Development of Proteomic Prediction Models for Transition to Psychotic Disorder in the Clinical High-Risk State and Psychotic Experiences in Adolescence | Adolescent Medicine | JAMA Psychiatry | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Derivation of Participants Included in the Initial EU-GEI Mass Spectrometry Experiment and Their Provision of Plasma Samples
Derivation of Participants Included in the Initial EU-GEI Mass Spectrometry Experiment and Their Provision of Plasma Samples

CHR indicates clinical high risk; CHR-NT, participants at clinical high risk who did not transition to psychosis; CHR-T, participants at clinical high risk who transitioned to first-episode psychosis; and EU-GEI, European Network of National Schizophrenia Networks Studying Gene-Environment Interactions.

Figure 2.  Model 1a Predicting Transition to Psychosis Using Clinical and Proteomic Data
Model 1a Predicting Transition to Psychosis Using Clinical and Proteomic Data

A, The algorithm score is a decision score used to determine the predicted outcome class. Herein, a score greater than 0 is assigned as CHR-T, and a score less than 0 is assigned as CHR-NT. The dashed lines divide the graph into quadrants according to predicted vs actual outcome (ie, top right is true positive, bottom left is true negative, top left is false positive, and bottom right is false negative). B, The dashed line is the line of no discrimination (area under the receiver operating characteristic curve, 0.5). CHR-NT indicates participants at clinical high risk who did not transition to psychosis; CHR-T, participants at clinical high risk who transitioned to first-episode psychosis; and EU-GEI, European Network of National Schizophrenia Networks Studying Gene-Environment Interactions.

Table 1.  Sample Characteristics for CHR-T and CHR-NT Groups in the Initial Experiment
Sample Characteristics for CHR-T and CHR-NT Groups in the Initial Experiment
Table 2.  Performance Metrics for Unadjusted Support Vector Machine Models
Performance Metrics for Unadjusted Support Vector Machine Models
Table 3.  Ten Percent Highest-Weighted Features for Model 1a, Model 3, and Model 4a
Ten Percent Highest-Weighted Features for Model 1a, Model 3, and Model 4a
Supplement 1.

eMethods.

eReferences.

eTable 1. Summary of Support Vector Machine Models

eTable 2. List of 69 Baseline EU-GEI Clinical Variables Included in Model 1a, Model 1b, and Model 3

eTable 3. Comparison of Baseline Characteristics for CHR Participants Who Attended at Least One Follow-up Interview vs CHR Participants Who Did Not in EU-GEI

eTable 4. Comparison of Characteristics for Participants Included in Initial Experiment (N = 133) From Total EU-GEI Clinical High-Risk Cohort (N = 344)

eTable 5. Results of ANCOVA (Adjusted for Age, Sex, BMI, and Years in Education) and Fold Changes (CHR-T vs CHR-NT) for Proteins Identified in EU-GEI Baseline Plasma Samples in the Initial Experiment

eTable 6. Coefficients of Variation for Proteins Across Quality Control Standards in EU-GEI Initial Experiment

eTable 7. Summary of Protein-Protein Interactions Identified From the BIOGRID Database for Significantly Differentially Expressed Proteins Between CHR-T and CHR-NT in EU-GEI Initial Experiment

eTable 8. Functional Enrichment Analysis of Differentially Expressed Proteins (Following False Discovery Rate Correction) Between CHR-T and CHR-NT in EU-GEI Initial Experiment: 6 KEGG Pathways Significantly Enriched

eTable 9. Results of Enzyme-Linked Immunosorbent Assays in CHR-T and CHR-NT Participants in EU-GEI Initial Experiment

eTable 10. Correlations Between 5 Proteins Assessed by ELISA and by Mass Spectrometry in EU-GEI Initial Experiment

eTable 11. Comparison of Characteristics for Participants Included in Replication Experiment (N = 135) From Total EU-GEI Clinical High-Risk Cohort (N = 344)

eTable 12. Sample Characteristics for CHR-T and CHR-NT Groups in the Replication Experiment

eTable 13. Results of ANCOVA (Adjusted for Age, Sex, BMI, Years in Education, Tobacco Use, and Ethnicity) and Fold Changes (CHR-T vs CHR-NT) for Proteins Identified in EU-GEI Baseline Plasma Samples in the Replication Experiment

eTable 14. Sample Characteristics for ALSPAC Subsample Cases and Controls

eTable 15. Results of ANCOVA (Adjusted for Sex, BMI, and Maternal Social Class) and Fold Changes (Definite PEs at 18 vs no PEs at 18) for Proteins Identified in ALSPAC Age 12 Plasma Samples

eTable 16. Ten Percent Highest-Weighted Features for Model S2 (Support Vector Machine Model Predicting Functional Outcome at 24 Months in EU-GEI)

eTable 17. Table Comparing Performance Metrics for Multiclass Site Prediction Models Based on 69 Clinical Features From Model 1b

eTable 18. Table Comparing Performance Metrics for Multiclass Site Prediction Models Based on 166 Proteomic Features From Model 1c

eTable 19. Comparison of Performance Metrics for Uncorrected vs Corrected Support Vector Machine Models

eTable 20. Proteins Differentially Expressed in CHR-T vs CHR-NT on ANCOVA (P < .05) in EU-GEI Baseline Plasma Samples in the Initial and Replication Experiment and Predicted Systemic Impact on Coagulation and Complement Activation and Regulation

eTable 21. Table Summarizing Directionality of Effect of Ten Percent Highest-Weighted Features in Model 1a (EU-GEI Initial Experiment), Model 3 (EU-GEI Replication Experiment), and Model 4 (ALSPAC Proteomic Data)

eFigure 1. Derivation and Testing of Model 2b: Parsimonious (10-Predictor) Proteomic Model

eFigure 2. STRING Functional Protein Association Network for Proteins Differentially Expressed (Following False Discovery Rate Correction) Between CHR-T and CHR-NT in EU-GEI Initial Experiment

eFigure 3. Mean Algorithm Scores and Class Predictions (A) and Receiver Operating Characteristic Curve (B) for Model 1b: Clinical Data

eFigure 4. Mean Algorithm Scores and Class Predictions (A) and Receiver Operating Characteristic Curve (B) for Model 1c: Proteomic Data

eFigure 5. Mean Algorithm Scores and Class Predictions (A) and Receiver Operating Characteristic Curve (B) for Model 2a: Proteomic (Non-London)

eFigure 6. Mean Algorithm Scores and Class Predictions (A) and Receiver Operating Characteristic Curve (B) for Model 2b: Parsimonious (10-Predictor) Proteomic Model, Training Data (Non-London)

eFigure 7. Mean Algorithm Scores and Class Predictions (A) and Receiver Operating Characteristic Curve (B) for Model 3: Replication

eFigure 8. Mean Algorithm Scores and Class Predictions (A) and Receiver Operating Characteristic Curve (B) for Model 4: ALSPAC Proteomic Data

eFigure 9. Mean Algorithm Scores and Class Predictions (A) and Receiver Operating Characteristic Curve (B) for Model S1: ELISA

eFigure 10. Mean Algorithm Scores and Class Predictions (A) and Receiver Operating Characteristic Curve (B) for Model S2: Functional Outcome

eFigure 11. Model 1b (Clinical Data) Decision Scores Stratified by EU-GEI Site

eFigure 12. Model 1c (Proteomic Data) Decision Scores Stratified by EU-GEI Site

eFigure 13. Age Stratified by EU-GEI Site

eFigure 14. Years in Education Stratified by EU-GEI Site

eFigure 15. Body Mass Index Stratified by EU-GEI Site

eFigure 16. Sex Stratified by EU-GEI Site

eFigure 17. General Assessment of Functioning Symptoms Subscale Stratified by EU-GEI Site

eFigure 18. General Assessment of Functioning Disability Subscale Stratified by EU-GEI Site

eFigure 19. Scale for Assessment of Negative Symptoms (Composite Score) Stratified by EU-GEI Site

eFigure 20. Scale for Assessment of Negative Symptoms (Global Score) Stratified by EU-GEI Site

eFigure 21. Brief Psychiatric Rating Scale Score Stratified by EU-GEI Site

eFigure 22. Montgomery-Asberg Depression Rating Scale Score Stratified by EU-GEI Site

eFigure 23. Multiclass Receiver Operating Curves for Site Prediction Based on 69 Clinical Features From Model 1b

eFigure 24. Multiclass Receiver Operating Curves for Site Prediction Based on 166 Proteomic Features From Model 1c

eFigure 25. Illustration of the Complement and Coagulation Pathways Depicting the Impact Model of Differentially Expressed Complement and Coagulation Proteins in CHR-T vs CHR-NT

