ࡱ > " > bjbj 8^ e e =
8 V " Z\ r! ! ! ! w" " " [ [ [ [ [ [ [ $ 2^ ` V [
" w" w" " " [
! ! \ !B !B !B "
!
! -U !B " [ !B !B : E F ! 8 mF
U *\ 0 Z\ wF R :a ? :a F F :a
G d
" " !B " " " " " [ [ !B " " " Z\ " " " " :a " " " " " " " " " : Genetic overlap between schizophrenia and developmental psychopathology: longitudinal and multivariate polygenic risk prediction of common psychiatric traits during development
Running title: Genetic correlates of schizophrenia during development
Michel G. Nivard, PhD1*, Suzanne H. Gage, PhD2,3,4, Jouke J. Hottenga, PhD1,5, Catharina E.M. van Beijsterveldt, PhD1, Abdel Abdellaoui, PhD1, Meike Bartels1,5, Bart M.L. Baselmans, MSc1,5, Lannie Ligthart, PhD1,5, Beate St Pourcain,PhD 3,6 Dorret I. Boomsma, PhD1,5,7, Marcus R. Munaf, PhD3,4 & Christel M. Middeldorp, PhD MD1,7,8
Biological Psychology, VU University, Amsterdam, The Netherlands
Department of Psychological Sciences, University of Liverpool, UK
MRC Integrative Epidemiology Unit at the University of Bristol, UK
UK Centre for Tobacco and Alcohol Studies, School of Experimental Psychology, University of Bristol, Bristol, UK
Amsterdam public health, Medical Center, Amsterdam, The Netherlands
Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands,
Amsterdam Neuroscience, Amsterdam, The Netherlands
Department of Child and Adolescent Psychiatry, GGZ inGeest / VU University Medical Centre, Amsterdam, The Netherlands
*: Correspondence should be addressed to Michel G Nivard m.g.nivard@vu.nl Van der Boechorstraat 1081 BT Amsterdam, The Netherlands.
Wordcount; abstract: 249; body: 4165.
Figures: 3
Tables 2
Supplementary material: 1 note, 3 figures and 4 tables
Abstract
Background: Several non-psychotic psychiatric disorders in childhood and adolescence can precede the onset of schizophrenia, but the etiology of this relationship remains unclear. We investigated to what extent the association between schizophrenia and psychiatric disorders in childhood is explained by correlated genetic risk factors.
Methods: Polygenic risk scores (PRS), reflecting an individuals genetic risk for schizophrenia, were constructed for 2,588 children from the Netherlands Twin Register (NTR) and 6,127 from the Avon Longitudinal Study of Parents And Children (ALSPAC)). The associations between schizophrenia PRS and measures of anxiety, depression, attention deficit hyperactivity disorder (ADHD), and oppositional defiant disorder/conduct disorder (ODD/CD) were estimated at age 7, 10, 12/13 and 15 years in the two cohorts. Results were then meta-analyzed, and a meta-regression analysis was performed to test differences in effects sizes over e.g., age and disorders.
Results: Schizophrenia PRS were associated with childhood and adolescent psychopathology. Meta-regression analysis showed differences in the associations over disorders, with the strongest association with childhood and adolescent depression and a weaker association for ODD/CD at age 7. The associations increased with age and this increase was steepest for ADHD and ODD/CD. Genetic correlations varied between 0.10 and 0.25.
Conclusion: By optimally using longitudinal data across diagnoses in a multivariate meta-analysis this study sheds light on the development of childhood disorders into severe adult psychiatric disorders. The results are consistent with a common genetic etiology of schizophrenia and developmental psychopathology as well as with a stronger shared genetic etiology between schizophrenia and adolescent onset psychopathology.
Key words: Developmental psychiatry, Schizophrenia prodrome, Genetic epidemiology
Introduction
The onset of schizophrenia generally occurs during adolescence or early adulthood, ADDIN REFMGR.CITE Kessler20073Age of onset of mental disorders: a review of recent literatureJournal3Age of onset of mental disorders: a review of recent literatureKessler,Ronald C.Amminger,G.PaulAguilar-Gaxiola,SergioAlonso,JordiLee,SingUstun,T.Bedirhan2007Not in File359Current opinion in psychiatry204NIH Public AccessCurrent opinion in psychiatry11 but it is well established that non-psychotic psychiatric symptoms can be present in the period before the first psychotic episode. The prodromal phase is characterized by neurodevelopmental deficits, ADDIN REFMGR.CITE Heinrichs199828Neurocognitive deficit in schizophrenia: a quantitative review of the evidenceJournal28Neurocognitive deficit in schizophrenia: a quantitative review of the evidenceHeinrichs,R.WalterZakzanis,Konstantine K.1998Not in File426Neuropsychology123American Psychological Association1931-1559Neuropsychology1Kahn201437The neurobiology and treatment of first-episode schizophreniaJournal37The neurobiology and treatment of first-episode schizophreniaKahn,R.S.Sommer,I.E.2014Not in FileMolecular psychiatryNature Publishing Group1359-4184Molecular psychiatry1Fusar-Poli201338At risk for schizophrenic or affective psychoses? A meta-analysis of DSM/ICD diagnostic outcomes in individuals at high clinical riskJournal38At risk for schizophrenic or affective psychoses? A meta-analysis of DSM/ICD diagnostic outcomes in individuals at high clinical riskFusar-Poli,PaoloBechdolf,AndreasTaylor,Matthew JohnBonoldi,IlariaCarpenter,William T.Yung,Alison RuthMcGuire,Philip2013Not in File923932Schizophrenia bulletin394MPRC0586-7614Schizophrenia bulletin12-4 cognitive learning and memory problems, ADDIN REFMGR.CITE Woodberry200827Premorbid IQ in schizophrenia: a meta-analytic reviewJournal27Premorbid IQ in schizophrenia: a meta-analytic reviewWoodberry,Kristen A.Giuliano,Anthony J.Seidman,Larry J.2008Not in File579587The American journal of psychiatry1655Am Psychiatric AssocThe American journal of psychiatry15 and elevated psychiatric symptoms. ADDIN REFMGR.CITE Cunningham Owens20066Precursors and prodromata of schizophrenia: findings from the Edinburgh High Risk Study and their literature contextJournal6Precursors and prodromata of schizophrenia: findings from the Edinburgh High Risk Study and their literature contextCunningham Owens,D.G.Johnstone,Eve C.2006Not in File15011514Psychological medicine3611Cambridge Univ Press1469-8978Psychological medicine16 Well before the prodromal phase, psychiatric symptoms or disorders are more prevalent in individuals who later develop schizophrenia, as becomes apparent from longitudinal population-based cohorts, ADDIN REFMGR.CITE Cannon200244Evidence