. To comply with these guidelines, patients were required to have low levels of suicidality and aggression, to have a duration of untreated psychosis of less than 6 months, and to be living in stable accommodations with social support. Additionally, patients who showed increased risk at any point or did not improve in clinical symptoms or functioning within 3 months of study intake were discontinued from the trial. These criteria resulted in removal of 17 patients by the 12-month time point.

Limitations

Our study had some limitations. Our strict inclusion criteria raise concerns that our sample comprises patients with a mild form of illness. This seems unlikely, however, as the mean baseline BPRS score of patients in our study would classify them as markedly ill,61 and the mean baseline SOFAS score was consistent with serious functional impairment, being comparable to epidemiologically representative cohorts of patients with FEP (Table).62 It is possible, however, that patients who remained in the study, particularly at the 12-month time point, have a less deteriorating form of psychotic illness. As such, the apparent normalization of specific brain systems seen in patients in the PIPT group by this time point may not generalize to individuals with a more deteriorating form. By the same reasoning, it is also possible that the illness-related differences that we identified represent a conservative estimate of brain dysconnectivity. Future work may extend our approach to include more rigorous monitoring of patients with a higher risk and evaluate the extent to which our findings generalize to that FEP subgroup. We also note that, although rates of substance and tobacco use did not differ between the treatment groups, use of these substances was not controlled for in our analyses.

Conclusions

Antipsychotic-naive patients with FEP showed widespread functional dysconnectivity at baseline, particularly in default mode and cortical limbic systems, with evidence for an improvement of these changes during the first 3 months of illness. Our results also suggest that antipsychotics may normalize dysconnectivity primarily by affecting thalamocortical and cortical limbic networks.

Back to top
Article Information

Accepted for Publication: April 26, 2021.

Published Online: June 23, 2021. doi:10.1001/jamapsychiatry.2021.1422

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Chopra S et al. JAMA Psychiatry.

Corresponding Author: Sidhant Chopra, BSc, Monash University, 770 Blackburn, Clayton, VIC 3168, Australia (sid.chopra@monash.edu).

Author Contributions: Mr Chopra and Dr Fornito had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Chopra, Francey, Yuen, Alvarez-Jimenez, Harrigan, Pantelis, Wood, McGorry, Fornito.

Acquisition, analysis, or interpretation of data: Chopra, Francey, O'Donoghue, Sabaroedin, Arnatkeviciute, Cropley, Nelson, Graham, Baldwin, Tahtalian, Yuen, Allott, Alvarez-Jimenez, Pantelis, Wood, McGorry, Fornito.

Drafting of the manuscript: Chopra, O'Donoghue, Nelson, Fornito.

Critical revision of the manuscript for important intellectual content: Chopra, Francey, Sabaroedin, Arnatkeviciute, Cropley, Nelson, Graham, Baldwin, Tahtalian, Yuen, Allott, Alvarez-Jimenez, Harrigan, Pantelis, Wood, McGorry, Fornito.

Statistical analysis: Chopra, Sabaroedin, Arnatkeviciute, Fornito.

Obtained funding: Francey, Nelson, Allott, Alvarez-Jimenez, Harrigan, McGorry, Fornito.

Administrative, technical, or material support: Chopra, Francey, O'Donoghue, Sabaroedin, Cropley, Graham, Baldwin, Tahtalian, Yuen, Pantelis, Fornito.

Supervision: Francey, Pantelis, McGorry, Fornito.

Conflict of Interest Disclosures: Dr Francey reported receiving grants from the Australian National Health & Medical Research Council (NHMRC) during the conduct of the study. Dr Allott reported receiving grants from the Australian National Health and Medical Research Council during the conduct of the study. Dr Pantelis reported receiving grants from the Australian NHMRC and from the Lundbeck Foundation and personal fees from Lundbeck Australia Pty Ltd Advisory Board for talks presented at educational meetings organized by Lundbeck. Dr McGorry reported receiving grants from the Australian NHMRC, the Colonial Foundation, the National Alliance for Research on Schizophrenia and Depression, the Stanley Foundation, the National Institutes of Health, Wellcome Trust, the Australian and Victorian governments, and Janssen-Cilag (unrestricted investigator-initiated grant) during the conduct of the study; past unrestricted grant funding from Janssen-Cilag, AstraZeneca, Eli Lilly, Novartis, and Pfizer; honoraria for consultancy and teaching from Janssen-Cilag, Eli Lilly, Pfizer, AstraZeneca, Roche, Bristol Myers Squibb, and Lundbeck. No other disclosures were reported.