1.
Larsen  TK, Melle  I, Auestad  B,  et al.  Early detection of psychosis: positive effects on 5-year outcome.   Psychol Med. 2011;41(7):1461-1469. doi:10.1017/S0033291710002023 PubMedGoogle ScholarCrossref
2.
Fusar-Poli  P, Borgwardt  S, Bechdolf  A,  et al.  The psychosis high-risk state: a comprehensive state-of-the-art review.   JAMA Psychiatry. 2013;70(1):107-120. doi:10.1001/jamapsychiatry.2013.269 PubMedGoogle ScholarCrossref
3.
Fusar-Poli  P, Bonoldi  I, Yung  AR,  et al.  Predicting psychosis: meta-analysis of transition outcomes in individuals at high clinical risk.   Arch Gen Psychiatry. 2012;69(3):220-229. doi:10.1001/archgenpsychiatry.2011.1472 PubMedGoogle ScholarCrossref
4.
Zammit  S, Kounali  D, Cannon  M,  et al.  Psychotic experiences and psychotic disorders at age 18 in relation to psychotic experiences at age 12 in a longitudinal population-based cohort study.   Am J Psychiatry. 2013;170(7):742-750. doi:10.1176/appi.ajp.2013.12060768 PubMedGoogle ScholarCrossref
5.
Healy  C, Brannigan  R, Dooley  N,  et al.  Childhood and adolescent psychotic experiences and risk of mental disorder: a systematic review and meta-analysis.   Psychol Med. 2019;49(10):1589-1599. doi:10.1017/S0033291719000485 PubMedGoogle ScholarCrossref
6.
Healy  C, Campbell  D, Coughlan  H, Clarke  M, Kelleher  I, Cannon  M.  Childhood psychotic experiences are associated with poorer global functioning throughout adolescence and into early adulthood.   Acta Psychiatr Scand. 2018;138(1):26-34. doi:10.1111/acps.12907 PubMedGoogle ScholarCrossref
7.
McGorry  P, Keshavan  M, Goldstone  S,  et al.  Biomarkers and clinical staging in psychiatry.   World Psychiatry. 2014;13(3):211-223. doi:10.1002/wps.20144 PubMedGoogle ScholarCrossref
8.
UK Public General Acts. Human Tissue Act 2004. Accessed July 7, 2020. http://www.legislation.gov.uk/ukpga/2004/30/contents
9.
van Os  J, Rutten  BP, Myin-Germeys  I,  et al; European Network of National Networks studying Gene-Environment Interactions in Schizophrenia (EU-GEI).  Identifying gene-environment interactions in schizophrenia: contemporary challenges for integrated, large-scale investigations.   Schizophr Bull. 2014;40(4):729-736. doi:10.1093/schbul/sbu069 PubMedGoogle ScholarCrossref
10.
Kraan  TC, Velthorst  E, Themmen  M,  et al; EU-GEI High Risk Study.  Child maltreatment and clinical outcome in individuals at ultra-high risk for psychosis in the EU-GEI High Risk Study.   Schizophr Bull. 2018;44(3):584-592. doi:10.1093/schbul/sbw162 PubMedGoogle ScholarCrossref
11.
Yung  AR, Yuen  HP, McGorry  PD,  et al.  Mapping the onset of psychosis: the Comprehensive Assessment of At-Risk Mental States.   Aust N Z J Psychiatry. 2005;39(11-12):964-971. doi:10.1080/j.1440-1614.2005.01714.x PubMedGoogle ScholarCrossref
12.
English  JA, Lopez  LM, O’Gorman  A,  et al.  Blood-based protein changes in childhood are associated with increased risk for later psychotic disorder: evidence from a nested case-control study of the ALSPAC longitudinal birth cohort.   Schizophr Bull. 2018;44(2):297-306. doi:10.1093/schbul/sbx075 PubMedGoogle ScholarCrossref
13.
Aas  IH.  Global Assessment of Functioning (GAF): properties and frontier of current knowledge.   Ann Gen Psychiatry. 2010;9:20. doi:10.1186/1744-859X-9-20 PubMedGoogle ScholarCrossref
14.
Goldman  HH, Skodol  AE, Lave  TR.  Revising axis V for DSM-IV: a review of measures of social functioning.   Am J Psychiatry. 1992;149(9):1148-1156. doi:10.1176/ajp.149.9.1148 PubMedGoogle ScholarCrossref
15.
Andreasen  NC.  The Scale for the Assessment of Negative Symptoms (SANS): conceptual and theoretical foundations.   Br J Psychiatry Suppl. 1989;(7):49-58. doi:10.1192/S0007125000291496 PubMedGoogle Scholar
16.
Overall  JE, Gorham  DR.  The Brief Psychiatric Rating Scale.   Psychological Reports. 1962;10(3):799-812. doi:10.2466/pr0.1962.10.3.799Google ScholarCrossref
17.
Montgomery  SA, Asberg  M.  A new depression scale designed to be sensitive to change.   Br J Psychiatry. 1979;134:382-389. doi:10.1192/bjp.134.4.382 PubMedGoogle ScholarCrossref
18.
English  JA, Fan  Y, Föcking  M,  et al.  Reduced protein synthesis in schizophrenia patient–derived olfactory cells.   Transl Psychiatry. 2015;5:e663. doi:10.1038/tp.2015.119 PubMedGoogle Scholar
19.
Föcking  M, Opstelten  R, Prickaerts  J,  et al.  Proteomic investigation of the hippocampus in prenatally stressed mice implicates changes in membrane trafficking, cytoskeletal, and metabolic function.   Dev Neurosci. 2014;36(5):432-442. doi:10.1159/000365327 PubMedGoogle ScholarCrossref
20.
Föcking  M, Sabherwal  S, Cates  HM,  et al.  Complement pathway changes at age 12 are associated with psychotic experiences at age 18 in a longitudinal population-based study: evidence for a role of stress.   Mol Psychiatry. Published online January 11, 2019. doi:10.1038/s41380-018-0306-z PubMedGoogle Scholar
21.
Boyd  A, Golding  J, Macleod  J,  et al.  Cohort profile: the “children of the 90s”: the index offspring of the Avon Longitudinal Study of Parents and Children.   Int J Epidemiol. 2013;42(1):111-127. doi:10.1093/ije/dys064 PubMedGoogle ScholarCrossref
22.
Fraser  A, Macdonald-Wallis  C, Tilling  K,  et al.  Cohort profile: the Avon Longitudinal Study of Parents and Children: ALSPAC mothers cohort.   Int J Epidemiol. 2013;42(1):97-110. doi:10.1093/ije/dys066 PubMedGoogle ScholarCrossref
23.
University of Bristol. Avon Longitudinal Study of Parents and Children. Accessed July 2020. http://www.bristol.ac.uk/alspac/researchers/our-data/
24.
Benjamini  Y, Hochberg  Y.  Controlling the false discovery rate: a practical and powerful approach to multiple testing.   J R Stat Soc B. 1995;57(1):289-300. doi:10.1111/j.2517-6161.1995.tb02031.x Google Scholar
25.
Cox  J, Mann  M.  Quantitative, high-resolution proteomics for data-driven systems biology.   Annu Rev Biochem. 2011;80:273-299. doi:10.1146/annurev-biochem-061308-093216 PubMedGoogle ScholarCrossref
26.
Cox  J, Mann  M.  MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification.   Nat Biotechnol. 2008;26(12):1367-1372. doi:10.1038/nbt.1511 PubMedGoogle ScholarCrossref
27.
Lazar  C. Package “imputeLCMD.” Version 2.0. Published February 20, 2015. Accessed April 2019. https://cran.r-project.org/web/packages/imputeLCMD/imputeLCMD.pdf
28.
RStudio. PBC. RStudio. Accessed April 2019. https://www.rstudio.com/
29.
Cearns  M, Hahn  T, Baune  BT.  Recommendations and future directions for supervised machine learning in psychiatry.   Transl Psychiatry. 2019;9(1):271. doi:10.1038/s41398-019-0607-2 PubMedGoogle ScholarCrossref
30.
Szklarczyk  D, Franceschini  A, Wyder  S,  et al.  STRING v10: protein-protein interaction networks, integrated over the tree of life.   Nucleic Acids Res. 2015;43(database issue):D447-D452. doi:10.1093/nar/gku1003 PubMedGoogle ScholarCrossref
31.
Radua  J, Ramella-Cravaro  V, Ioannidis  JPA,  et al.  What causes psychosis? an umbrella review of risk and protective factors.   World Psychiatry. 2018;17(1):49-66. doi:10.1002/wps.20490 PubMedGoogle ScholarCrossref
32.
Studerus  E, Ramyead  A, Riecher-Rössler  A.  Prediction of transition to psychosis in patients with a clinical high risk for psychosis: a systematic review of methodology and reporting.   Psychol Med. 2017;47(7):1163-1178. doi:10.1017/S0033291716003494 PubMedGoogle ScholarCrossref
33.
Malda  A, Boonstra  N, Barf  H,  et al.  Individualized prediction of transition to psychosis in 1,676 individuals at clinical high risk: development and validation of a multivariable prediction model based on individual patient data meta-analysis.   Front Psychiatry. 2019;10:345. doi:10.3389/fpsyt.2019.00345 PubMedGoogle ScholarCrossref
34.
Mechelli  A, Lin  A, Wood  S,  et al.  Using clinical information to make individualized prognostic predictions in people at ultra high risk for psychosis.   Schizophr Res. 2017;184:32-38. doi:10.1016/j.schres.2016.11.047 PubMedGoogle ScholarCrossref
35.
Schmidt  A, Cappucciati  M, Radua  J,  et al.  Improving prognostic accuracy in subjects at clinical high risk for psychosis: systematic review of predictive models and meta-analytical sequential testing simulation.   Schizophr Bull. 2017;43(2):375-388.PubMedGoogle Scholar
36.
Ruhrmann  S, Schultze-Lutter  F, Salokangas  RK,  et al.  Prediction of psychosis in adolescents and young adults at high risk: results from the prospective European Prediction of Psychosis Study.   Arch Gen Psychiatry. 2010;67(3):241-251. doi:10.1001/archgenpsychiatry.2009.206 PubMedGoogle ScholarCrossref
37.
Cannon  TD, Yu  C, Addington  J,  et al.  An individualized risk calculator for research in prodromal psychosis.   Am J Psychiatry. 2016;173(10):980-988. doi:10.1176/appi.ajp.2016.15070890 PubMedGoogle ScholarCrossref
38.
Koutsouleris  N, Meisenzahl  EM, Davatzikos  C,  et al.  Use of neuroanatomical pattern classification to identify subjects in at-risk mental states of psychosis and predict disease transition.   Arch Gen Psychiatry. 2009;66(7):700-712. doi:10.1001/archgenpsychiatry.2009.62 PubMedGoogle ScholarCrossref
39.
Koutsouleris  N, Borgwardt  S, Meisenzahl  EM, Bottlender  R, Möller  HJ, Riecher-Rössler  A.  Disease prediction in the at-risk mental state for psychosis using neuroanatomical biomarkers: results from the FePsy study.   Schizophr Bull. 2012;38(6):1234-1246. doi:10.1093/schbul/sbr145 PubMedGoogle ScholarCrossref
40.
Koutsouleris  N, Kambeitz-Ilankovic  L, Ruhrmann  S,  et al; PRONIA Consortium.  Prediction models of functional outcomes for individuals in the clinical high-risk state for psychosis or with recent-onset depression: a multimodal, multisite machine learning analysis.   JAMA Psychiatry. 2018;75(11):1156-1172. doi:10.1001/jamapsychiatry.2018.2165 PubMedGoogle ScholarCrossref
41.
Das  T, Borgwardt  S, Hauke  DJ,  et al.  Disorganized gyrification network properties during the transition to psychosis.   JAMA Psychiatry. 2018;75(6):613-622. doi:10.1001/jamapsychiatry.2018.0391 PubMedGoogle ScholarCrossref
42.
Koutsouleris  N, Davatzikos  C, Bottlender  R,  et al.  Early recognition and disease prediction in the at-risk mental states for psychosis using neurocognitive pattern classification.   Schizophr Bull. 2012;38(6):1200-1215. doi:10.1093/schbul/sbr037 PubMedGoogle ScholarCrossref
43.
Perkins  DO, Jeffries  CD, Addington  J,  et al.  Towards a psychosis risk blood diagnostic for persons experiencing high-risk symptoms: preliminary results from the NAPLS project.   Schizophr Bull. 2015;41(2):419-428. doi:10.1093/schbul/sbu099 PubMedGoogle ScholarCrossref
44.
Chan  MK, Krebs  MO, Cox  D,  et al.  Development of a blood-based molecular biomarker test for identification of schizophrenia before disease onset.   Transl Psychiatry. 2015;5:e601. doi:10.1038/tp.2015.91 PubMedGoogle Scholar
45.
Ruhrmann  S, Schultze-Lutter  F, Schmidt  SJ, Kaiser  N, Klosterkötter  J.  Prediction and prevention of psychosis: current progress and future tasks.   Eur Arch Psychiatry Clin Neurosci. 2014;264(suppl 1):S9-S16. doi:10.1007/s00406-014-0541-5 PubMedGoogle ScholarCrossref
46.
Woo  JJ, Pouget  JG, Zai  CC, Kennedy  JL.  The complement system in schizophrenia: where are we now and what’s next?   Mol Psychiatry. 2020;25(1):114-130. doi:10.1038/s41380-019-0479-0PubMedGoogle ScholarCrossref
47.
Sabherwal  S, English  JA, Föcking  M, Cagney  G, Cotter  DR.  Blood biomarker discovery in drug-free schizophrenia: the contribution of proteomics and multiplex immunoassays.   Expert Rev Proteomics. 2016;13(12):1141-1155. doi:10.1080/14789450.2016.1252262 PubMedGoogle ScholarCrossref
48.
Li  Y, Zhou  K, Zhang  Z,  et al.  Label-free quantitative proteomic analysis reveals dysfunction of complement pathway in peripheral blood of schizophrenia patients: evidence for the immune hypothesis of schizophrenia.   Mol Biosyst. 2012;8(10):2664-2671. doi:10.1039/c2mb25158b PubMedGoogle ScholarCrossref
49.
Kopczynska  M, Zelek  W, Touchard  S,  et al.  Complement system biomarkers in first episode psychosis.   Schizophr Res. 2019;204:16-22. doi:10.1016/j.schres.2017.12.012 PubMedGoogle Scholar
50.
Levin  Y, Wang  L, Schwarz  E, Koethe  D, Leweke  FM, Bahn  S.  Global proteomic profiling reveals altered proteomic signature in schizophrenia serum.   Mol Psychiatry. 2010;15(11):1088-1100. doi:10.1038/mp.2009.54 PubMedGoogle ScholarCrossref
51.
Föcking  M, Dicker  P, Lopez  LM,  et al.  Differential expression of the inflammation marker IL12p40 in the at-risk mental state for psychosis: a predictor of transition to psychotic disorder?   BMC Psychiatry. 2016;16(1):326. doi:10.1186/s12888-016-1039-7 PubMedGoogle ScholarCrossref
52.
Miller  BJ, Buckley  P, Seabolt  W, Mellor  A, Kirkpatrick  B.  Meta-analysis of cytokine alterations in schizophrenia: clinical status and antipsychotic effects.   Biol Psychiatry. 2011;70(7):663-671. doi:10.1016/j.biopsych.2011.04.013 PubMedGoogle ScholarCrossref
53.
Khandaker  GM, Pearson  RM, Zammit  S, Lewis  G, Jones  PB.  Association of serum interleukin 6 and C-reactive protein in childhood with depression and psychosis in young adult life: a population-based longitudinal study.   JAMA Psychiatry. 2014;71(10):1121-1128. doi:10.1001/jamapsychiatry.2014.1332 PubMedGoogle ScholarCrossref
54.
Laskaris  L, Zalesky  A, Weickert  CS,  et al.  Investigation of peripheral complement factors across stages of psychosis.   Schizophr Res. 2019;204:30-37. doi:10.1016/j.schres.2018.11.035 PubMedGoogle ScholarCrossref
55.
Baumeister  D, Russell  A, Pariante  CM, Mondelli  V.  Inflammatory biomarker profiles of mental disorders and their relation to clinical, social and lifestyle factors.   Soc Psychiatry Psychiatr Epidemiol. 2014;49(6):841-849. doi:10.1007/s00127-014-0887-z PubMedGoogle ScholarCrossref
56.
Sekar  A, Bialas  AR, de Rivera  H,  et al; Schizophrenia Working Group of the Psychiatric Genomics Consortium.  Schizophrenia risk from complex variation of complement component 4.   Nature. 2016;530(7589):177-183. doi:10.1038/nature16549 PubMedGoogle ScholarCrossref
57.
Hoirisch-Clapauch  S, Amaral  OB, Mezzasalma  MA, Panizzutti  R, Nardi  AE.  Dysfunction in the coagulation system and schizophrenia.   Transl Psychiatry. 2016;6:e704. doi:10.1038/tp.2015.204 PubMedGoogle Scholar
58.
Rehman  AA, Ahsan  H, Khan  FH.  α-2-Macroglobulin: a physiological guardian.   J Cell Physiol. 2013;228(8):1665-1675. doi:10.1002/jcp.24266 PubMedGoogle ScholarCrossref
59.
Upthegrove  R, Manzanares-Teson  N, Barnes  NM.  Cytokine function in medication-naive first episode psychosis: a systematic review and meta-analysis.   Schizophr Res. 2014;155(1-3):101-108. doi:10.1016/j.schres.2014.03.005 PubMedGoogle ScholarCrossref
60.
Borth  W.  Alpha 2-macroglobulin, a multifunctional binding protein with targeting characteristics.   FASEB J. 1992;6(15):3345-3353. doi:10.1096/fasebj.6.15.1281457 PubMedGoogle ScholarCrossref
61.
Baker  SK, Chen  ZL, Norris  EH, Revenko  AS, MacLeod  AR, Strickland  S.  Blood-derived plasminogen drives brain inflammation and plaque deposition in a mouse model of Alzheimer’s disease.   Proc Natl Acad Sci U S A. 2018;115(41):E9687-E9696. doi:10.1073/pnas.1811172115 PubMedGoogle ScholarCrossref
62.
Amara  U, Flierl  MA, Rittirsch  D,  et al.  Molecular intercommunication between the complement and coagulation systems.   J Immunol. 2010;185(9):5628-5636. doi:10.4049/jimmunol.0903678 PubMedGoogle ScholarCrossref
63.
Ryu  JK, Petersen  MA, Murray  SG,  et al.  Blood coagulation protein fibrinogen promotes autoimmunity and demyelination via chemokine release and antigen presentation.   Nat Commun. 2015;6:8164. doi:10.1038/ncomms9164 PubMedGoogle ScholarCrossref
64.
Pollak  TA, Drndarski  S, Stone  JM, David  AS, McGuire  P, Abbott  NJ.  The blood-brain barrier in psychosis.   Lancet Psychiatry. 2018;5(1):79-92. doi:10.1016/S2215-0366(17)30293-6 PubMedGoogle ScholarCrossref
65.
Ryu  JK, Rafalski  VA, Meyer-Franke  A,  et al.  Fibrin-targeting immunotherapy protects against neuroinflammation and neurodegeneration.   Nat Immunol. 2018;19(11):1212-1223. doi:10.1038/s41590-018-0232-x PubMedGoogle ScholarCrossref
66.
Comes  AL, Papiol  S, Mueller  T, Geyer  PE, Mann  M, Schulze  TG.  Proteomics for blood biomarker exploration of severe mental illness: pitfalls of the past and potential for the future.   Transl Psychiatry. 2018;8(1):160. doi:10.1038/s41398-018-0219-2 PubMedGoogle ScholarCrossref
67.
Westwood  S, Leoni  E, Hye  A,  et al.  Blood-based biomarker candidates of cerebral amyloid using PiB PET in non-demented elderly.   J Alzheimers Dis. 2016;52(2):561-572. doi:10.3233/JAD-151155 PubMedGoogle ScholarCrossref
68.
Ripke  S, Neale  BM, Corvin  A,  et al; Schizophrenia Working Group of the Psychiatric Genomics Consortium.  Biological insights from 108 schizophrenia-associated genetic loci.   Nature. 2014;511(7510):421-427. doi:10.1038/nature13595 PubMedGoogle ScholarCrossref
69.
Fung  TC, Olson  CA, Hsiao  EY.  Interactions between the microbiota, immune and nervous systems in health and disease.   Nat Neurosci. 2017;20(2):145-155. doi:10.1038/nn.4476 PubMedGoogle ScholarCrossref
70.
Khandaker  GM, Zimbron  J, Lewis  G, Jones  PB.  Prenatal maternal infection, neurodevelopment and adult schizophrenia: a systematic review of population-based studies.   Psychol Med. 2013;43(2):239-257. doi:10.1017/S0033291712000736 PubMedGoogle ScholarCrossref
71.
Rasmussen  LJH, Moffitt  TE, Arseneault  L,  et al.  Association of adverse experiences and exposure to violence in childhood and adolescence with inflammatory burden in young people.   JAMA Pediatr. 2019;174(1):1-11. doi:10.1001/jamapediatrics.2019.3875PubMedGoogle Scholar
72.
van Os  J, Kenis  G, Rutten  BP.  The environment and schizophrenia.   Nature. 2010;468(7321):203-212. doi:10.1038/nature09563 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    August 26, 2020