for early-childhood, pan-developmental impairment specific to schizophreniform disorder: results from a longitudinal birth cohortJournal44Evidence for early-childhood, pan-developmental impairment specific to schizophreniform disorder: results from a longitudinal birth cohortCannon,MaryCaspi,AvshalomMoffitt,Terrie E.Harrington,HonaLeeTaylor,AlanMurray,Robin M.Poulton,Richie2002Not in File449456Archives of general psychiatry595American Medical Association0003-990XArchives of general psychiatry1Kim-Cohen200345Prior juvenile diagnoses in adults with mental disorder: developmental follow-back of a prospective-longitudinal cohortJournal45Prior juvenile diagnoses in adults with mental disorder: developmental follow-back of a prospective-longitudinal cohortKim-Cohen,JuliaCaspi,AvshalomMoffitt,Terrie E.Harrington,HonaLeeMilne,Barry J.Poulton,Richie2003Not in File709717Archives of general psychiatry607American Medical Association0003-990XArchives of general psychiatry17, 8 retrospective assesments of schizophrenia cases, ADDIN REFMGR.CITE Rossi20007Behavioral neurodevelopment abnormalities and schizophrenic disorder: a retrospective evaluation with the Childhood Behavior Checklist (CBCL)Journal7Behavioral neurodevelopment abnormalities and schizophrenic disorder: a retrospective evaluation with the Childhood Behavior Checklist (CBCL)Rossi,A.Pollice,R.Daneluzzo,E.Marinangeli,M.G.Stratta,P.2000Not in File121128Schizophrenia research442Elsevier0920-9964Schizophrenia research19 and from studies on populations at risk for developing schizophrenia. ADDIN REFMGR.CITE Miller20028Childhood behaviour, psychotic symptoms and psychosis onset in young people at high risk of schizophrenia: early findings from the Edinburgh High Risk StudyJournal8Childhood behaviour, psychotic symptoms and psychosis onset in young people at high risk of schizophrenia: early findings from the Edinburgh High Risk StudyMiller,P.M.Byrne,M.Hodges,A.Lawrie,S.M.Johnstone,E.C.2002Not in File173179Psychological medicine3201Cambridge Univ Press1469-8978Psychological medicine110 Both externalizing symptoms or disorders, including attention deficit hyperactivity disorder, conduct disorder, aggression, and anti-social behavior, ADDIN REFMGR.CITE Muratori200539Childhood psychopathological antecedents in early onset schizophreniaJournal39Childhood psychopathological antecedents in early onset schizophreniaMuratori,F.Salvadori,F.D-Arcangelo,G.Viglione,V.Picchi,L.2005Not in File309314European psychiatry204Elsevier0924-9338European psychiatry1Keshavan200526Premorbid indicators and risk for schizophrenia: a selective review and updateJournal26Premorbid indicators and risk for schizophrenia: a selective review and updateKeshavan,Matcheri S.Diwadkar,Vaibhav A.Montrose,Debra M.Rajarethinam,RajaprabhakaranSweeney,John A.2005Not in File4557Schizophrenia research791Elsevier0920-9964Schizophrenia research111, 12 and internalizing symptoms or disorders, including anxiety and depression, are associated with a higher risk of schizophrenia. ADDIN REFMGR.CITE Cannon200244Evidence for early-childhood, pan-developmental impairment specific to schizophreniform disorder: results from a longitudinal birth cohortJournal44Evidence for early-childhood, pan-developmental impairment specific to schizophreniform disorder: results from a longitudinal birth cohortCannon,MaryCaspi,AvshalomMoffitt,Terrie E.Harrington,HonaLeeTaylor,AlanMurray,Robin M.Poulton,Richie2002Not in File449456Archives of general psychiatry595American Medical Association0003-990XArchives of general psychiatry1Kim-Cohen200345Prior juvenile diagnoses in adults with mental disorder: developmental follow-back of a prospective-longitudinal cohortJournal45Prior juvenile diagnoses in adults with mental disorder: developmental follow-back of a prospective-longitudinal cohortKim-Cohen,JuliaCaspi,AvshalomMoffitt,Terrie E.Harrington,HonaLeeMilne,Barry J.Poulton,Richie2003Not in File709717Archives of general psychiatry607American Medical Association0003-990XArchives of general psychiatry1Muratori200539Childhood psychopathological antecedents in early onset schizophreniaJournal39Childhood psychopathological antecedents in early onset schizophreniaMuratori,F.Salvadori,F.D-Arcangelo,G.Viglione,V.Picchi,L.2005Not in File309314European psychiatry204Elsevier0924-9338European psychiatry1Meyer200510The psychosis prodrome in adolescent patients viewed through the lens of DSM-IVJournal10The psychosis prodrome in adolescent patients viewed through the lens of DSM-IVMeyer,Stephanie E.Bearden,Carrie E.Lux,Sabrina R.Gordon,Jamie L.Johnson,Jennifer K.O'Brien,Mary P.Niendam,Tara A.Loewy,Rachel L.Ventura,JosephCannon,Tyrone D.2005Not in File434451Journal of Child & Adolescent Psychopharmacology153Mary Ann Liebert, Inc. 2 Madison Avenue Larchmont, NY 10538 USA1044-5463Journal of Child & Adolescent Psychopharmacology1Fusar-Poli201435Comorbid depressive and anxiety disorders in 509 individuals with an at-risk mental state: impact on psychopathology and transition to psychosisJournal35Comorbid depressive and anxiety disorders in 509 individuals with an at-risk mental state: impact on psychopathology and transition to psychosisFusar-Poli,PaoloNelson,BarnabyValmaggia,LuciaYung,Alison R.McGuire,Philip K.2014Not in File120131Schizophrenia bulletin401MPRC0586-7614Schizophrenia bulletin1Gajwani201336Attachment: Developmental pathways to affective dysregulation in young people at ultra-high risk of developing psychosisJournal36Attachment: Developmental pathways to affective dysregulation in young people at ultra-high risk of developing psychosisGajwani,RuchikaPatterson,PaulBirchwood,Max2013Not in File424437British Journal of Clinical Psychology524Wiley Online Library2044-8260British Journal of Clinical Psychology1Maibing201472Risk of schizophrenia increases after all child and adolescent psychiatric disorders: a nationwide studyJournal72Risk of schizophrenia increases after all child and adolescent psychiatric disorders: a nationwide studyMaibing,Cecilie FrejstrupPedersen,Carsten B+©ckerBenros,Michael EriksenMortensen,Preben BoDalsgaard,S.+Nordentoft,Merete2014Not in Filesbu119Schizophrenia bulletinMPRC0586-7614Schizophrenia bulletin17, 8, 11, 13-16 In sum, these studies indicate that the onset of schizophrenia can be preceded by a broad range of childhood and adolescent psychopathology.