Funding/Support: This study was supported in part by grants 1105825 and 1196508 from the NHMRC Senior Principal Research Fellowship and the NHMRC L3 Investigator Grant, respectively (Prof Pantelis), and grant 1137687 from the NHMRC Senior Research Fellowship (Prof Nelson). Janssen-Cilag partially supported the early years of this study with an unrestricted investigator-initiated grant and provided risperidone, paliperidone, and matched placebo for the first 30 participants. The study was then funded by grant 1064704 from the Australian NHMRC Project (Dr Francey, Prof Nelson, Dr Allott, Dr Alvarez-Jimenez, Dr Harrigan, Prof McGorry, and Dr Fornito). Control participants were funded by NHMRC project grant 1065742 (Dr Cropley and Prof Pantelis). Dr Cropley was supported by NHMRC EL2 Investigator Grant 1177370.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 4.

Additional Contributions: We thank the clinical staff at Orygen Youth Health: Craig Macneil, DClinPsy, Kingsley Crisp, BSW, Dylan Alexander, MPsych, Tina Proffitt, PhD, Rachel Tindall, PhD, Jennifer Hall, MPsych, Lisa Rumney, MPsych, Franco Scalzo, MPsych, Melissa Pane, BAppSci, Linda Kader, MBBS, MD, Frank Hughes, MSc, Clare Shelton, MPsych, Ryan Kaplan, PhD, David Hallford, MPsych, Bridget Moller, MPsych, Rick Fraser, MBBS, and research assistants Daniela Cagliarini, MPsych, Suzanne Wiltink, MPsych, Janine Ward, MPsych, and Sumudu Mallawaarachichi, MSc.

Additional Information: The trial took place at the Early Psychosis Prevention and Intervention Centre, which is part of Orygen Youth Health, Melbourne, Australia.