    Development of Proteomic Prediction Models for Transition to Psychotic Disorder in the Clinical High-Risk State and Psychotic Experiences in Adolescence

    Author Affiliations
    • 1Department of Psychiatry, Royal College of Surgeons in Ireland, Dublin, Ireland
    • 2School of Pharmacy and Pharmaceutical Sciences, Cardiff University, Cardiff, United Kingdom
    • 3School of Biomolecular and Biomedical Science, Conway Institute, University College Dublin, Dublin, Ireland
    • 4Centre for Youth Mental Health, University of Melbourne, Parkville, Victoria, Australia
    • 5Mental Health Centre Copenhagen, Copenhagen University Hospital, Copenhagen, Denmark
    • 6University Paris Descartes, Groupe Hospitalier Universitaire (GHU) Paris–Sainte Anne, Evaluation Centre for Young Adults and Adolescents (C’JAAD), Service Hospitalov–Universitaire, Institut National de la Santé et de la Recherche Medicale (INSERM) U1266, Institut de Psychiatrie (Centre National de la Recherche Scientifique [CNRS] 3557), Paris, France
    • 7Department of Psychiatry, Medical Faculty, University of Basel, Basel, Switzerland
    • 8LiNC–Lab Interdisciplinar Neurociências Clínicas, Depto Psiquiatria, Escola Paulista de Medicina, Universidade Federal de São Paulo (UNIFESP), São Paulo, Brazil
    • 9Departament de Psicologia Clínica i de la Salut (Universitat Autònoma de Barcelona), Fundació Sanitària Sant Pere Claver (Spain), Spanish Mental Health Research Network (Centro de Investigación Biomédica en Red de Salud Mental [CIBERSAM]), Barcelona, Spain
    • 10Department of Psychiatry and Psychotherapy, Translational Psychiatry Unit, University zu Lübeck, Lübeck, Germany
    • 11Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
    • 12Department of Psychiatry and Psychotherapy, Medical University of Vienna, Vienna, Austria
    • 13Melbourne Neuropsychiatry Centre, Department of Psychiatry, University of Melbourne and Melbourne Health, Carlton South, Victoria, Australia
    • 14Faculty of Behavioural and Movement Sciences, Department of Clinical Psychology and EMGO+ Institute for Health and Care Research, Vrije Universiteit (VU) University, Amsterdam, the Netherlands
    • 15Department of Psychosis Research, Parnassia Psychiatric Institute, The Hague, the Netherlands
    • 16Academic Medical Centre (AMC), Academic Psychiatric Centre, Department Early Psychosis, Amsterdam, the Netherlands
    • 17Institute of Psychiatry, Psychology & Neuroscience, Department of Psychology, King’s College London, London, United Kingdom
    • 18Institute of Psychiatry, Psychology & Neuroscience, Department of Psychosis Studies, King’s College London, London, United Kingdom
    • 19Department of Psychiatry and Neuropsychology, School for Mental Health and Neuroscience, Maastricht University Medical Centre, Maastricht, the Netherlands
    • 20Trinity Institute of Neuroscience, Trinity College Dublin, Dublin, Ireland
    • 21Medical Research Council (MRC) Centre for Neuropsychiatric Genetics and Genomics, Cardiff University, Cardiff, United Kingdom
    • 22Bristol Medical School, University of Bristol, Bristol, United Kingdom
    JAMA Psychiatry. 2021;78(1):77-90. doi:10.1001/jamapsychiatry.2020.2459
    Key Points

    Question  Can plasma proteomic biomarkers aid prediction of transition to psychotic disorder in people at clinical high risk (CHR) of psychosis and adolescent psychotic experiences in the general population?

    Findings  In this diagnostic study of 133 individuals at CHR in EU-GEI and 121 individuals from the general population in ALSPAC, models were developed based on baseline proteomic data, with excellent predictive performance for transition to psychotic disorder in individuals at CHR. In a general population sample, models based on proteomic data at age 12 years had fair predictive performance for psychotic experiences at age 18 years.

    Meaning  Predictive models based on proteomic biomarkers may contribute to personalized prognosis and stratification strategies in individuals at risk of psychosis.

    Abstract

    Importance  Biomarkers that are predictive of outcomes in individuals at risk of psychosis would facilitate individualized prognosis and stratification strategies.

    Objective  To investigate whether proteomic biomarkers may aid prediction of transition to psychotic disorder in the clinical high-risk (CHR) state and adolescent psychotic experiences (PEs) in the general population.

    Design, Setting, and Participants  This diagnostic study comprised 2 case-control studies nested within the European Network of National Schizophrenia Networks Studying Gene-Environment Interactions (EU-GEI) and the Avon Longitudinal Study of Parents and Children (ALSPAC). EU-GEI is an international multisite prospective study of participants at CHR referred from local mental health services. ALSPAC is a United Kingdom–based general population birth cohort. Included were EU-GEI participants who met CHR criteria at baseline and ALSPAC participants who did not report PEs at age 12 years. Data were analyzed from September 2018 to April 2020.

    Main Outcomes and Measures  In EU-GEI, transition status was assessed by the Comprehensive Assessment of At-Risk Mental States or contact with clinical services. In ALSPAC, PEs at age 18 years were assessed using the Psychosis-Like Symptoms Interview. Proteomic data were obtained from mass spectrometry of baseline plasma samples in EU-GEI and plasma samples at age 12 years in ALSPAC. Support vector machine learning algorithms were used to develop predictive models.