The early detection of schizophrenia can improve outcomes, ADDIN REFMGR.CITE Kim-Cohen200345Prior juvenile diagnoses in adults with mental disorder: developmental follow-back of a prospective-longitudinal cohortJournal45Prior juvenile diagnoses in adults with mental disorder: developmental follow-back of a prospective-longitudinal cohortKim-Cohen,JuliaCaspi,AvshalomMoffitt,Terrie E.Harrington,HonaLeeMilne,Barry J.Poulton,Richie2003Not in File709717Archives of general psychiatry607American Medical Association0003-990XArchives of general psychiatry18 and preventive treatment for individuals at risk for schizophrenia can reduce the risk of psychosis. ADDIN REFMGR.CITE Perkins200522Relationship between duration of untreated psychosis and outcome in first-episode schizophrenia: a critical review and meta-analysisJournal22Relationship between duration of untreated psychosis and outcome in first-episode schizophrenia: a critical review and meta-analysisPerkins,Diana O.Gu,HongbinBoteva,KalinaLieberman,Jeffrey A.2005Not in File17851804American Journal of Psychiatry16210Am Psychiatric Assoc0002-953XAmerican Journal of Psychiatry1Stafford201330Early interventions to prevent psychosis: systematic review and meta-analysisJournal30Early interventions to prevent psychosis: systematic review and meta-analysisStafford,Megan R.Jackson,HannahMayo-Wilson,EvanMorrison,Anthony P.Kendall,Tim2013Not in Filef185Bmj346BMJ Publishing Group Ltd1756-1833Bmj117, 18 Insight into the risk factors associated with the predictors of schizophrenia may facilitate early detection. Here, we focused on the role of genetic risk factors. Schizophrenia is highly heritability (approximately 80% ADDIN REFMGR.CITE Sullivan200343Schizophrenia as a complex trait: evidence from a meta-analysis of twin studiesJournal43Schizophrenia as a complex trait: evidence from a meta-analysis of twin studiesSullivan,Patrick F.Kendler,Kenneth S.Neale,Michael C.2003Not in File11871192Archives of general psychiatry6012American Medical Association0003-990XArchives of general psychiatry119) and molecular genetic and twin and family studies generally found evidence for a genetic association between childhood and adult psychopathologies, ADDIN REFMGR.CITE Sanchez-Gistau201571Psychiatric disorders in child and adolescent offspring of patients with schizophrenia and bipolar disorder: A controlled studyJournal71Psychiatric disorders in child and adolescent offspring of patients with schizophrenia and bipolar disorder: A controlled studySanchez-Gistau,VanessaRomero,SoledadMoreno,Doloresde la Serna,ElenaBaeza,InmaculadaSugranyes,GiselaMoreno,CarmenSanchez-Gutierrez,TeresaRodriguez-Toscano,ElisaCastro-Fornieles,Josefina2015Not in File197203Schizophrenia research1681Elsevier0920-9964Schizophrenia research1Nivard201415Stability in symptoms of anxiety and depression as a function of genotype and environment: a longitudinal twin study from ages 3 to 63 yearsJournal15Stability in symptoms of anxiety and depression as a function of genotype and environment: a longitudinal twin study from ages 3 to 63 yearsNivard,M.G.Dolan,C.V.Kendler,K.S.Kan,K.J.Willemsen,G.van Beijsterveldt,C.E.Lindauer,R.J.van Beek,J.H.D.A.Geels,L.M.Bartels,M.2014Not in File111Psychological medicineCambridge Univ Press0033-2917Psychological medicine1Kan201317Genetic and environmental stability in attention problems across the lifespan: evidence from the Netherlands twin registerJournal17Genetic and environmental stability in attention problems across the lifespan: evidence from the Netherlands twin registerKan,Kees JanDolan,Conor V.Nivard,Michel G.Middeldorp,Christel M.van Beijsterveldt,Catharina EMWillemsen,GonnekeBoomsma,Dorret I.2013Not in File1225Journal of the American Academy of Child & Adolescent Psychiatry521Elsevier0890-8567Journal of the American Academy of Child & Adolescent Psychiatry1Benke201418A genome-wide association meta-analysis of preschool internalizing problemsJournal18A genome-wide association meta-analysis of preschool internalizing problemsBenke,Kelly S.Nivard,Michel G.Velders,Fleur P.Walters,Raymond K.Pappa,IreneScheet,Paul A.Xiao,XiangjunEhli,Erik A.Palmer,Lyle J.Whitehouse,Andrew JO2014Not in File667676Journal of the American Academy of Child & Adolescent Psychiatry536Elsevier0890-8567Journal of the American Academy of Child & Adolescent Psychiatry1Cross-Disorder Group of the Psychiatric Genomics Consortium201314Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsJournal14Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsCross-Disorder Group of the Psychiatric Genomics Consortium2013Not in File984994Nature genetics459Nature Publishing Group1061-4036Nature genetics1Cross-Disorder Group of the Psychiatric Genomics Consortium201346Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysisJournal46Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysisCross-Disorder Group of the Psychiatric Genomics Consortium2013Not in File13711379The Lancet3819875Elsevier0140-6736The Lancet1Hamshere201381Shared polygenic contribution between childhood attention-deficit hyperactivity disorder and adult schizophreniaJournal81Shared polygenic contribution between childhood attention-deficit hyperactivity disorder and adult schizophreniaHamshere,Marian L.Stergiakouli,EvangeliaLangley,KateMartin,JoannaHolmans,PeterKent,LindseyOwen,Michael J.Gill,MichaelThapar,AnitaO Donovan,Mick2013Not in File107111The British Journal of Psychiatry2032RCP0007-1250The British Journal of Psychiatry1Jones201682Phenotypic manifestation of genetic risk for schizophrenia
during adolescence in the general population.Journal82Phenotypic manifestation of genetic risk for schizophrenia
during adolescence in the general population.Jones,H.J.Stergiakouli,E.Tansey,K.E.Hubbard,L.Heron,J.Cannon,M.Holmans,P.Lewis,G.Linden,D.E.J.Jones,P.B.Smith,G.D.O'Donnovan,M.Owen,M.J.Walter,J.T.Zammit,S.2016Not in FileJAMA Psychiatry.733JAMA Psychiatry.120-27 with one exception. ADDIN REFMGR.CITE Krapohl201569Phenome-wide analysis of genome-wide polygenic scoresJournal69Phenome-wide analysis of genome-wide polygenic scoresKrapohl,E.Euesden,J.Zabaneh,D.Pingault,J.B.Rimfeld,K.von Stumm,S.Dale,P.S.Breen,G.O'Reilly,P.F.Plomin,R.2015Not in FileMolecular psychiatryAOPNature Publishing Group1359-4184Molecular psychiatry128 Consequently, we hypothesized that genetic risk factors for schizophrenia are associated with childhood and adolescent psychopathology. We further expected this association to become stronger from childhood into adolescence, since the the prevalence rates of prodromal symptoms and of psychiatric disorders genetically correlated to schizophrenia (i.e., major depression and bipolar disorder) show a marked increase during adolescence. ADDIN REFMGR.CITE Kessler20073Age of onset of mental disorders: a review of recent literatureJournal3Age of onset of mental disorders: a review of recent literatureKessler,Ronald C.Amminger,G.PaulAguilar-Gaxiola,SergioAlonso,JordiLee,SingUstun,T.Bedirhan2007Not in File359Current opinion in psychiatry204NIH Public AccessCurrent opinion in psychiatry11 Previous molecular genetics studies ADDIN REFMGR.CITE Jones201682Phenotypic manifestation of genetic risk for schizophrenia
during adolescence in the general population.Journal82Phenotypic manifestation of genetic risk for schizophrenia
during adolescence in the general population.Jones,H.J.Stergiakouli,E.Tansey,K.E.Hubbard,L.Heron,J.Cannon,M.Holmans,P.Lewis,G.Linden,D.E.J.Jones,P.B.Smith,G.D.O'Donnovan,M.Owen,M.J.Walter,J.T.Zammit,S.2016Not in FileJAMA Psychiatry.733JAMA Psychiatry.1Krapohl201569Phenome-wide analysis of genome-wide polygenic scoresJournal69Phenome-wide analysis of genome-wide polygenic scoresKrapohl,E.Euesden,J.Zabaneh,D.Pingault,J.B.Rimfeld,K.von Stumm,S.Dale,P.S.Breen,G.O'Reilly,P.F.Plomin,R.2015Not in FileMolecular psychiatryAOPNature Publishing Group1359-4184Molecular psychiatry127, 28 have also related schizophrenia to psychopathology in childhood or adolecence, but never considered how the genetics associations with psychopathology developed from childhood into adolescence.