References
1.
Rubinov  M, Bullmore  E.  Fledgling pathoconnectomics of psychiatric disorders.   Trends Cogn Sci. 2013;17(12):641-647. doi:10.1016/j.tics.2013.10.007 PubMedGoogle ScholarCrossref
2.
Fornito  A, Zalesky  A, Pantelis  C, Bullmore  ET.  Schizophrenia, neuroimaging and connectomics.   Neuroimage. 2012;62(4):2296-2314. doi:10.1016/j.neuroimage.2011.12.090 PubMedGoogle ScholarCrossref
3.
Fornito  A, Harrison  BJ, Goodby  E,  et al.  Functional dysconnectivity of corticostriatal circuitry as a risk phenotype for psychosis.   JAMA Psychiatry. 2013;70(11):1143-1151. doi:10.1001/jamapsychiatry.2013.1976 PubMedGoogle ScholarCrossref
4.
Anticevic  A, Hu  X, Xiao  Y,  et al.  Early-course unmedicated schizophrenia patients exhibit elevated prefrontal connectivity associated with longitudinal change.   J Neurosci. 2015;35(1):267-286. doi:10.1523/JNEUROSCI.2310-14.2015 PubMedGoogle ScholarCrossref
5.
Baker  JT, Dillon  DG, Patrick  LM,  et al.  Functional connectomics of affective and psychotic pathology.   Proc Natl Acad Sci U S A. 2019;116(18):9050-9059. doi:10.1073/pnas.1820780116 PubMedGoogle ScholarCrossref
6.
Baker  JT, Holmes  AJ, Masters  GA,  et al.  Disruption of cortical association networks in schizophrenia and psychotic bipolar disorder.   JAMA Psychiatry. 2014;71(2):109-118. doi:10.1001/jamapsychiatry.2013.3469 PubMedGoogle ScholarCrossref
7.
Rapoport  JL, Giedd  JN, Gogtay  N.  Neurodevelopmental model of schizophrenia: update 2012.   Mol Psychiatry. 2012;17(12):1228-1238. doi:10.1038/mp.2012.23 PubMedGoogle ScholarCrossref
8.
Howes  OD, Kapur  S.  The dopamine hypothesis of schizophrenia: version III–the final common pathway.   Schizophr Bull. 2009;35(3):549-562. doi:10.1093/schbul/sbp006 PubMedGoogle ScholarCrossref
9.
McCutcheon  RA, Jauhar  S, Pepper  F,  et al.  The topography of striatal dopamine and symptoms in psychosis: an integrative positron emission tomography and magnetic resonance imaging study.   Biol Psychiatry Cogn Neurosci Neuroimaging. 2020;5(11):1040-1051. doi:10.1016/j.bpsc.2020.04.004 PubMedGoogle Scholar
10.
Howes  OD, Murray  RM.  Schizophrenia: an integrated sociodevelopmental-cognitive model.   Lancet. 2014;383(9929):1677-1687. doi:10.1016/S0140-6736(13)62036-X PubMedGoogle ScholarCrossref
11.
González-Vivas  C, Soldevila-Matías  P, Sparano  O,  et al.  Longitudinal studies of functional magnetic resonance imaging in first-episode psychosis: a systematic review.   Eur Psychiatry. 2019;59:60-69. doi:10.1016/j.eurpsy.2019.04.009 PubMedGoogle ScholarCrossref
12.
Duan  X, Hu  M, Huang  X,  et al.  Effects of risperidone monotherapy on the default-mode network in antipsychotic-naïve first-episode schizophrenia: posteromedial cortex heterogeneity and relationship with the symptom improvements.   Schizophr Res. 2020;218:201-208. doi:10.1016/j.schres.2020.01.001 PubMedGoogle ScholarCrossref
13.
Kraguljac  NV, White  DM, Hadley  N,  et al.  Aberrant hippocampal connectivity in unmedicated patients with schizophrenia and effects of antipsychotic medication: a longitudinal resting state functional MRI study.   Schizophr Bull. 2016;42(4):1046-1055. doi:10.1093/schbul/sbv228 PubMedGoogle ScholarCrossref
14.
O’Neill  A, Mechelli  A, Bhattacharyya  S.  Dysconnectivity of large-scale functional networks in early psychosis: a meta-analysis.   Schizophr Bull. 2019;45(3):579-590. doi:10.1093/schbul/sby094 PubMedGoogle ScholarCrossref
15.
Anticevic  A, Haut  K, Murray  JD,  et al.  Association of thalamic dysconnectivity and conversion to psychosis in youth and young adults at elevated clinical risk.   JAMA Psychiatry. 2015;72(9):882-891. doi:10.1001/jamapsychiatry.2015.0566 PubMedGoogle ScholarCrossref
16.
Avram  M, Brandl  F, Bäuml  J, Sorg  C.  Cortico-thalamic hypo- and hyperconnectivity extend consistently to basal ganglia in schizophrenia.   Neuropsychopharmacology. 2018;43(11):2239-2248. doi:10.1038/s41386-018-0059-z PubMedGoogle ScholarCrossref