    Results  The EU-GEI subsample (133 participants at CHR (mean [SD] age, 22.6 [4.5] years; 68 [51.1%] male) comprised 49 (36.8%) who developed psychosis and 84 (63.2%) who did not. A model based on baseline clinical and proteomic data demonstrated excellent performance for prediction of transition outcome (area under the receiver operating characteristic curve [AUC], 0.95; positive predictive value [PPV], 75.0%; and negative predictive value [NPV], 98.6%). Functional analysis of differentially expressed proteins implicated the complement and coagulation cascade. A model based on the 10 most predictive proteins accurately predicted transition status in training (AUC, 0.99; PPV, 76.9%; and NPV, 100%) and test (AUC, 0.92; PPV, 81.8%; and NPV, 96.8%) data. The ALSPAC subsample (121 participants from the general population with plasma samples available at age 12 years (61 [50.4%] male) comprised 55 participants (45.5%) with PEs at age 18 years and 61 (50.4%) without PEs at age 18 years. A model using proteomic data at age 12 years predicted PEs at age 18 years, with an AUC of 0.74 (PPV, 67.8%; and NPV, 75.8%).

    Conclusions and Relevance  In individuals at risk of psychosis, proteomic biomarkers may contribute to individualized prognosis and stratification strategies. These findings implicate early dysregulation of the complement and coagulation cascade in the development of psychosis outcomes.

    Introduction

    Early detection of psychosis may improve clinical outcomes.1 Clinical high-risk (CHR) criteria2 enable identification of vulnerable groups with 3-year transition rates to first-episode psychosis (FEP) of 16% to 35%.3 However, it is difficult to predict outcomes individually. Previous studies have also characterized an extended psychosis phenotype that includes individuals with psychotic experiences (PEs).4 These subthreshold symptoms are associated with an increased risk of psychotic and nonpsychotic disorders5 and reduced global functioning.6

    Biomarkers may augment prognosis and stratification strategies.7 We aimed to compare plasma protein expression in individuals at CHR who do and do not develop psychosis and to develop models incorporating proteomic data for individualized prediction of transition to FEP. This study also aimed to apply similar methods for prediction of PEs in a general population sample.

    Methods

    Ethical approval for this diagnostic study was granted by the Royal College of Surgeons in Ireland. Ethics committees of participating sites granted approval for the European Network of National Schizophrenia Networks Studying Gene-Environment Interactions (EU-GEI). Approval was also obtained from the Avon Longitudinal Study of Parents and Children (ALSPAC) Ethics and Law Committee and local research ethics committees. Informed consent for collection of biological samples was obtained in accordance with the Human Tissue Act 2004.8 Informed consent for use of questionnaire and clinic data was obtained following recommendations of the ALSPAC Ethics and Law Committee at the time.

    Study 1: CHR Sample
    Participants and Study Design

    EU-GEI study includes a prospective cohort of 344 participants at CHR recruited across 11 international sites.9,10 Individuals with CHR symptoms who were referred by local mental health services were eligible to participate if they met CHR criteria according to the Comprehensive Assessment of At-Risk Mental States11 (CAARMS) and provided written informed consent. Exclusion criteria were current or past psychotic disorder, symptoms explained by a medical disorder or drug or alcohol use, and IQ less than 60.

    Plasma samples were obtained at baseline, and clinical assessments were performed at baseline, 12 months, and 24 months. After 24 months, or if a face-to-face interview was not possible, attempts were made to confirm transition status via the clinical team or records. Assessors were not systematically blinded to transition status because, in some cases, clinical services contacted the research team in advance to advise that transition had occurred. Accrual began in September 2010. The last baseline assessment was performed in July 2015.

    The present investigation comprised a nested case-control study comparing plasma proteins from participants at CHR who transitioned to psychosis on follow-up (CHR-T) (n = 49) with a control group of randomly selected participants who did not (CHR-NT) (n = 84) (Figure 1). Based on previous experience,12 the experiment was limited to this number to ensure optimal technical performance across mass spectrometry runs.

    Outcome and Clinical Measures

    Transition was defined as the onset of nonorganic psychotic disorder as assessed either by CAARMS interview11 or by contact with the clinical team or review of clinical records. Sixty-five of 344 participants at CHR (18.9%) developed psychosis on follow-up, 57 within 24 months and 8 after 24 months.

    Baseline clinical measures were recorded. These included age, sex, body mass index (BMI; calculated as weight in kilograms divided by height in meters squared), years of education, General Assessment of Functioning (GAF) subscales for symptoms and disability,13,14 the Scale for the Assessment of Negative Symptoms (SANS),15 the Brief Psychiatric Rating Scale (BPRS),16 and the Montgomery-Åsberg Depression Rating Scale (MADRS).17

    Sample Preparation, Proteomics, Validation, and Replication

    Laboratory procedures were conducted blind to case-control status. Protein depletion, digestion, and peptide purification were performed using baseline plasma samples. Discovery-based proteomic methods were used.12 Briefly, 5 μL from each prepared sample was injected on a Q Exactive (Thermo Scientific) mass spectrometer operated in data-dependent acquisition mode for label-free liquid chromatography mass spectrometry12,18-20 (eMethods in Supplement 1 and eAppendix in Supplement 2).

    Nine proteins in plasma samples from the same participants at CHR described above (Figure 1) were assessed using enzyme-linked immunosorbent assay (ELISA). Details are available in eMethods in Supplement 1.

    In an effort to reproduce our findings, we conducted a partial replication of the initial mass spectrometry experiment by analyzing baseline plasma samples from 49 CHR-T cases (2 of these cases were different from the initial experiment) and an entirely new group of 86 CHR-NT control cases. Details are available in eMethods in Supplement 1.

    Study 2: General Population Sample
    Participants and Study Design

    The ALSPAC is a prospective birth cohort.21-23 Pregnant women in Avon, United Kingdom, with delivery dates between April 1, 1991, and December 31, 1992, were invited to participate, and 14 541 pregnancies were enrolled. When the oldest children were approximately age 7 years, an attempt was made to bolster the sample with children who did not join originally. The sample size for analyses using data from age 7 years is 15 454 pregnancies (14 901 children alive at 1 year).

    Plasma samples obtained at age 12 years from ALSPAC participants who did or did not report PEs at age 18 years were previously investigated.12,20 In data-independent acquisition analyses focused on proteins of the complement pathway, several proteins were differentially expressed. Herein, we performed data-dependent acquisition analyses (rather than data-independent acquisition) in this sample to achieve broader proteome coverage.

    Outcome, Sample Preparation, and Proteomics

    Psychotic experiences were assessed in participants at age 12 years and age 18 years using the Psychosis-Like Symptoms Interview4 and were rated as not present, suspected, or definite. Of 4060 participants assessed at both time points, 190 (4.7%) had suspected or definite PEs at age 18 years but not at age 12 years.4 The present study was based on a subsample of case participants (who did not report PEs at age 12 years but reported at least 1 definite PE at age 18 years) and randomly selected control participants (who did not report PEs at either age 12 years or age 18 years).

    Plasma samples at age 12 years were prepared as previously described.12 Data-dependent acquisition proteomic analyses were performed as for study 1.

    Data and Statistical Analyses

    Data were analyzed from September 2018 to April 2020. Clinical data were tested for differences using the 2-sided t test for continuous variables and χ2 test for categorical variables in SPSS, version 25 (IBM). P values were corrected for multiple comparisons using the Benjamini-Hochberg procedure24 with a 5% false discovery rate (FDR). The threshold for statistical significance was FDR-corrected P < .05.

    Label-free quantification was performed in MaxQuant, version 1.5.2.8 (Max Planck Institute of Biochemistry).25,26 Proteins identified with at least 2 peptides (1 uniquely assignable to the protein) and quantified in more than 80% of samples were taken forward for analysis and log2 transformed. Missing values were imputed using imputeLCMD (version 2.0)27 in RStudio.28 Label-free quantification values were converted to z scores and winsorized within ±3 z.

    Analysis of covariance was performed in Stata, version 15 (StataCorp LLC), comparing the mean label-free quantification for each protein in cases and controls. Covariates included age, sex, BMI, and years of education in study 1 and sex, BMI at age 12 years, and maternal social class in study 2. P values were corrected for multiple comparisons with a 5% FDR.

    Predictive Models

    Neurominer, version 1.0, for MatLab 2018a (MathWorks Inc) was used to develop support vector machine (SVM) models (eMethods in Supplement 1). The development of each model is summarized in eTable 1 in Supplement 1.

    Models 1a-c: Predicting Transition Using Clinical and Proteomic Data

    First, we developed a model predicting transition using clinical and proteomic data together (model 1a). eTable 2 in Supplement 1 lists the included clinical features. Geographical generalizability was incorporated using leave-site-out cross-validation (eMethods in Supplement 1) as recommended for multisite consortia.29 To assess the relative contribution of clinical and proteomic data, we next developed models using the same cross-validation and training framework but based on clinical (model 1b) and proteomic (model 1c) features separately.

    Model 2a and b: Parsimonious Model

    We sought to generate a parsimonious model based on the 10 highest-weighted proteomic predictors and internally validate this model in unseen data (eFigure 1 in Supplement 1). As the largest site, London, United Kingdom, was chosen as the test site, and data for these participants were held out.

    To derive the 10 highest-weighted proteins, a model (model 2a) was generated using proteomic data from all sites except London (n = 30 for CHR-T and n = 50 for CHR-NT). A reduced model was then developed based solely on data for these 10 proteins in the non-London data set (model 2b) and then tested in the held-out London data (n = 19 for CHR-T and n = 34 for CHR-NT). Both models used leave-site-out cross-validation.

    Model 3: Replication

    Because of differences in protein identifications, it was not possible to apply models 1a-c and 2a-b to the replication data set. We instead sought to replicate our initial findings by performing a second discovery analysis, generating a new model (with leave-site-out cross-validation) predicting transition based on clinical and proteomic data in the replication data set.

    Model 4: Predicting PEs Using Proteomic Data

    We developed a model predicting PEs at age 18 years in the ALSPAC based on proteomic data at age 12 years. Repeated nested cross-validation with 5 inner folds and 5 outer folds was used.

    Supplementary Analyses

    Several supplementary analyses (eMethods in Supplement 1) were performed. These included the following: the development of a model predicting transition in EU-GEI based on ELISA data (model S1), the development of a model predicting functional outcome in EU-GEI (GAF disability subscale score ≤60 [poor functional outcome] vs >60 [good functional outcome] at 24 months) based on clinical and proteomic data (model S2), investigation of potential EU-GEI site associations for clinical and proteomic data, and the development of multivariate-corrected versions of SVM models whereby the variance associated with multiple covariates was extracted using principal components analysis.

    Results
    Study 1: CHR Sample

    Of 344 participants at CHR who were recruited, 152 (44.2%) attended face-to-face interviews at 12 months and 105 (30.5%) at 24 months. Baseline characteristics of participants who did or did not attend at least 1 follow-up interview are compared in eTable 3 in Supplement 1. After FDR correction, participants who attended interviews had a mean of 1 more year of education and a lower mean SANS total global score than those who did not attend interviews but were otherwise comparable.

    The subsample for the initial experiment comprised 133 (49 CHR-T and 84 CHR-NT) participants with baseline plasma samples available, of whom 49 (36.8%) developed psychosis (Figure 1). The mean (SD) age of the participants was 22.6 (4.5) years; 68 participants (51.1%) were male. After FDR correction, participants included in the subsample had a higher mean SANS total composite, SANS total global, and BPRS total scores than nonincluded participants but were otherwise comparable on baseline characteristics (eTable 4 in Supplement 1).

    Subsample characteristics are listed in Table 1. After FDR correction, there were no statistically significant group differences for CHR-T vs CHR-NT based on baseline characteristics. The median duration from baseline to transition was 219 days (interquartile range, 424 days). The CHR-T participants had lower mean functional outcome scores at 2 years compared with CHR-NT participants (mean GAF symptoms score at 2 years, 42.3 in CHR-T vs 62.2 in CHR-NT; FDR-corrected P < .007; mean GAF disability score at 2 years, 44.7 in CHR-T vs 64.5 in CHR-NT; FDR-corrected P < .007).