We tested these hypotheses with a novel approach which involves the meta-analysis of multiple polygenic risk score (PRS) analyses of the genetic associations between schizophrenia and longitudinal psychopathology measures. The PRS were based on the most recent schizophrenia GWA meta-analysis ADDIN REFMGR.CITE Schizophrenia Working Group of the Psychiatric Genomics Consortium20145Biological insights from 108 schizophrenia-associated genetic lociJournal5Biological insights from 108 schizophrenia-associated genetic lociSchizophrenia Working Group of the Psychiatric Genomics Consortium2014Not in File421427Nature5117510Nature Publishing Group0028-0836Nature129 that yielded 108 genome wide associations, which provides an excellent starting point to investigate the genetic overlap between schizophrenia and other traits (for a review of PRS analyses see: ADDIN REFMGR.CITE Wray201453Research Review: Polygenic methods and their application to psychiatric traitsJournal53Research Review: Polygenic methods and their application to psychiatric traitsWray,Naomi R.Lee,Sang HongMehta,DivyaVinkhuyzen,Anna AEDudbridge,FrankMiddeldorp,Christel M.2014Not in File10681087Journal of child psychology and psychiatry5510Wiley Online Library1469-7610Journal of child psychology and psychiatry1Purcell200912Common polygenic variation contributes to risk of schizophrenia and bipolar disorderJournal12Common polygenic variation contributes to risk of schizophrenia and bipolar disorderPurcell,Shaun M.Wray,Naomi R.Stone,Jennifer L.Visscher,Peter M.O'Donovan,Michael C.Sullivan,Patrick F.Sklar,PamelaRuderfer,Douglas M.McQuillin,AndrewMorris,Derek W.2009Not in File748752Nature4607256Nature Publishing Group0028-0836Nature1Wray201313Pitfalls of predicting complex traits from SNPsJournal13Pitfalls of predicting complex traits from SNPsWray,Naomi R.Yang,JianHayes,Ben J.Price,Alkes L.Goddard,Michael E.Visscher,Peter M.2013Not in File507515Nature Reviews Genetics147Nature Publishing Group1471-0056Nature Reviews Genetics130-32). The schizophrenia PRS were used to predict DSM-IV ADDIN REFMGR.CITE American Psychiatric Association199455Diagnostic and statistical manual of mental diseasesJournal55Diagnostic and statistical manual of mental diseasesAmerican Psychiatric Association1994Not in FileDSM-IV.4th edn.Washington (DC): American Psychiatric AssociationDSM-IV.4th edn.Washington (DC): American Psychiatric Association133 based measures of anxiety, depression, attention deficit hyperactivity disorder (ADHD) and oppositional deviant disorder and conduct disorder (ODD/CD) assessed at ages 7, 10, 12/13 and 15 years, in two large cohorts. The 192 univariate PRS predictions were subjected to a multivariate meta-analysis, and a meta-regression analysis. The multivariate meta-regression framework provided the opportunity to test for differences in the association between schizophrenia PRS and childhood psychopathology across cohorts, disorders and over age.
Methods
Subjects
The Netherlands Twin Register (NTR) (HYPERLINK "http://www.tweelingenregister.org"www.tweelingenregister.org) follows newborn and adult twins. In the Young NTR (YNTR), twins are registered by their parents and followed from birth onwards. Until age 12, parents complete surveys to report on their twins. From age 14 onwards, information is collected by means of self-report. ADDIN REFMGR.CITE van Beijsterveldt201324The Young Netherlands Twin Register (YNTR): longitudinal twin and family studies in over 70,000 childrenJournal24The Young Netherlands Twin Register (YNTR): longitudinal twin and family studies in over 70,000 childrenvan Beijsterveldt,Catharina EMGroen-Blokhuis,MariaHottenga,Jouke JanFranic,SanjaHudziak,James J.Lamb,DianeHuppertz,Charlottede Zeeuw,EvelineNivard,MichelSchutte,Nienke2013Not in File252267Twin Research and Human Genetics1601Cambridge Univ Press1839-2628Twin Research and Human Genetics134 In the current study, maternal ratings of childhood psychopathology collected at age 7, 10, and 12 years were analyzed as well as self-report data collected between ages 14-16 years. The number of genotyped children with scores available varied between 1,223 and 2,588 depending on age group (Supplementary Table S1). Informed consent was obtained from all participants. The study was approved by the Central Ethics Committee on Research Involving Human Subjects of the VU University Medical Centre, Amsterdam, an Institutional Review Board certified by the U.S. Office of Human Research Protections (IRB number IRB-2991 under Federal-wide Assurance-3703; IRB/institute codes, NTR 03-180).