17.
Guo  W, Liu  F, Chen  J,  et al.  Hyperactivity of the default-mode network in first-episode, drug-naive schizophrenia at rest revealed by family-based case-control and traditional case-control designs.   Medicine (Baltimore). 2017;96(13):e6223. doi:10.1097/MD.0000000000006223 PubMedGoogle Scholar
18.
Sarpal  DK, Robinson  DG, Lencz  T,  et al.  Antipsychotic treatment and functional connectivity of the striatum in first-episode schizophrenia.   JAMA Psychiatry. 2015;72(1):5-13. doi:10.1001/jamapsychiatry.2014.1734 PubMedGoogle ScholarCrossref
19.
Hu  ML, Zong  XF, Zheng  JJ,  et al.  Short-term effects of risperidone monotherapy on spontaneous brain activity in first-episode treatment-naïve schizophrenia patients: a longitudinal fMRI study.   Sci Rep. 2016;6:34287. doi:10.1038/srep34287 PubMedGoogle ScholarCrossref
20.
Bergé  D, Carmona  S, Salgado  P, Rovira  M, Bulbena  A, Vilarroya  O.  Limbic activity in antipsychotic naïve first-episode psychotic subjects during facial emotion discrimination.   Eur Arch Psychiatry Clin Neurosci. 2014;264(4):271-283. doi:10.1007/s00406-013-0465-5 PubMedGoogle ScholarCrossref
21.
Reske  M, Kellermann  T, Habel  U,  et al.  Stability of emotional dysfunctions? a long-term fMRI study in first-episode schizophrenia.   J Psychiatr Res. 2007;41(11):918-927. doi:10.1016/j.jpsychires.2007.02.009 PubMedGoogle ScholarCrossref
22.
Li  F, Lui  S, Yao  L,  et al.  Longitudinal changes in resting-state cerebral activity in patients with first-episode schizophrenia: a 1-year follow-up functional MR imaging study.   Radiology. 2016;279(3):867-875. doi:10.1148/radiol.2015151334 PubMedGoogle ScholarCrossref
23.
Chopra  S, Fornito  A, Francey  SM,  et al.  Differentiating the effect of antipsychotic medication and illness on brain volume reductions in first-episode psychosis: a longitudinal, randomised, triple-blind, placebo-controlled MRI study.   Neuropsychopharmacology. Published online February 26, 2021. doi:10.1038/s41386-021-00980-0PubMedGoogle Scholar
24.
Francey  SM, O’Donoghue B, Nelson B,  et al. Psychosocial intervention with or without antipsychotic medication for first-episode psychosis: a randomized noninferiority clinical trial. SBO. Published online March 20, 2020:sgaa015. doi:10.1093/schizbullopen/sgaa015
25.
Schaefer  A, Kong  R, Gordon  EM,  et al.  Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI.   Cereb Cortex. 2018;28(9):3095-3114. doi:10.1093/cercor/bhx179 PubMedGoogle ScholarCrossref
26.
Tian  Y, Margulies  DS, Breakspear  M, Zalesky  A.  Topographic organization of the human subcortex unveiled with functional connectivity gradients.   Nat Neurosci. 2020;23(11):1421-1432. doi:10.1038/s41593-020-00711-6 PubMedGoogle ScholarCrossref
27.
Guillaume  B, Hua  X, Thompson  PM, Waldorp  L, Nichols  TE; Alzheimer’s Disease Neuroimaging Initiative.  Fast and accurate modelling of longitudinal and repeated measures neuroimaging data.   Neuroimage. 2014;94:287-302. doi:10.1016/j.neuroimage.2014.03.029 PubMedGoogle ScholarCrossref
28.
Zalesky  A, Fornito  A, Bullmore  ET.  Network-based statistic: identifying differences in brain networks.   Neuroimage. 2010;53(4):1197-1207. doi:10.1016/j.neuroimage.2010.06.041 PubMedGoogle ScholarCrossref
29.
Alberton  BAV, Nichols  TE, Gamba  HR, Winkler  AM.  Multiple testing correction over contrasts for brain imaging.   Neuroimage. 2020;216:116760. doi:10.1016/j.neuroimage.2020.116760 PubMedGoogle Scholar
30.
Winkler  AM, Renaud  O, Smith  SM, Nichols  TE.  Permutation inference for canonical correlation analysis.   Neuroimage. 2020;220:117065. doi:10.1016/j.neuroimage.2020.117065 PubMedGoogle Scholar
31.
Stephan  KE, Friston  KJ, Frith  CD.  Dysconnection in schizophrenia: from abnormal synaptic plasticity to failures of self-monitoring.   Schizophr Bull. 2009;35(3):509-527. doi:10.1093/schbul/sbn176 PubMedGoogle ScholarCrossref