    Differential Expression

    Of 345 proteins identified, 166 were quantified in more than 80% of plasma samples. There was nominally statistically significant (P < .05) differential expression of 56 proteins in CHR-T vs CHR-NT, of which 35 remained statistically significant after FDR correction (eTables 5 and 6 in Supplement 1). eFigure 2 in Supplement 1 shows a functional association network30 for these proteins, and eTable 7 in Supplement 1 lists protein-protein interactions. On functional enrichment analysis, the topmost implicated pathway was the complement and coagulation cascade (eTable 8 in Supplement 1).

    Model 1a: Predicting Transition Using Clinical and Proteomic Data

    An SVM model predicted transition status based on clinical and proteomic features (model 1a), with excellent performance (area under the receiver operating characteristic curve [AUC], 0.95; [P < .001]; sensitivity, 98.0%; specificity, 81.0%; positive predictive value [PPV], 75.0%; and negative predictive value [NPV], 98.6%). Performance metrics are listed in Table 2. Figure 2A shows the mean algorithm scores and predicted outcomes stratified by site. The receiver operating characteristic curve is shown in Figure 2B. Table 3 lists the 10% highest-weighted features according to the mean feature weight. For example, the 5 highest-ranked predictive features were alpha-2-macroglobulin (A2M) (mean weight, −0.330), immunoglobulin heavy constant mu (IGHM) (mean weight, −0.256), C4b-binding protein alpha chain (C4BPA) (mean weight, −0.161), complement component 8 alpha chain (C8A) (mean weight, 0.158), and phospholipid transfer protein (PLTP) (mean weight, −0.146).

    Model 1b and 1c: Clinical and Proteomic Data

    The clinical model (model 1b) demonstrated poor predictive performance (AUC, 0.48; P = .63). These results are summarized in Table 2 and eFigure 3 in Supplement 1. For example, sensitivity was 46.9%, specificity was 53.6%, PPV was 37.1%, and NPV was 63.4%.

    The proteomic model (model 1c) demonstrated excellent predictive performance (AUC, 0.96; P < .001). These results are summarized in Table 2 and eFigure 4 in Supplement 1. For example, sensitivity was 100%, specificity was 84.5%, PPV was 79.0%, and NPV was 100%.

    Model 2a and b: Parsimonious Model

    The AUC for the model based on proteomic data from all sites except London (model 2a) was 0.94 (P < .001) (Table 2 and eFigure 5 in Supplement 1). The 10 highest-weighted features were alpha-2-macroglobulin (A2M), immunoglobulin heavy constant mu (IGHM), C4b-binding protein alpha chain (C4BPA), vitamin K–dependent protein S, fibulin 1, transthyretin, N-acetylmuramoyl-l-alanine amidase, vitamin D–binding protein, clusterin, and complement component 6 (C6).

    A reduced model based solely on these 10 most predictive proteins was developed using data from all sites except London (model 2b), with an AUC of 0.99 (P < .001), sensitivity of 100%, specificity of 82.0%, PPV of 76.9%, and NPV of 100%) (Table 2 and eFigure 6 in Supplement 1). This model predicted transition status in the held-out London data, with an AUC of 0.92, sensitivity of 94.7%, specificity of 88.2%, PPV of 81.8%, and NPV of 96.8% (Table 2).

    ELISA Validation

    After FDR correction, 2 proteins assessed by ELISA showed statistically significant mean differences between CHR-T and CHR-NT. These were A2M and complement component 1r (C1r) (eTables 9 and 10 in Supplement 1). The A2M mean in CHR-T was 1173.1 μg/mL vs 11 501.7 μg/mL in CHR-T (FDR-corrected P = .02), and the C1r mean in CHR-T was 65 008.9 μg/mL vs 52 803.9 μg/mL in CHR-NT (FDR-corrected P = .04).

    Model 3: Replication

    Replication subsample characteristics are listed in eTables 11 and 12 in Supplement 1. Of 485 proteins identified, 119 were quantified in more than 80% of plasma samples. There was nominally statistically significant (P < .05) differential expression of 82 proteins, of which 78 remained statistically significant after FDR correction (eTable 13 in Supplement 1).

    Model 3 demonstrated excellent performance for prediction of transition in the replication data set (AUC, 0.98 [P < .001]; sensitivity, 98.0%; specificity, 89.5%; PPV, 84.2%; and NPV, 98.7%) (Table 2 and eFigure 7 in Supplement 1). The highest-weighted 10% of features are listed in Table 3. For example, the 5 highest-ranked predictive features were A2M (mean weight, −0.286), carboxypeptidase N subunit 2 (mean weight, 0.210), IGHM (mean weight, −0.193), complement C1s subcomponent (mean weight, −0.181), and alpha-1-antichymotrypsin (mean weight, 0.168). Proteins among the highest-weighted 10% of features in both model 1a and model 3 (and weighted in similar directions) included A2M, IGHM, C4BPA, plasminogen, and C6.

    Study 2: General Population Sample

    The initial subsample was composed of plasma samples from 132 participants (65 case and 67 control samples). Eleven plasma samples were excluded because of poor protein identification profiles, resulting in 55 case and 66 control samples from 121 participants (61 [50.4%] male). Case samples were more likely to be from female participants. There was no evidence for differences in BMI, race/ethnicity, or maternal social class (eTable 14 in Supplement 1).

    Differential Expression

    Of 506 proteins identified, 265 were quantified in more than 80% of samples. There was nominally statistically significant (P < .05) differential expression of 40 proteins at age 12 years (eTable 15 in Supplement 1), of which the following 5 remained statistically significant after FDR correction: C4BPA (ratio of means in PE vs no PE, 0.77), serum paraoxonase/arylesterase 1 (ratio of means, 0.80), IGHM (ratio of means, 0.78), inhibin beta chain (ratio of means, 1.31), and clusterin (ratio of means, 0.92).

    Model 4: Predicting PEs Using Proteomic Data

    An SVM model using 265 proteomic features from plasma samples obtained at age 12 years predicted PEs at age 18 years, with an AUC of 0.74 (P < .001), sensitivity of 72.7%, specificity of 71.2%, PPV of 67.8%, and NPV of 75.8% (Table 2 and eFigure 8 in Supplement 1). For example, the 5 highest-ranked predictive features were C4BPA (mean weight, −0.227), serum paraoxonase/arylesterase 1 (mean weight, −0.180), complement factor H–related protein 1 (mean weight, −0.152), vitamin K–dependent protein S (mean weight, −0.145), and lysozyme C (mean weight, −0.142) (Table 3).

    Supplementary Analyses

    Model S1 used ELISA data to predict transition status in EU-GEI, with an AUC of 0.76 (P < .001). These results are summarized in Table 2 and eFigure 9 in Supplement 1.

    Model S2 used clinical and proteomic data to predict poor (GAF disability subscale score ≤60) vs good (>60) functional outcome at 2 years in EU-GEI, with an AUC of 0.74 (P = .003) (Table 2 and eFigure 10 in Supplement 1). The 10% highest-weighted features are listed in eTable 16 in Supplement 1.

    There was evidence of differences for the clinical data between the London and the Netherlands sites compared with others (eTable 17, eFigure 11, and eFigure 23 in Supplement 1), likely because of group differences in age, years in education, and BPRS score (eMethods and eFigures 13-22 in Supplement 1). There was no strong evidence of systematic site associations for the proteomic data (eTable 18, eFigure 12, and eFigure 24 in Supplement 1).

    Performance metrics of multivariate-corrected SVM models are listed in eTable 19 in Supplement 1. There were generally slight reductions in AUCs of the corrected models compared with their uncorrected counterparts (median change in AUC, 0.04; range, 0.01-0.10), although in all cases the 95% CIs overlapped.

    Discussion

    We described evidence of differential baseline plasma protein expression in individuals at CHR who developed psychosis compared with those who did not. Machine learning algorithms that incorporated clinical and proteomic data were used to predict transition outcome (AUC, 0.95). Proteomic features were of greater predictive value than clinical features. A parsimonious model based on 10 highly predictive proteins showed excellent performance in training (AUC, 0.99) and test (AUC, 0.92) data. Furthermore, a predictive model was developed using proteomic data at age 12 years for PEs at age 18 years in a general population sample (AUC, 0.74).

    Although only 16% to 35% of individuals at CHR transition to FEP,3 the CHR state remains a strong risk factor.31 Clinical data have previously shown value for prediction of transition,32-37 and the poor performance of the clinical features in our study does not imply that clinical data in general are of little prognostic use. Previous studies have attempted to augment accuracy using neuroimaging38-41 and neurocognitive42 data, but blood-based tests have the advantage of greater accessibility. Perkins et al43 derived a panel of 15 proteins using immunoassays that distinguished between CHR-T and CHR-NT, with an AUC of 0.88. Chan et al44 used 22 blood-based biomarkers to predict schizophrenia onset, with an AUC of 0.82 that increased to 0.90 with incorporation of the CAARMS positive symptoms subscale. Our parsimonious model used data for 10 proteins, and, with further validation, may contribute to individualized prognosis and treatment stratification strategies.45

    eTable 20 in Supplement 1 summarizes our findings of differential expression in CHR-T vs CHR-NT and the predicted functional implications (modeled in eFigure 25 in Supplement 1). We found particularly strong evidence for dysregulation of the complement and coagulation cascade, previously implicated in schizophrenia.46-50 Similar processes have been previously implicated in proteomic studies of the development of PEs in the general population.12,20 Changes in the present CHR study that were consistent with results from these previous PE studies include increases in plasminogen, C1r, clusterin, and complement factor H and decreases in A2M and IGHM. The primary causes of these changes remain unknown but are consistent with evidence of enhanced inflammatory tone preceding psychosis and other mental disorders43,44,51-55 and schizophrenia risk associated with genetic variation of complement C4.56

    Several complement proteins emerged as important predictors of transition, including C4BPA, C1r of the antibody-antigen complex mediated pathway, key regulatory protease complement factor I, and terminal pathway components C6 and C8A. These arise from common pathways or functionally interact with coagulation proteins plasminogen and vitamin K–dependent protein S, supporting hypotheses of coagulation activation in psychosis.57 In both the initial and replication experiments, the most highly weighted predictor of transition was A2M (decreased in CHR-T vs CHR-NT), a protease inhibitor with diverse functions, including inhibition of proinflammatory cytokines such as interleukin 1β58 (consistently elevated in FEP59). A2M is a key coagulation inhibitor60 and thus links functionally to our observations of elevated plasminogen in CHR-T. This finding is intriguing given the evidence that blood-derived plasminogen is associated with brain inflammation61 and complement activation.62 In models of multiple sclerosis, blood-brain barrier disruption facilitates transfer of fibrinogen into the brain, where it is deposited as fibrin, causing local inflammation.63 Given evidence for blood-brain barrier disruption in psychosis,64 fibrin may be associated with etiopathogenic mechanisms providing novel therapeutic avenues,65 but this hypothesis requires further investigation.

    We validated differential expression of A2M and C1r using ELISA. The ELISA-based model (model S1) demonstrated fair, although reduced, predictive accuracy. This finding may reflect reduced sensitivity of ELISA and the inability to accurately quantify specific protein isoforms. Several proteins in the highest-weighted 10% of features for transition in study 1 were similarly highly weighted for PEs in study 2, including C4BPA, vitamin K–dependent protein S, A2M, and IGHM (eTable 21 in Supplement 1 summarizes the directionality of association of the 10% highest predictors in model 1a, model 3, and model 4). This observation may suggest a degree of similarity in proteomic changes between young people in the general population who develop PEs and help-seeking individuals at CHR who develop psychosis, but this hypothesis requires confirmation.

    Outside of psychosis outcomes, several proteomic features contributed to prediction of functional outcome (model S2). A2M, IGHM, phospholipid transfer protein, and clusterin were among the 10% highest-weighted predictors. The results of the present study are also in keeping with studies in bipolar disorder and depression reporting decreased A2M, IgM, and C4BPA.66 At least some of these proteomic changes may be common to multiple clinical phenotypes, including neurodegenerative disorders, such as Alzheimer disease.67 Rather than considering such changes as biomarkers of individual disorders, phenotypic manifestations may be clinical markers of a variety of overlapping neuroimmune abnormalities that have their origin in combined genetic56,68 and environmental69-72 factors.