The Avon Longitudinal Study of Parents And Children (ALSPAC) (HYPERLINK "http://www.bristol.ac.uk/alspac/"www.bristol.ac.uk/alspac/) consists of mothers and their children, born between 1990 and 1991 in the Avon area in southwest England, UK. ADDIN REFMGR.CITE Boyd201225Cohort profile: the `children of the 90s` the index offspring of the Avon Longitudinal Study of Parents and ChildrenJournal25Cohort profile: the `children of the 90s` the index offspring of the Avon Longitudinal Study of Parents and ChildrenBoyd,AndyGolding,JeanMacleod,JohnLawlor,Debbie A.Fraser,AbigailHenderson,JohnMolloy,LynnNess,AndyRing,SusanSmith,George Davey2012Not in Filedys064International journal of epidemiologyIEA0300-5771International journal of epidemiology135 The ALSPAC cohort includes maternal ratings of psychopathology at age 7, 10, 13, and 15 and self-ratings at 15 years. The number of genotyped children at each age group varied between 4445 and 6127 (Table S1). Ethical approval for the study was obtained from the ALSPAC Ethics and Law Committee and the Local Research Ethics Committees. The study website contains details of all data available through a fully searchable data dictionary (HYPERLINK "http://www.bris.ac.uk/alspac/researchers/data-access/data-dictionary"www.bris.ac.uk/alspac/researchers/data-access/data-dictionary).
Measures
In the NTR, psychopathology was measured with DSM-IV based symptom scales ADDIN REFMGR.CITE Ebesutani201064Concurrent validity of the Child Behavior Checklist DSM-oriented scales: Correspondence with DSM diagnoses and comparison to syndrome scalesJournal64Concurrent validity of the Child Behavior Checklist DSM-oriented scales: Correspondence with DSM diagnoses and comparison to syndrome scalesEbesutani,ChadBernstein,AdamNakamura,Brad J.Chorpita,Bruce F.Higa-McMillan,Charmaine K.Weisz,John R.Research Network on Youth Mental Health2010Not in File373384Journal of Psychopathology and Behavioral Assessment323Springer0882-2689Journal of Psychopathology and Behavioral Assessment136 of the age appropriate versions of the Achenbach System of Empirically Based Assessment (ASEBA). For ages 7 to 12, maternal Child Behavior Checklist (CBCL) ratings of the anxiety disorder scale (anxiety), the affective disorder scale (depression), the attention deficit hyperactivity disorder scale (ADHD), and a combined oppositional deviant disorder and conduct disorder ((ODD/CD) scale were analyzed. From age 14 onwards, self-ratings of these scales were analyzed.
In ALSPAC, psychopathology was assessed using the development and wellbeing assessment (DAWBA), which measures the presence of symptoms required for a DSM-IV diagnosis. ADDIN REFMGR.CITE Goodman200033The Development and Well-Being Assessment: description and initial validation of an integrated assessment of child and adolescent psychopathologyJournal33The Development and Well-Being Assessment: description and initial validation of an integrated assessment of child and adolescent psychopathologyGoodman,RobertFord,TamsinRichards,HilaryGatward,RebeccaMeltzer,Howard2000Not in File645655Journal of child psychology and psychiatry4105Cambridge Univ Press1469-7610Journal of child psychology and psychiatry137 Disorders comparable to the ASEBA scales were included in the analyses: any anxiety disorder (anxiety), major depression (depression), attention deficit hyperactivity disorder (ADHD), and combined oppositional deviant disorder and conduct disorder (ODD/CD). ADDIN REFMGR.CITE Goodman201134The 'DAWBA bands` as an ordered-categorical measure of child mental health: description and validation in British and Norwegian samplesJournal34The 'DAWBA bands` as an ordered-categorical measure of child mental health: description and validation in British and Norwegian samplesGoodman,AnnaHeiervang,EinarCollishaw,StephanGoodman,Robert2011Not in File521532Social psychiatry and psychiatric epidemiology466Springer0933-7954Social psychiatry and psychiatric epidemiology138 Any anxiety disorder included generalized anxiety disorder, specific phobia, social phobia (at age 7, 10, 13, and 15), separation anxiety disorder (at age 7, 10, and 13), and panic disorder and agoraphobia (at age 15). At ages 7, 10, and 13 all ratings were maternal ratings. At age 15, ADHD and CD/ODD were rated by mothers, and anxiety and depression were self-ratings. The DAWBA yields a diagnosis, but also a more finely grained indicator of disease risk, the DAWBA band. DAWBA band scores, which range from 0 to 5, correspond to probabilities of <0.01%, 0.5%, 3%, 15%, 50%, and >70% of satisfying DSM-IV diagnostic criteria.
Genotyping
Genotyping and genotype quality control were performed in accordance with common standards to (for a detailed description see Supplementary Note 1).
Polygenic risk scores
Polygenic risk scores (PRS) were calculated by summing the number of risk alleles across all genetic loci (coded as 0,1,2), weighted by the schizophrenia risk conferred by each locus. The risk conferred by each locus was based on the results from the most recent genome-wide association meta-analysis for schizophrenia (PGC-SCZ2, available online: HYPERLINK "http://www.med.unc.edu/pgc/downloads"http://www.med.unc.edu/pgc/downloads). ADDIN REFMGR.CITE Schizophrenia Working Group of the Psychiatric Genomics Consortium20145Biological insights from 108 schizophrenia-associated genetic lociJournal5Biological insights from 108 schizophrenia-associated genetic lociSchizophrenia Working Group of the Psychiatric Genomics Consortium2014Not in File421427Nature5117510Nature Publishing Group0028-0836Nature129 For all participants, we calculated PRS using LDpred, ADDIN REFMGR.CITE Vilhjalmsson201567Modeling Linkage Disequilibrium Increases Accuracy of Polygenic Risk ScoresJournal67Modeling Linkage Disequilibrium Increases Accuracy of Polygenic Risk ScoresVilhjalmsson,BjarniYang,JianFinucane,Hilary KiyoGusev,AlexanderLindstrom,SaraRipke,StephanGenovese,GiulioLoh,Po RuBhatia,GauravDo,Ron2015Not in File015859bioRxivCold Spring Harbor Labs JournalsbioRxiv139 a method which accounts for correlations between adjacent genetic loci, and adjusts the effect size for each locus in a Linkage Disequilibrium (LD) block to avoid inflation due to LD. LDpred further uses an expectation for the per locus risk, which is based on the expected degree of polygenicity in a trait, i.e., in the case of low polygenicity, a small proportion of the SNPs, e.g. 1%, explains the total genetic variation and the per locus effect is high, whereas in the case of high polygenicity a large proportion of the SNP, e.g., 50% explains the genetic variation and the per locus effect is low. Since it is of interest to know the proportion of the genome that exerts an influence on a trait, we computed 6 PRS at 6 different priors for the proportion of SNPs with a casual effect (0.01, 0.05, 0.1, 0.25, 0.5, 1) and investigated at which prior the predictions were most optimal. Setting a prior on the polygenicity determines for how many loci LDpred expects the effect size to be zero. The discovery markers were not pruned and neither were markers a priori eliminated based on thresholds. Instead, the effect of all markers, the LD between markers and the prior expectation of the degree of polygenicity were leveraged to obtain optimal weights for all markers. For each prior a set of weights was obtained, which was converted in a polygenic score for each subject. The inclusion criteria for SNPs were minor allele frequency above 5% and high imputation quality (R2 > .9). The PRS were scaled to unit variance and mean centered within cohort.