32.
Pettersson-Yeo  W, Allen  P, Benetti  S, McGuire  P, Mechelli  A.  Dysconnectivity in schizophrenia: where are we now?   Neurosci Biobehav Rev. 2011;35(5):1110-1124. doi:10.1016/j.neubiorev.2010.11.004 PubMedGoogle ScholarCrossref
33.
Nelson  B, Fornito  A, Harrison  BJ,  et al.  A disturbed sense of self in the psychosis prodrome: linking phenomenology and neurobiology.   Neurosci Biobehav Rev. 2009;33(6):807-817. doi:10.1016/j.neubiorev.2009.01.002 PubMedGoogle ScholarCrossref
34.
Nelson  B, Whitford  TJ, Lavoie  S, Sass  LA.  What are the neurocognitive correlates of basic self-disturbance in schizophrenia? integrating phenomenology and neurocognition: part 2 (aberrant salience).   Schizophr Res. 2014;152(1):20-27. doi:10.1016/j.schres.2013.06.033 PubMedGoogle ScholarCrossref
35.
Catani  M, Dell’acqua  F, Thiebaut de Schotten  M.  A revised limbic system model for memory, emotion and behaviour.   Neurosci Biobehav Rev. 2013;37(8):1724-1737. doi:10.1016/j.neubiorev.2013.07.001 PubMedGoogle ScholarCrossref
36.
Kapur  S.  Psychosis as a state of aberrant salience: a framework linking biology, phenomenology, and pharmacology in schizophrenia.   Am J Psychiatry. 2003;160(1):13-23. doi:10.1176/appi.ajp.160.1.13 PubMedGoogle ScholarCrossref
37.
Kee  KS, Green  MF, Mintz  J, Brekke  JS.  Is emotion processing a predictor of functional outcome in schizophrenia?   Schizophr Bull. 2003;29(3):487-497. doi:10.1093/oxfordjournals.schbul.a007021 PubMedGoogle ScholarCrossref
38.
Yung  AR, McGorry  PD.  The prodromal phase of first-episode psychosis: past and current conceptualizations.   Schizophr Bull. 1996;22(2):353-370. doi:10.1093/schbul/22.2.353 PubMedGoogle ScholarCrossref
39.
Blessing  EM, Murty  VP, Zeng  B, Wang  J, Davachi  L, Goff  DC.  Anterior hippocampal-cortical functional connectivity distinguishes antipsychotic naïve first-episode psychosis patients from controls and may predict response to second-generation antipsychotic treatment.   Schizophr Bull. 2020;46(3):680-689. doi:10.1093/schbul/sbz076 PubMedGoogle ScholarCrossref
40.
Nelson  EA, Kraguljac  NV, Maximo  JO,  et al.  Hippocampal dysconnectivity and altered glutamatergic modulation of the default mode network: a combined resting-state connectivity and magnetic resonance spectroscopy study in schizophrenia.   Biol Psychiatry Cogn Neurosci Neuroimaging. 2020;S2451-9022(20)30108-7. doi:10.1016/j.bpsc.2020.04.014 PubMedGoogle Scholar
41.
Xia  CH, Ma  Z, Ciric  R,  et al.  Linked dimensions of psychopathology and connectivity in functional brain networks.   Nat Commun. 2018;9(1):3003. doi:10.1038/s41467-018-05317-y PubMedGoogle ScholarCrossref
42.
Dandash  O, Fornito  A, Lee  J,  et al.  Altered striatal functional connectivity in subjects with an at-risk mental state for psychosis.   Schizophr Bull. 2014;40(4):904-913. doi:10.1093/schbul/sbt093 PubMedGoogle ScholarCrossref
43.
Li  A, Zalesky  A, Yue  W,  et al.  A neuroimaging biomarker for striatal dysfunction in schizophrenia.   Nat Med. 2020;26(4):558-565. doi:10.1038/s41591-020-0793-8 PubMedGoogle ScholarCrossref
44.
Sabaroedin  K, Tiego  J, Parkes  L,  et al.  Functional connectivity of corticostriatal circuitry and psychosis-like experiences in the general community.   Biol Psychiatry. 2019;86(1):16-24. doi:10.1016/j.biopsych.2019.02.013 PubMedGoogle ScholarCrossref
45.
O’Donoghue  B, Francey  SM, Nelson  B,  et al.  Staged treatment and acceptability guidelines in early psychosis study (STAGES): a randomized placebo controlled trial of intensive psychosocial treatment plus or minus antipsychotic medication for first-episode psychosis with low-risk of self-harm or aggression: study protocol and baseline characteristics of participants.   Early Interv Psychiatry. 2019;13(4):953-960. doi:10.1111/eip.12716 PubMedGoogle ScholarCrossref
46.