    Limitations

    This study has some limitations. First, these models require validation in independent cohorts to assess generalizability and real-world applicability. Second, differences in protein identifications precluded application of models between studies. However, there are valid reasons not to do so, including differences in outcome (psychotic disorder vs PEs) and age (postpubertal vs peripubertal). Third, data on duration of follow-up and reasons for dropout were not systematically collected in EU-GEI, and we were unable to fully assess the potential implications of these factors. Fourth, the replication experiment was partial because only 2 CHR-T cases were different from the initial experiment. Although our findings were generally replicated, no statement can be made regarding generalizability of model sensitivity. Fifth, participants were nonfasting, and there were no restrictions on time of sample collection. Sixth, other factors, such as childhood adversity, may have contributed to the proteomic changes that we observed,10,71 but these factors require further study.

    Conclusions

    We developed models incorporating proteomic data predicting transition to psychotic disorder in the CHR state. In a general population sample, several of the same proteins contributed to prediction of PEs. Further studies are required to validate these findings, evaluate their causes, and elucidate tractable targets for prediction and prevention of psychosis.

    Back to top
    Article Information

    Accepted for Publication: June 15, 2020.

    Published Online: August 26, 2020. doi:10.1001/jamapsychiatry.2020.2459

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Mongan D et al. JAMA Psychiatry.

    Corresponding Author: David R. Cotter, PhD, Department of Psychiatry, Royal College of Surgeons in Ireland, Education and Research Centre, Beaumont Hospital, Dublin 9, Ireland (drcotter@rcsi.ie).

    Author Contributions: Drs Cotter and McGuire are co–senior authors. Drs Mongan and Cotter 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: Healy, Amminger, Krebs, Bressan, Borgwardt, Ruhrmann, de Haan, Pollak, Rutten, Cannon, Cagney, Cotter, McGuire.

    Acquisition, analysis, or interpretation of data: Mongan, Föcking, Healy, Susai, Heurich, Wynne, Nelson, McGorry, Amminger, Nordentoft, Krebs, Riecher-Rössler, Bressan, Barrantes-Vidal, Borgwardt, Ruhrmann, Sachs, Pantelis, van der Gaag, de Haan, Valmaggia, Pollak, Kempton, Rutten, Whelan, Zammit, Cagney, Cotter, McGuire.

    Drafting of the manuscript: Mongan, Föcking, Healy, Wynne, Cannon, Cotter, McGuire.

    Critical revision of the manuscript for important intellectual content: Mongan, Föcking, Healy, Susai, Heurich, Nelson, McGorry, Amminger, Nordentoft, Krebs, Riecher-Rössler, Bressan, Barrantes-Vidal, Borgwardt, Ruhrmann, Sachs, Pantelis, van der Gaag, de Haan, Valmaggia, Pollak, Kempton, Rutten, Whelan, Zammit, Cagney, Cotter, McGuire.

    Statistical analysis: Mongan, Healy, Whelan, Cagney, Cotter.

    Obtained funding: Nelson, McGorry, Amminger, Nordentoft, Krebs, Riecher-Rössler, Bressan, Borgwardt, Pantelis, Valmaggia, Rutten, Cagney, Cotter, McGuire.

    Administrative, technical, or material support: Susai, Wynne, Riecher-Rössler, Bressan, Barrantes-Vidal, Borgwardt, Pantelis, van der Gaag, Valmaggia, Pollak, Kempton, Cagney, Cotter, McGuire.

    Supervision: Amminger, Nordentoft, Bressan, Borgwardt, Ruhrmann, Sachs, de Haan, Cannon, Cagney, Cotter, McGuire.

    Conflict of Interest Disclosures: A United Kingdom (UK) patent application has been filed in relation to the development of a prognostic test derived from this work (UK patent application 1919155.0). Dr Mongan reported receiving grants from the Wellcome Trust and Health Research Board Ireland and having UK patent application 1919155.0 pending. Mr Healy reported receiving grants from the European Research Council. Dr Krebs reported receiving grants from the French Ministry Programme Hospitalier de Recherche Clinique AOM07-118 and Eisai, receiving grants and personal fees from Otsuka-Lundbeck and Janssen, and having a patent pending. Dr Borgwardt reported receiving grants from the European Community’s Seventh Framework Programme under grant agreement HEALTH-F2-2010-241909 (Project EU-GEI). Dr Ruhrmann reported receiving grants from the European Commission and receiving nonfinancial support from Boehringer Ingelheim. Dr Sachs reported receiving honoraria for consulting and lectures on the topic of schizophrenia. Dr Pantelis reported receiving grants from the Australian National Health and Medical Research Council (NHMRC) and The Lundbeck Foundation and receiving personal fees from Lundbeck Australia Pty Ltd. Dr Kempton reported receiving grants from the European Commission and the Medical Research Council. Dr Cagney reported receiving grants from Health Research Board Ireland and having a patent for a biomarker panel pending. Dr Cotter reported receiving grants from Health Research Board Ireland and having UK patent 1919155.0 pending. No other disclosures were reported.

    Funding/Support: EU-GEI was funded by a Framework 7 Grant (HEALTH-F2-2010-241909) for the European Network of National Schizophrenia Networks Studying Gene-Environment Interactions (EU-GEI) study and by Health Research Board Ireland through a Clinician Scientist Award to Dr Cotter. Additional support was provided by a Medical Research Council Fellowship to Dr Kempton (grant MR/J008915/1) and by the Ministerio de Ciencia, Innovación e Universidades (grant PSI2017-87512-C2-1-R) and Generalitat de Catalunya (grant 2017SGR1612 and Catalan Institution for Research and Advanced Studies [ICREA] Academia award) to Dr Barrantes-Vidal. The UK Medical Research Council and the Wellcome Trust (grant 102215/2/13/2) and the University of Bristol provide core support for the Avon Longitudinal Study of Parents and Children (ALSPAC). A comprehensive list of grant funding is available on the ALSPAC website (http://www.bristol.ac.uk/alspac/external/documents/grant-acknowledgements.pdf). The outcomes data collected in the ALSPAC study that were used in the present study were specifically funded by the Medical Research Council (grant G0701503/85179). Dr Zammit is supported by the Bristol National Institute for Health Research Biomedical Research Centre. Dr Mongan is a fellow of the Irish Clinical Academic Training (ICAT) Programme, which is supported by the Wellcome Trust and Health Research Board Ireland (grant 203930/B/16/Z), the Health Service Executive National Doctors Training and Planning, and the Health and Social Care Research and Development Division, Northern Ireland.

    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.

    Group Information: EU-GEI High Risk Study Group members are Philip McGuire, Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London; Lucia Valmaggia, Department of Psychology, Institute of Psychiatry, Psychology & Neuroscience, King’s College London; Matthew J. Kempton, Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London; Thomas A. Pollak, Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London; Conrad Iyegbe, Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London; Stefania Tognin, Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London; Gemma Modinos, Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London; Lieuwe de Haan, Academic Medical Centre (AMC), Academic Psychiatric Centre, Department Early Psychosis; Mark van der Gaag, VU University, Faculty of Behavioural and Movement Sciences, Department of Clinical Psychology and EMGO+ Institute for Health and Care Research, and Parnassia Psychiatric Institute, Department of Psychosis Research; Eva Velthorst, AMC, Academic Psychiatric Centre, Department Early Psychosis, and Icahn School of Medicine at Mount Sinai, Department of Psychiatry; Tamar C. Kraan, AMC, Academic Psychiatric Centre, Department Early Psychosis; Daniella S. van Dam, AMC, Academic Psychiatric Centre, Department Early Psychosis; Nadine Burger, Parnassia Psychiatric Institute, Department of Psychosis Research; Barnaby Nelson, Centre for Youth Mental Health, University of Melbourne; Patrick D. McGorry, Centre for Youth Mental Health, University of Melbourne; G. Paul Amminger, Centre for Youth Mental Health, University of Melbourne; Christos Pantelis, Center for Neuropsychiatric Schizophrenia Research (CNSR) and Center for Clinical Intervention and Neuropsychiatric Schizophrenia Research (CINS), University of Copenhagen, and University of Copenhagen, Faculty of Health and Medical Sciences, Department of Clinical Medicine; Athena Politis, Centre for Youth Mental Health, University of Melbourne; Joanne Goodall, Centre for Youth Mental Health, University of Melbourne; Anita Riecher-Rössler, University of Basel Psychiatric Hospital; Stefan Borgwardt, University of Basel Psychiatric Hospital; Charlotte Rapp, University of Basel Psychiatric Hospital; Sarah Ittig, University of Basel Psychiatric Hospital; Erich Studerus, University of Basel Psychiatric Hospital; Renata Smieskova, University of Basel Psychiatric Hospital; Rodrigo A. Bressan, LiNC–Lab Interdisciplinar Neurociências Clínicas, Depto Psiquiatria, Escola Paulista de Medicina, Universidade Federal de São Paulo–UNIFESP; Ary Gadelha, LiNC, Depto Psiquiatria, Escola Paulista de Medicina, UNIFESP; Elisa Brietzke, Depto Psiquiatria, Escola Paulista de Medicina, UNIFESP; Graccielle Asevedo, LiNC, Depto Psiquiatria, Escola Paulista de Medicina, UNIFESP; Elson Asevedo, LiNC, Depto Psiquiatria, Escola Paulista de Medicina, UNIFESP; Andre Zugman, LiNC, Depto Psiquiatria, Escola Paulista de Medicina, UNIFESP; Neus Barrantes-Vidal, Departament de Psicologia Clínica i de la Salut (Universitat Autònoma de Barcelona), Fundació Sanitària Sant Pere Claver, Spanish Mental Health Research Network (CIBERSAM); Tecelli Domínguez-Martínez, CONACYT–Dirección de Investigaciones Epidemiológicas y Psicosociales, Instituto Nacional de Psiquiatría Ramón de la Fuente Muñiz; Anna Racioppi, Departament de Psicologia Clínica i de la Salut (Universitat Autònoma de Barcelona); Thomas R. Kwapil, Department of Psychology, University of Illinois at Urbana–Champaign; Manel Monsonet, Departament de Psicologia Clínica i de la Salut (Universitat Autònoma de Barcelona); Araceli Rosa, Departament de Biologia Evolutiva, Ecologia i Ciències Ambientals (Universitat de Barcelona); Oussama Kebir, University Paris Descartes, Hôpital Sainte-Anne, C’JAAD, Service Hospitalo–Universitaire, INSERM U894, Institut de Psychiatrie (CNRS 3557); Claire Daban, University of Paris, GHU-Paris, Sainte-Anne, C’JAAD, Hospitalo–Universitaire Department SHU; Julie Bourgin, University of Paris, GHU-Paris, Sainte-Anne, C’JAAD, Hospitalo–Universitaire Department SHU; Boris Chaumette, University Paris Descartes, Hôpital Sainte-Anne, C’JAAD, Service Hospitalo–Universitaire, INSERM U894, Institut de Psychiatrie (CNRS 3557); Célia Mam-Lam-Fook, University of Paris, GHU-Paris, Sainte-Anne, C’JAAD, Hospitalo–Universitaire Department SHU; Marie-Odile Krebs, University of Paris, GHU-Paris, Sainte-Anne, C’JAAD, Hospitalo–Universitaire Department SHU; Dorte Nordholm, Mental Health Center Copenhagen and CINS, Mental Health Center Glostrup, Mental Health Services in the Capital Region of Copenhagen, University of Copenhagen; Lasse Randers, Mental Health Center Copenhagen and CINS, Mental Health Center Glostrup, Mental Health Services in the Capital Region of Copenhagen, University of Copenhagen; Kristine Krakauer, Mental Health Center Copenhagen and CINS, Mental Health Center Glostrup, Mental Health Services in the Capital Region of Copenhagen, University of Copenhagen; Louise Birkedal Glenthøj, Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, and CNSR and CINS, University of Copenhagen; Birte Glenthøj, CNSR and CINS, University of Copenhagen; Merete Nordentoft, Mental Health Center Copenhagen and CINS, Mental Health Center Glostrup, Mental Health Services in the Capital Region of Copenhagen, University of Copenhagen; Stephan Ruhrmann, Department of Psychiatry and Psychotherapy, University of Cologne; Dominika Gebhard, Department of Psychiatry and Psychotherapy, University of Cologne; Julia Arnhold, psyberlin, Berlin, Germany; Joachim Klosterkötter, Department of Psychiatry and Psychotherapy, University of Cologne; Gabriele Sachs, Department of Psychiatry and Psychotherapy, Medical University of Vienna; Iris Lasser, Department of Psychiatry and Psychotherapy, Medical University of Vienna; Bernadette Winklbaur, Department of Psychiatry and Psychotherapy, Medical University of Vienna; Philippe A. Delespaul, Department of Psychiatry and Neuropsychology, School for Mental Health and Neuroscience, Maastricht University Medical Centre, and Mondriaan Mental Health Trust; Bart P. F. Rutten, Department of Psychiatry and Neuropsychology, School for Mental Health and Neuroscience, Maastricht University Medical Centre; and Jim van Os, Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, and Department of Psychiatry, UMC Utrecht Brain Center, Utrecht University Medical Centre.