Statistical analyses
In NTR and in ALSPAC, 96 (4 age bins x 4 disorders x 6 polygenic scores) regression analyses were performed to analyze the prediction of the psychopathology measures by the schizophrenia PRS. Psychopathology measures were scaled to unit variance. As the NTR contained related individuals, the linear regression was performed using a generalized estimation equation with exchangeable background correlations within family, and robust standard errors. This procedure adequately corrects for the presence of related individuals in the sample. ADDIN REFMGR.CITE Minicâ201441Sandwich corrected standard errors in family-based genome-wide association studiesJournal41Sandwich corrected standard errors in family-based genome-wide association studiesMinicâ,Camelia C.Dolan,Conor V.Kampert,Maarten MDBoomsma,Dorret I.Vink,Jacqueline M.2014Not in FileEuropean Journal of Human GeneticsNature Publishing Group1018-4813European Journal of Human Genetics140 In the ALSPAC sample, an ordered logistic regression was performed since the DAWBA bands are ordered categorical variables. The ordered logistic regression in ALSPAC was transformed to a scale where the underlying latent variable has variance 1. This results in comparable betas in the two cohorts, as in both samples an 1 SD increase in the schizophrenia PRS results in an 1 SD increase in the (latent) phenotype. The expression of the effect sizes on a common scale enabled a meta-analysis of regression coefficients from the 96 NTR and 96 ALSPAC analyses. Meta-analyses were performed in the metaphor R-package. ADDIN REFMGR.CITE Viechtbauer201068Conducting meta-analyses in R with the metafor packageJournal68Conducting meta-analyses in R with the metafor packageViechtbauer,Wolfgang2010Not in File148Journal of Statistical Software363Journal of Statistical Software141 In contrast to most meta-analyses, the outcome variables were correlated since within the NTR and ALSPAC the same individuals were repeatedly assessed. The polygenic risk scores were also correlated since they were based on a common set of effect sizes, and only differ in the degree of polygenicity assumed in their construction. These correlations result in dependencies between the parameters to be meta-analyzed. We accounted for this in the meta-analysis by specifying the error covariance matrix as the observed correlations between traits and PRS (see Supplementary Note 1 for type 1 error simulation and sensitivity analysis).
In a meta-analysis, we tested whether the effect sizes obtained from the 192 univariate PRS analyses departed from zero. We subsequently used a meta-regression to model differences in the effect sizes of the association between PRS and childhood psychopathology as a function of cohort, age, prior and disorder. To clarify, by adding the variable age to the meta-regression model, it is tested whether the observed effect sizes differ over the four age groups, i.e., become smaller or larger with age. Cohort was coded 1 for ALSPAC and 0 for NTR. Age was coded in years over seven (age seven was coded as zero). We considered 4 meta-regression models, which include an increasing number of predictors (see Table 1). The most comprehensive model included cohort, age, prior, prior2, disorder, age X disorder, age2, age2 X disorder. To guard against over fitting, which is a risk in meta-regression, ADDIN REFMGR.CITE Higgins200485Controlling the risk of spurious findings from meta-regressionJournal85Controlling the risk of spurious findings from meta-regressionHiggins,JulianThompson,Simon G.2004Not in File16631682Statistics in medicine2311Wiley Online Library1097-0258Statistics in medicine142 we performed 1,000 parametric resamples of the data and performed the model selection on each resample (see Supplementary Note 1). We report the proportion of resamples in which each model is selected based on the AIC. To check whether the presence of undetected random effects influenced the fixed effects meta-regression, we performed two additional random effects meta-regression analyses as a robustness check.(see Supplemental Note 1).
To summarize our methodology, first, regression analyses were performed to estimate the effect sizes for the associations between schizophrenia PRS and each disorder at each age at multiple SNP priors across the two studies. Subsequently, these results were analyzed in a multivariate meta-analytic model. In this model, we tested whether schizophrenia PRS are associated with childhood psychopathology and whether the magnitude of this association depends on factors such as age or disorder. This meta-analysis resulted in a best-fitting model describing the relationship between childhood psychopathology and schizophrenia PRS.
The regression coefficients and the variance explained by the PRS are expected to be small, as their effect is determined largely by the ratio of the discovery sample size to the number of independent genetic effects. The number of independent genetic effects on schizophrenia has been show to be large (up to ~70% of the regions in the human genome could affect schizophrenia ADDIN REFMGR.CITE Loh201566Contrasting genetic architectures of schizophrenia and other complex diseases using fast variance-components analysisJournal66Contrasting genetic architectures of schizophrenia and other complex diseases using fast variance-components analysisLoh,Po RuBhatia,GauravGusev,AlexanderFinucane,Hilary K.Bulik-Sullivan,Brendan K.Pollack,Samuela J.de Candia,Teresa R.Lee,Sang HongWray,Naomi R.Kendler,Kenneth S.2015Not in FileNature geneticsNature Publishing Group1061-4036Nature genetics143). By accounting for the sample size in the discovery study and the number of expected, independent genetic loci, regression coefficients as obtained in polygenic risk score analyses can be transformed to genetic correlations between traits. ADDIN REFMGR.CITE Dudbridge201387Power and predictive accuracy of polygenic risk scoresJournal87Power and predictive accuracy of polygenic risk scoresDudbridge,Frank2013Not in Filee1003348PLoS Genet93Public Library of Science1553-7404PLoS Genet144 We calculated the genetic correlations between schizophrenia and childhood psychopathology based on the best fitting meta-regression model. Since this calculation relies on assumptions regarding the variance explained by the SNPs and the number of independent SNPs influencing the disorders, we also present the results of the calculations while making different assumptions on the variance explained by all SNPs.
Results
The descriptive statistics of the psychopathology measures revealed the expected sex differences of adolescent girls scoring higher on internalizing disorders than boys and boys scoring generally higher than girls on ADHD and ODD/CD (Table S1-S2 & Supplementary Note 1). Consequently, sex was included as a covariate in the PRS analyses.
Meta-analysis of the associations between all schizophrenia PRS and all childhood psychopathology measures revealed a significant positive association (estimate = 0.0182, se= 0.005, Z = 3.66, p = 0.0002). This meta-analysis accounted for the dependence between outcomes.
Given the association between childhood psychopathology and schizophrenia PRS, we continued with meta-regression analyses. Model fit statistics, and comparative model fit statistics for the four meta-analytic models are presented in Table 1. The model in which the effect sizes of the associations between childhood psychopathology and schizophrenia PRS were predicted by age, prior, prior2, disorder and age x disorder (i.e., a different relationship between disorder and schizophrenia PRS over age) outperformed the basic model which only allowed for differences in effect sizes between cohorts. Inclusion of nonlinear age effects did not yield an improvement in fit. Likelihood-ratio testing and AIC suggested that Model 3 provided the best balance between parsimony and model complexity. In 77.1% of parametric resamples, model 4 provided the best fit to the data according to the AIC, in 99.5% of the resampled datasets either model 3 or 4 provided the best fit. The increased complexity in model 4 by the addition of non-linear age effects yielded little extra information.