Mason  L, Peters  ER, Dima  D, Williams  SC, Kumari  V.  Cognitive behavioral therapy normalizes functional connectivity for social threat in psychosis.   Schizophr Bull. 2016;42(3):684-692. doi:10.1093/schbul/sbv153 PubMedGoogle ScholarCrossref
47.
Pantelis  C, Velakoulis  D, McGorry  PD,  et al.  Neuroanatomical abnormalities before and after onset of psychosis: a cross-sectional and longitudinal MRI comparison.   Lancet. 2003;361(9354):281-288. doi:10.1016/S0140-6736(03)12323-9 PubMedGoogle ScholarCrossref
48.
Grace  AA.  Dysregulation of the dopamine system in the pathophysiology of schizophrenia and depression.   Nat Rev Neurosci. 2016;17(8):524-532. doi:10.1038/nrn.2016.57 PubMedGoogle ScholarCrossref
49.
Röder  CH, Hoogendam  JM, van der Veen  FM.  FMRI, antipsychotics and schizophrenia: influence of different antipsychotics on BOLD-signal.   Curr Pharm Des. 2010;16(18):2012-2025. doi:10.2174/138161210791293088 PubMedGoogle ScholarCrossref
50.
van Dellen  E, Börner  C, Schutte  M,  et al.  Functional brain networks in the schizophrenia spectrum and bipolar disorder with psychosis.   NPJ Schizophr. 2020;6(1):22. doi:10.1038/s41537-020-00111-6 PubMedGoogle ScholarCrossref
51.
Hwang  K, Bertolero  MA, Liu  WB, D’Esposito  M.  The human thalamus is an integrative hub for functional brain networks.   J Neurosci. 2017;37(23):5594-5607. doi:10.1523/JNEUROSCI.0067-17.2017 PubMedGoogle ScholarCrossref
52.
Creese  I, Burt  DR, Snyder  SH.  Dopamine receptor binding predicts clinical and pharmacological potencies of antischizophrenic drugs.   Science. 1976;192(4238):481-483. doi:10.1126/science.3854 PubMedGoogle ScholarCrossref
53.
Carlsson  A, Waters  N, Carlsson  ML.  Neurotransmitter interactions in schizophrenia–therapeutic implications.   Biol Psychiatry. 1999;46(10):1388-1395. doi:10.1016/S0006-3223(99)00117-1 PubMedGoogle ScholarCrossref
54.
NICE. Psychosis and schizophrenia in adults: prevention and management. Accessed January 15, 2021. https://www.nice.org.uk/guidance/cg178/chapter/1-Recommendations
55.
Johnstone  EC, Crow  TJ, Frith  CD, Owens  DG.  The Northwick Park “functional” psychosis study: diagnosis and treatment response.   Lancet. 1988;2(8603):119-125. doi:10.1016/S0140-6736(88)90682-4 PubMedGoogle ScholarCrossref
56.
Legge  SE, Jones  HJ, Kendall  KM,  et al.  Association of genetic liability to psychotic experiences with neuropsychotic disorders and traits.   JAMA Psychiatry. 2019;76(12):1256-1265. doi:10.1001/jamapsychiatry.2019.2508 PubMedGoogle ScholarCrossref
57.
Torrent  C, Reinares  M, Martinez-Arán  A,  et al; PEPS group.  Affective versus non-affective first episode psychoses: a longitudinal study.   J Affect Disord. 2018;238:297-304. doi:10.1016/j.jad.2018.06.005 PubMedGoogle ScholarCrossref
58.
Jauhar  S, Krishnadas  R, Nour  MM, Cunningham-Owens  D, Johnstone  EC, Lawrie  SM.  Is there a symptomatic distinction between the affective psychoses and schizophrenia? a machine learning approach.   Schizophr Res. 2018;202:241-247. doi:10.1016/j.schres.2018.06.070 PubMedGoogle ScholarCrossref
59.
Craddock  N, Owen  MJ.  The Kraepelinian dichotomy—going, going... but still not gone.   Br J Psychiatry. 2010;196(2):92-95. doi:10.1192/bjp.bp.109.073429 PubMedGoogle ScholarCrossref
60.
Brockington  IF, Kendell  RE, Wainwright  S, Hillier  VF, Walker  J.  The distinction between the affective psychoses and schizophrenia.   Br J Psychiatry. 1979;135:243-248. doi:10.1192/bjp.135.3.243 PubMedGoogle ScholarCrossref
61.
Leucht  S, Kane  JM, Kissling  W, Hamann  J, Etschel  E, Engel  R.  Clinical implications of Brief Psychiatric Rating Scale scores.   Br J Psychiatry. 2005;187:366-371. doi:10.1192/bjp.187.4.366 PubMedGoogle ScholarCrossref
62.
Henry  LP, Amminger  GP, Harris  MG,  et al.  The EPPIC follow-up study of first-episode psychosis: longer-term clinical and functional outcome 7 years after index admission.   J Clin Psychiatry. 2010;71(6):716-728. doi:10.4088/JCP.08m04846yel PubMedGoogle ScholarCrossref
×