    Additional Contributions: We thank Magda Hryniewiecka, MSc (Royal College of Surgeons in Ireland), for technical assistance and contributions in preparing samples for mass spectrometry. She was not compensated for her contributions. We also thank the Mass Spectrometry Core Facility at the Conway Institute, University College Dublin, for support in the development of our proteomics workflows. We are extremely grateful to all the participants, clinical teams, and research staff who contributed to the EU-GEI project. We are extremely grateful to all the families who took part in the ALSPAC, the midwives for their help in recruiting them, and the whole ALSPAC team, which includes interviewers, computer and laboratory technicians, clerical workers, research scientists, volunteers, managers, receptionists, and nurses.

    References
    1.
    Larsen  TK, Melle  I, Auestad  B,  et al.  Early detection of psychosis: positive effects on 5-year outcome.   Psychol Med. 2011;41(7):1461-1469. doi:10.1017/S0033291710002023 PubMedGoogle ScholarCrossref
    2.
    Fusar-Poli  P, Borgwardt  S, Bechdolf  A,  et al.  The psychosis high-risk state: a comprehensive state-of-the-art review.   JAMA Psychiatry. 2013;70(1):107-120. doi:10.1001/jamapsychiatry.2013.269 PubMedGoogle ScholarCrossref
    3.
    Fusar-Poli  P, Bonoldi  I, Yung  AR,  et al.  Predicting psychosis: meta-analysis of transition outcomes in individuals at high clinical risk.   Arch Gen Psychiatry. 2012;69(3):220-229. doi:10.1001/archgenpsychiatry.2011.1472 PubMedGoogle ScholarCrossref
    4.
    Zammit  S, Kounali  D, Cannon  M,  et al.  Psychotic experiences and psychotic disorders at age 18 in relation to psychotic experiences at age 12 in a longitudinal population-based cohort study.   Am J Psychiatry. 2013;170(7):742-750. doi:10.1176/appi.ajp.2013.12060768 PubMedGoogle ScholarCrossref
    5.
    Healy  C, Brannigan  R, Dooley  N,  et al.  Childhood and adolescent psychotic experiences and risk of mental disorder: a systematic review and meta-analysis.   Psychol Med. 2019;49(10):1589-1599. doi:10.1017/S0033291719000485 PubMedGoogle ScholarCrossref
    6.
    Healy  C, Campbell  D, Coughlan  H, Clarke  M, Kelleher  I, Cannon  M.  Childhood psychotic experiences are associated with poorer global functioning throughout adolescence and into early adulthood.   Acta Psychiatr Scand. 2018;138(1):26-34. doi:10.1111/acps.12907 PubMedGoogle ScholarCrossref
    7.
    McGorry  P, Keshavan  M, Goldstone  S,  et al.  Biomarkers and clinical staging in psychiatry.   World Psychiatry. 2014;13(3):211-223. doi:10.1002/wps.20144 PubMedGoogle ScholarCrossref
    8.
    UK Public General Acts. Human Tissue Act 2004. Accessed July 7, 2020. http://www.legislation.gov.uk/ukpga/2004/30/contents
    9.
    van Os  J, Rutten  BP, Myin-Germeys  I,  et al; European Network of National Networks studying Gene-Environment Interactions in Schizophrenia (EU-GEI).  Identifying gene-environment interactions in schizophrenia: contemporary challenges for integrated, large-scale investigations.   Schizophr Bull. 2014;40(4):729-736. doi:10.1093/schbul/sbu069 PubMedGoogle ScholarCrossref
    10.
    Kraan  TC, Velthorst  E, Themmen  M,  et al; EU-GEI High Risk Study.  Child maltreatment and clinical outcome in individuals at ultra-high risk for psychosis in the EU-GEI High Risk Study.   Schizophr Bull. 2018;44(3):584-592. doi:10.1093/schbul/sbw162 PubMedGoogle ScholarCrossref
    11.
    Yung  AR, Yuen  HP, McGorry  PD,  et al.  Mapping the onset of psychosis: the Comprehensive Assessment of At-Risk Mental States.   Aust N Z J Psychiatry. 2005;39(11-12):964-971. doi:10.1080/j.1440-1614.2005.01714.x PubMedGoogle ScholarCrossref
    12.
    English  JA, Lopez  LM, O’Gorman  A,  et al.  Blood-based protein changes in childhood are associated with increased risk for later psychotic disorder: evidence from a nested case-control study of the ALSPAC longitudinal birth cohort.   Schizophr Bull. 2018;44(2):297-306. doi:10.1093/schbul/sbx075 PubMedGoogle ScholarCrossref
    13.
    Aas  IH.  Global Assessment of Functioning (GAF): properties and frontier of current knowledge.   Ann Gen Psychiatry. 2010;9:20. doi:10.1186/1744-859X-9-20 PubMedGoogle ScholarCrossref
    14.
    Goldman  HH, Skodol  AE, Lave  TR.  Revising axis V for DSM-IV: a review of measures of social functioning.   Am J Psychiatry. 1992;149(9):1148-1156. doi:10.1176/ajp.149.9.1148 PubMedGoogle ScholarCrossref
    15.
    Andreasen  NC.  The Scale for the Assessment of Negative Symptoms (SANS): conceptual and theoretical foundations.   Br J Psychiatry Suppl. 1989;(7):49-58. doi:10.1192/S0007125000291496 PubMedGoogle Scholar
    16.
    Overall  JE, Gorham  DR.  The Brief Psychiatric Rating Scale.   Psychological Reports. 1962;10(3):799-812. doi:10.2466/pr0.1962.10.3.799Google ScholarCrossref
    17.
    Montgomery  SA, Asberg  M.  A new depression scale designed to be sensitive to change.   Br J Psychiatry. 1979;134:382-389. doi:10.1192/bjp.134.4.382 PubMedGoogle ScholarCrossref
    18.
    English  JA, Fan  Y, Föcking  M,  et al.  Reduced protein synthesis in schizophrenia patient–derived olfactory cells.   Transl Psychiatry. 2015;5:e663. doi:10.1038/tp.2015.119 PubMedGoogle Scholar
    19.
    Föcking  M, Opstelten  R, Prickaerts  J,  et al.  Proteomic investigation of the hippocampus in prenatally stressed mice implicates changes in membrane trafficking, cytoskeletal, and metabolic function.   Dev Neurosci. 2014;36(5):432-442. doi:10.1159/000365327 PubMedGoogle ScholarCrossref
    20.
    Föcking  M, Sabherwal  S, Cates  HM,  et al.  Complement pathway changes at age 12 are associated with psychotic experiences at age 18 in a longitudinal population-based study: evidence for a role of stress.   Mol Psychiatry. Published online January 11, 2019. doi:10.1038/s41380-018-0306-z PubMedGoogle Scholar
    21.
    Boyd  A, Golding  J, Macleod  J,  et al.  Cohort profile: the “children of the 90s”: the index offspring of the Avon Longitudinal Study of Parents and Children.   Int J Epidemiol. 2013;42(1):111-127. doi:10.1093/ije/dys064 PubMedGoogle ScholarCrossref
    22.
    Fraser  A, Macdonald-Wallis  C, Tilling  K,  et al.  Cohort profile: the Avon Longitudinal Study of Parents and Children: ALSPAC mothers cohort.   Int J Epidemiol. 2013;42(1):97-110. doi:10.1093/ije/dys066 PubMedGoogle ScholarCrossref
    23.
    University of Bristol. Avon Longitudinal Study of Parents and Children. Accessed July 2020. http://www.bristol.ac.uk/alspac/researchers/our-data/
    24.
    Benjamini  Y, Hochberg  Y.  Controlling the false discovery rate: a practical and powerful approach to multiple testing.   J R Stat Soc B. 1995;57(1):289-300. doi:10.1111/j.2517-6161.1995.tb02031.x Google Scholar
    25.
    Cox  J, Mann  M.  Quantitative, high-resolution proteomics for data-driven systems biology.   Annu Rev Biochem. 2011;80:273-299. doi:10.1146/annurev-biochem-061308-093216 PubMedGoogle ScholarCrossref
    26.
    Cox  J, Mann  M.  MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification.   Nat Biotechnol. 2008;26(12):1367-1372. doi:10.1038/nbt.1511 PubMedGoogle ScholarCrossref
    27.
    Lazar  C. Package “imputeLCMD.” Version 2.0. Published February 20, 2015. Accessed April 2019. https://cran.r-project.org/web/packages/imputeLCMD/imputeLCMD.pdf
    28.
    RStudio. PBC. RStudio. Accessed April 2019. https://www.rstudio.com/
    29.
    Cearns  M, Hahn  T, Baune  BT.  Recommendations and future directions for supervised machine learning in psychiatry.   Transl Psychiatry. 2019;9(1):271. doi:10.1038/s41398-019-0607-2 PubMedGoogle ScholarCrossref
    30.
    Szklarczyk  D, Franceschini  A, Wyder  S,  et al.  STRING v10: protein-protein interaction networks, integrated over the tree of life.   Nucleic Acids Res. 2015;43(database issue):D447-D452. doi:10.1093/nar/gku1003 PubMedGoogle ScholarCrossref
    31.
    Radua  J, Ramella-Cravaro  V, Ioannidis  JPA,  et al.  What causes psychosis? an umbrella review of risk and protective factors.   World Psychiatry. 2018;17(1):49-66. doi:10.1002/wps.20490 PubMedGoogle ScholarCrossref
    32.
    Studerus  E, Ramyead  A, Riecher-Rössler  A.  Prediction of transition to psychosis in patients with a clinical high risk for psychosis: a systematic review of methodology and reporting.   Psychol Med. 2017;47(7):1163-1178. doi:10.1017/S0033291716003494 PubMedGoogle ScholarCrossref
    33.
    Malda  A, Boonstra  N, Barf  H,  et al.  Individualized prediction of transition to psychosis in 1,676 individuals at clinical high risk: development and validation of a multivariable prediction model based on individual patient data meta-analysis.   Front Psychiatry. 2019;10:345. doi:10.3389/fpsyt.2019.00345 PubMedGoogle ScholarCrossref
    34.
    Mechelli  A, Lin  A, Wood  S,  et al.  Using clinical information to make individualized prognostic predictions in people at ultra high risk for psychosis.   Schizophr Res. 2017;184:32-38. doi:10.1016/j.schres.2016.11.047 PubMedGoogle ScholarCrossref
    35.
    Schmidt  A, Cappucciati  M, Radua  J,  et al.  Improving prognostic accuracy in subjects at clinical high risk for psychosis: systematic review of predictive models and meta-analytical sequential testing simulation.   Schizophr Bull. 2017;43(2):375-388.PubMedGoogle Scholar
    36.
    Ruhrmann  S, Schultze-Lutter  F, Salokangas  RK,  et al.  Prediction of psychosis in adolescents and young adults at high risk: results from the prospective European Prediction of Psychosis Study.   Arch Gen Psychiatry. 2010;67(3):241-251. doi:10.1001/archgenpsychiatry.2009.206 PubMedGoogle ScholarCrossref