Because fixed effects meta-regression can yield false positive results if random effects are ignored, we further performed random effects meta-regressions as a robustness check. The random effects meta-regression did not substantially change the parameter estimates nor the conclusions drawn based on the meta-regression (see Supplemental Note 1). Further sensitivity analyses showed that the model selection and model parameters were robust to considerable misspecification of the error covariance matrix (see Supplementary Note 1).
Continuing with meta-regression model 3, we tested the degree of polygenicity of the association between schizophrenia and childhood psychopathology by varying the covariate values for prior and prior2 while keeping the other covariates fixed at their inverse variance weighted means. The prediction accuracy as a function of prior peaked between the prior values of .50 and 1, suggesting the optimal prior can be found in this range (Figure S1). This result suggests that the relationship between childhood psychopathology and schizophrenia is highly polygenic in nature, i.e., a large portion of the genome is involved in the relationship between schizophrenia and childhood psychopathology. The forest plot (Figure 1), which contains both the empirical and model predicted estimates for the PRS predictions (for PRS prior = 0.50), reveals that the meta-regression predictions were close to the observed PRS regression coefficients.
Figure 2 shows, based on model 3, the associations between schizophrenia PRS and childhood psychopathology as a function of age, age x disorder, while keeping all other predictors (i.e, cohort, prior and prior2) fixed at their respective inverse variance weighed mean value. The associations increased with age, confirming our hypothesis that the genetic relationship between schizophrenia and developmental psychopathology is stronger in adolescence than in early childhood. Post-hoc inspection of the parameters obtained from 3 (Table 2) further indicate that the association with schizophrenia at age 7 was highest and significant for depression (0.0262, Z= 2.227, p < 0.03). The effect sizes were lower for ODD/CD compared to depression (Z=-2.49., p < 0.02). The predictions for ADHD (Z = -1.61, P < 0.11) and anxiety (Z = - .38, p = .70) did not differ with depression. The increase in association with schizophrenia with age was significant for depression (Z = 2.93, p < 0.003) and even stronger in ADHD (Z = 4.18, p < 0.001) and ODD/CD (Z = 2.17, p < 0.03) compared to depression. The increase was of similar magnitude for anxiety compared to depression (Z = 0.97, p = .30).
Finally, based on the relationship between the outcomes of PRS analyses and genetic correlations as described by Dudbridge, ADDIN REFMGR.CITE Dudbridge201387Power and predictive accuracy of polygenic risk scoresJournal87Power and predictive accuracy of polygenic risk scoresDudbridge,Frank2013Not in Filee1003348PLoS Genet93Public Library of Science1553-7404PLoS Genet144 we computed the expected genetic correlation between developmental psychopathology and schizophrenia as a function of age and split over disorders based on the betas obtained in the meta-regression (for details see: Supplementary Note 1). We assumed that 15% of the variance in childhood psychopathology is captured by the genetic markers used to compute the scores ADDIN REFMGR.CITE Pappa201588Single nucleotide polymorphism heritability of behavior problems in childhood: Genome-wide complex trait analysisJournal88Single nucleotide polymorphism heritability of behavior problems in childhood: Genome-wide complex trait analysisPappa,IreneFedko,Iryna O.Mileva-Seitz,Viara R.Hottenga,Jouke JanBakermans-Kranenburg,Marian J.Bartels,Meikevan Beijsterveldt,Catharina EMJaddoe,Vincent WVMiddeldorp,Christel M.Rippe,Ralph CA2015Not in File737744Journal of the American Academy of Child & Adolescent Psychiatry549Elsevier0890-8567Journal of the American Academy of Child & Adolescent Psychiatry1Benke201418A genome-wide association meta-analysis of preschool internalizing problemsJournal18A genome-wide association meta-analysis of preschool internalizing problemsBenke,Kelly S.Nivard,Michel G.Velders,Fleur P.Walters,Raymond K.Pappa,IreneScheet,Paul A.Xiao,XiangjunEhli,Erik A.Palmer,Lyle J.Whitehouse,Andrew JO2014Not in File667676Journal of the American Academy of Child & Adolescent Psychiatry536Elsevier0890-8567Journal of the American Academy of Child & Adolescent Psychiatry1Trzaskowski201392First genome-wide association study on anxiety-related behaviours in childhoodJournal92First genome-wide association study on anxiety-related behaviours in childhoodTrzaskowski,MaciejEley,Thalia C.Davis,Oliver SPDoherty,Sophia J.Hanscombe,Ken B.Meaburn,Emma L.Haworth,Claire MAPrice,ThomasPlomin,Robert2013Not in Filee58676PloS one84Public Library of Science1932-6203PloS one123, 45, 46, that 35% of variance in the schizophrenia liability is explained by the markers included in the score, ADDIN REFMGR.CITE Golan201490Measuring missing heritability: Inferring the contribution of common variantsJournal90Measuring missing heritability: Inferring the contribution of common variantsGolan,DavidLander,Eric S.Rosset,Saharon2014Not in FileE5272E5281Proceedings of the National Academy of Sciences11149National Acad Sciences0027-8424Proceedings of the National Academy of Sciences1Cross-Disorder Group of the Psychiatric Genomics Consortium201314Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsJournal14Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsCross-Disorder Group of the Psychiatric Genomics Consortium2013Not in File984994Nature genetics459Nature Publishing Group1061-4036Nature genetics124, 47 and that 200,000 independent genetic effects are captured by the markers included in the PRS. Given these assumptions genetic correlations increased from around 0.10 at age 7 and around 0.25 at age 16, differences in genetic correlations with schizophrenia between disorders were modest (Figure 3). The influence of assuming lower (10%) or higher (20%) true variance explained by all measured genetic markers in childhood psychopathology on the estimated genetic correlation is quantified in Figures S2-S3.
Discussion
We investigated whether associations between schizophrenia PRS and childhood psychopathology were explained by shared genetic risk factors. Our meta-analysis revealed a significant association between genetic schizophrenia risk and childhood psychopathology.. Further meta-regression analysis revealed that the genetic overlap between schizophrenia PRS and childhood psychopathology became stronger with age. The associations differed between disorders, with a weaker association for ODD/CD at age 7, and a stronger age related increase for ADHD and ODD/CD. We further found evidence for a high degree of polygenicity in the relationship between schizophrenia PRS and childhood psychopathology, as was evident from the increase in effect with an increase in the prior used in computing the polygenic risk scores.