    37.
    Cannon  TD, Yu  C, Addington  J,  et al.  An individualized risk calculator for research in prodromal psychosis.   Am J Psychiatry. 2016;173(10):980-988. doi:10.1176/appi.ajp.2016.15070890 PubMedGoogle ScholarCrossref
    38.
    Koutsouleris  N, Meisenzahl  EM, Davatzikos  C,  et al.  Use of neuroanatomical pattern classification to identify subjects in at-risk mental states of psychosis and predict disease transition.   Arch Gen Psychiatry. 2009;66(7):700-712. doi:10.1001/archgenpsychiatry.2009.62 PubMedGoogle ScholarCrossref
    39.
    Koutsouleris  N, Borgwardt  S, Meisenzahl  EM, Bottlender  R, Möller  HJ, Riecher-Rössler  A.  Disease prediction in the at-risk mental state for psychosis using neuroanatomical biomarkers: results from the FePsy study.   Schizophr Bull. 2012;38(6):1234-1246. doi:10.1093/schbul/sbr145 PubMedGoogle ScholarCrossref
    40.
    Koutsouleris  N, Kambeitz-Ilankovic  L, Ruhrmann  S,  et al; PRONIA Consortium.  Prediction models of functional outcomes for individuals in the clinical high-risk state for psychosis or with recent-onset depression: a multimodal, multisite machine learning analysis.   JAMA Psychiatry. 2018;75(11):1156-1172. doi:10.1001/jamapsychiatry.2018.2165 PubMedGoogle ScholarCrossref
    41.
    Das  T, Borgwardt  S, Hauke  DJ,  et al.  Disorganized gyrification network properties during the transition to psychosis.   JAMA Psychiatry. 2018;75(6):613-622. doi:10.1001/jamapsychiatry.2018.0391 PubMedGoogle ScholarCrossref
    42.
    Koutsouleris  N, Davatzikos  C, Bottlender  R,  et al.  Early recognition and disease prediction in the at-risk mental states for psychosis using neurocognitive pattern classification.   Schizophr Bull. 2012;38(6):1200-1215. doi:10.1093/schbul/sbr037 PubMedGoogle ScholarCrossref
    43.
    Perkins  DO, Jeffries  CD, Addington  J,  et al.  Towards a psychosis risk blood diagnostic for persons experiencing high-risk symptoms: preliminary results from the NAPLS project.   Schizophr Bull. 2015;41(2):419-428. doi:10.1093/schbul/sbu099 PubMedGoogle ScholarCrossref
    44.
    Chan  MK, Krebs  MO, Cox  D,  et al.  Development of a blood-based molecular biomarker test for identification of schizophrenia before disease onset.   Transl Psychiatry. 2015;5:e601. doi:10.1038/tp.2015.91 PubMedGoogle Scholar
    45.
    Ruhrmann  S, Schultze-Lutter  F, Schmidt  SJ, Kaiser  N, Klosterkötter  J.  Prediction and prevention of psychosis: current progress and future tasks.   Eur Arch Psychiatry Clin Neurosci. 2014;264(suppl 1):S9-S16. doi:10.1007/s00406-014-0541-5 PubMedGoogle ScholarCrossref
    46.
    Woo  JJ, Pouget  JG, Zai  CC, Kennedy  JL.  The complement system in schizophrenia: where are we now and what’s next?   Mol Psychiatry. 2020;25(1):114-130. doi:10.1038/s41380-019-0479-0PubMedGoogle ScholarCrossref
    47.
    Sabherwal  S, English  JA, Föcking  M, Cagney  G, Cotter  DR.  Blood biomarker discovery in drug-free schizophrenia: the contribution of proteomics and multiplex immunoassays.   Expert Rev Proteomics. 2016;13(12):1141-1155. doi:10.1080/14789450.2016.1252262 PubMedGoogle ScholarCrossref
    48.
    Li  Y, Zhou  K, Zhang  Z,  et al.  Label-free quantitative proteomic analysis reveals dysfunction of complement pathway in peripheral blood of schizophrenia patients: evidence for the immune hypothesis of schizophrenia.   Mol Biosyst. 2012;8(10):2664-2671. doi:10.1039/c2mb25158b PubMedGoogle ScholarCrossref
    49.
    Kopczynska  M, Zelek  W, Touchard  S,  et al.  Complement system biomarkers in first episode psychosis.   Schizophr Res. 2019;204:16-22. doi:10.1016/j.schres.2017.12.012 PubMedGoogle Scholar
    50.
    Levin  Y, Wang  L, Schwarz  E, Koethe  D, Leweke  FM, Bahn  S.  Global proteomic profiling reveals altered proteomic signature in schizophrenia serum.   Mol Psychiatry. 2010;15(11):1088-1100. doi:10.1038/mp.2009.54 PubMedGoogle ScholarCrossref
    51.
    Föcking  M, Dicker  P, Lopez  LM,  et al.  Differential expression of the inflammation marker IL12p40 in the at-risk mental state for psychosis: a predictor of transition to psychotic disorder?   BMC Psychiatry. 2016;16(1):326. doi:10.1186/s12888-016-1039-7 PubMedGoogle ScholarCrossref
    52.
    Miller  BJ, Buckley  P, Seabolt  W, Mellor  A, Kirkpatrick  B.  Meta-analysis of cytokine alterations in schizophrenia: clinical status and antipsychotic effects.   Biol Psychiatry. 2011;70(7):663-671. doi:10.1016/j.biopsych.2011.04.013 PubMedGoogle ScholarCrossref
    53.
    Khandaker  GM, Pearson  RM, Zammit  S, Lewis  G, Jones  PB.  Association of serum interleukin 6 and C-reactive protein in childhood with depression and psychosis in young adult life: a population-based longitudinal study.   JAMA Psychiatry. 2014;71(10):1121-1128. doi:10.1001/jamapsychiatry.2014.1332 PubMedGoogle ScholarCrossref
    54.
    Laskaris  L, Zalesky  A, Weickert  CS,  et al.  Investigation of peripheral complement factors across stages of psychosis.   Schizophr Res. 2019;204:30-37. doi:10.1016/j.schres.2018.11.035 PubMedGoogle ScholarCrossref
    55.
    Baumeister  D, Russell  A, Pariante  CM, Mondelli  V.  Inflammatory biomarker profiles of mental disorders and their relation to clinical, social and lifestyle factors.   Soc Psychiatry Psychiatr Epidemiol. 2014;49(6):841-849. doi:10.1007/s00127-014-0887-z PubMedGoogle ScholarCrossref
    56.
    Sekar  A, Bialas  AR, de Rivera  H,  et al; Schizophrenia Working Group of the Psychiatric Genomics Consortium.  Schizophrenia risk from complex variation of complement component 4.   Nature. 2016;530(7589):177-183. doi:10.1038/nature16549 PubMedGoogle ScholarCrossref
    57.
    Hoirisch-Clapauch  S, Amaral  OB, Mezzasalma  MA, Panizzutti  R, Nardi  AE.  Dysfunction in the coagulation system and schizophrenia.   Transl Psychiatry. 2016;6:e704. doi:10.1038/tp.2015.204 PubMedGoogle Scholar
    58.
    Rehman  AA, Ahsan  H, Khan  FH.  α-2-Macroglobulin: a physiological guardian.   J Cell Physiol. 2013;228(8):1665-1675. doi:10.1002/jcp.24266 PubMedGoogle ScholarCrossref
    59.
    Upthegrove  R, Manzanares-Teson  N, Barnes  NM.  Cytokine function in medication-naive first episode psychosis: a systematic review and meta-analysis.   Schizophr Res. 2014;155(1-3):101-108. doi:10.1016/j.schres.2014.03.005 PubMedGoogle ScholarCrossref
    60.
    Borth  W.  Alpha 2-macroglobulin, a multifunctional binding protein with targeting characteristics.   FASEB J. 1992;6(15):3345-3353. doi:10.1096/fasebj.6.15.1281457 PubMedGoogle ScholarCrossref
    61.
    Baker  SK, Chen  ZL, Norris  EH, Revenko  AS, MacLeod  AR, Strickland  S.  Blood-derived plasminogen drives brain inflammation and plaque deposition in a mouse model of Alzheimer’s disease.   Proc Natl Acad Sci U S A. 2018;115(41):E9687-E9696. doi:10.1073/pnas.1811172115 PubMedGoogle ScholarCrossref
    62.
    Amara  U, Flierl  MA, Rittirsch  D,  et al.  Molecular intercommunication between the complement and coagulation systems.   J Immunol. 2010;185(9):5628-5636. doi:10.4049/jimmunol.0903678 PubMedGoogle ScholarCrossref
    63.
    Ryu  JK, Petersen  MA, Murray  SG,  et al.  Blood coagulation protein fibrinogen promotes autoimmunity and demyelination via chemokine release and antigen presentation.   Nat Commun. 2015;6:8164. doi:10.1038/ncomms9164 PubMedGoogle ScholarCrossref
    64.
    Pollak  TA, Drndarski  S, Stone  JM, David  AS, McGuire  P, Abbott  NJ.  The blood-brain barrier in psychosis.   Lancet Psychiatry. 2018;5(1):79-92. doi:10.1016/S2215-0366(17)30293-6 PubMedGoogle ScholarCrossref
    65.
    Ryu  JK, Rafalski  VA, Meyer-Franke  A,  et al.  Fibrin-targeting immunotherapy protects against neuroinflammation and neurodegeneration.   Nat Immunol. 2018;19(11):1212-1223. doi:10.1038/s41590-018-0232-x PubMedGoogle ScholarCrossref
    66.
    Comes  AL, Papiol  S, Mueller  T, Geyer  PE, Mann  M, Schulze  TG.  Proteomics for blood biomarker exploration of severe mental illness: pitfalls of the past and potential for the future.   Transl Psychiatry. 2018;8(1):160. doi:10.1038/s41398-018-0219-2 PubMedGoogle ScholarCrossref
    67.
    Westwood  S, Leoni  E, Hye  A,  et al.  Blood-based biomarker candidates of cerebral amyloid using PiB PET in non-demented elderly.   J Alzheimers Dis. 2016;52(2):561-572. doi:10.3233/JAD-151155 PubMedGoogle ScholarCrossref
    68.
    Ripke  S, Neale  BM, Corvin  A,  et al; Schizophrenia Working Group of the Psychiatric Genomics Consortium.  Biological insights from 108 schizophrenia-associated genetic loci.   Nature. 2014;511(7510):421-427. doi:10.1038/nature13595 PubMedGoogle ScholarCrossref
    69.
    Fung  TC, Olson  CA, Hsiao  EY.  Interactions between the microbiota, immune and nervous systems in health and disease.   Nat Neurosci. 2017;20(2):145-155. doi:10.1038/nn.4476 PubMedGoogle ScholarCrossref
    70.
    Khandaker  GM, Zimbron  J, Lewis  G, Jones  PB.  Prenatal maternal infection, neurodevelopment and adult schizophrenia: a systematic review of population-based studies.   Psychol Med. 2013;43(2):239-257. doi:10.1017/S0033291712000736 PubMedGoogle ScholarCrossref
    71.
    Rasmussen  LJH, Moffitt  TE, Arseneault  L,  et al.  Association of adverse experiences and exposure to violence in childhood and adolescence with inflammatory burden in young people.   JAMA Pediatr. 2019;174(1):1-11. doi:10.1001/jamapediatrics.2019.3875PubMedGoogle Scholar
    72.
    van Os  J, Kenis  G, Rutten  BP.  The environment and schizophrenia.   Nature. 2010;468(7321):203-212. doi:10.1038/nature09563 PubMedGoogle ScholarCrossref
    ×