The variance explained by the polygenic risk score was small. This was expected since the predictive accuracy of a polygenic risk score is not only a function of the genetic correlation between the trait investigated in the discovery GWAS and the target sample, but also is a function of the sample size in the discovery and the number of genetic markers studied. ADDIN REFMGR.CITE Dudbridge201387Power and predictive accuracy of polygenic risk scoresJournal87Power and predictive accuracy of polygenic risk scoresDudbridge,Frank2013Not in Filee1003348PLoS Genet93Public Library of Science1553-7404PLoS Genet144 This is illustrated by other studies analyzing PRS who reported similarly small effect sizes, i.e., explained variance ranging from 0.001 to 0.03.30 More interpretable units of the magnitude of the associations are the genetic correlations derived from the betas. These ranged from 0.1 to 0.25, depending on the age of measurement and the assumptions in the formulae. These correlations indicate a modest but robust genetic association between schizophrenia and childhood psychopathology. This is an important finding in the search for factors influencing the persistence of symptoms from childhood into adulthood and the development into severe mental illness. The genetic correlations between schizophrenia and childhood psychopathology seem to be lower than the correlations between adult psychiatric disorders (MDD-SCZ r = .43, ADDIN REFMGR.CITE Cross-Disorder Group of the Psychiatric Genomics Consortium201314Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsJournal14Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsCross-Disorder Group of the Psychiatric Genomics Consortium2013Not in File984994Nature genetics459Nature Publishing Group1061-4036Nature genetics124 r = .51, ADDIN REFMGR.CITE Bulik-Sullivan2015102An atlas of genetic correlations across human diseases and traitsJournal102An atlas of genetic correlations across human diseases and traitsBulik-Sullivan,BrendanFinucane,Hilary K.Anttila,VerneriGusev,AlexanderDay,Felix R.Loh,Po RuDuncan,LaramiePerry,John RBPatterson,NickRobinson,Elise B.2015Not in FileNature geneticsNature Publishing Group1061-4036Nature genetics148 bipolar disorder-SCZ; r = .68 ADDIN REFMGR.CITE Cross-Disorder Group of the Psychiatric Genomics Consortium201314Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsJournal14Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsCross-Disorder Group of the Psychiatric Genomics Consortium2013Not in File984994Nature genetics459Nature Publishing Group1061-4036Nature genetics124). This is probably not surprising given that, around 50% of the children with psychiatric disorders are disorder free in adulthood ADDIN REFMGR.CITE Copeland2015101Adult functional outcomes of common childhood psychiatric problems: a prospective, longitudinal studyJournal101Adult functional outcomes of common childhood psychiatric problems: a prospective, longitudinal studyCopeland,William E.Wolke,DieterShanahan,LillyCostello,E Jane2015Not in File892899JAMA psychiatry729American Medical Association2168-622XJAMA psychiatry149 so the association with schizophrenia to begin with was already smaller than for adult disorders.
Strengths of our study were the substantial sample sizes of the discovery and target samples. The schizophrenia PRS were based on a large discovery set, a GWAS which revealed 108 genome wide significant loci ADDIN REFMGR.CITE Schizophrenia Working Group of the Psychiatric Genomics Consortium20145Biological insights from 108 schizophrenia-associated genetic lociJournal5Biological insights from 108 schizophrenia-associated genetic lociSchizophrenia Working Group of the Psychiatric Genomics Consortium2014Not in File421427Nature5117510Nature Publishing Group0028-0836Nature129, The target samples varied between 5,354 and 8,253 at different ages, which is substantially higher than the required number of ~2,000 subjects generally indicated as sufficient for PRS analysis. ADDIN REFMGR.CITE Wray201453Research Review: Polygenic methods and their application to psychiatric traitsJournal53Research Review: Polygenic methods and their application to psychiatric traitsWray,Naomi R.Lee,Sang HongMehta,DivyaVinkhuyzen,Anna AEDudbridge,FrankMiddeldorp,Christel M.2014Not in File10681087Journal of child psychology and psychiatry5510Wiley Online Library1469-7610Journal of child psychology and psychiatry130 Innovative strengths of our analyses were the explicit modeling of all univariate analysis results in the meta-regression approach which accounted for the covariance between disorders and ages, correction for cohort specific effects and the simultaneous consideration of multiple risk scores trained at different priors.
We note that the heterogeneity in assessments that also required slightly different regression methods in the NTR and ALSPAC may have biased the results downwards, beyond the point where the effect can be mitigated by transformations and covariates. Heterogeneity in phenotypes is often mentioned in genetic studies as a reason for few or no significant findings and, although both measures used in this study were based on ratings that are consistently related to clinical DSM-IV diagnoses, ADDIN REFMGR.CITE Ebesutani201064Concurrent validity of the Child Behavior Checklist DSM-oriented scales: Correspondence with DSM diagnoses and comparison to syndrome scalesJournal64Concurrent validity of the Child Behavior Checklist DSM-oriented scales: Correspondence with DSM diagnoses and comparison to syndrome scalesEbesutani,ChadBernstein,AdamNakamura,Brad J.Chorpita,Bruce F.Higa-McMillan,Charmaine K.Weisz,John R.Research Network on Youth Mental Health2010Not in File373384Journal of Psychopathology and Behavioral Assessment323Springer0882-2689Journal of Psychopathology and Behavioral Assessment1Bellina201363The ability of CBCL DSM-oriented scales to predict DSM-IV diagnoses in a referred sample of children and adolescentsJournal63The ability of CBCL DSM-oriented scales to predict DSM-IV diagnoses in a referred sample of children and adolescentsBellina,MonicaBrambilla,PaoloGarzitto,MarcoNegri,Gioia ALMolteni,MassimoNobile,Maria2013Not in File235246European child & adolescent psychiatry224Springer1018-8827European child & adolescent psychiatry1Goodman200033The Development and Well-Being Assessment: description and initial validation of an integrated assessment of child and adolescent psychopathologyJournal33The Development and Well-Being Assessment: description and initial validation of an integrated assessment of child and adolescent psychopathologyGoodman,RobertFord,TamsinRichards,HilaryGatward,RebeccaMeltzer,Howard2000Not in File645655Journal of child psychology and psychiatry4105Cambridge Univ Press1469-7610Journal of child psychology and psychiatry136, 37, 50 combining these measures has probably increased heterogeneity. It follows that it would be preferable for cohorts to use the same measurement instruments. This does not withstand the fact that to get to large enough sample sizes, it is preferable to combine the data collected in different cohorts. This is supported by studies that show that different measures of a phenotype are associated with the same risk factors. Genetic factors for clinically diagnosed ADHD, for example, overlap with genetic factors influencing continuous ADHD measures assessed in the general population. ADDIN REFMGR.CITE Groen-Blokhuis201442Attention-deficit/hyperactivity disorder polygenic risk scores predict attention problems in a population-based sample of childrenJournal42Attention-deficit/hyperactivity disorder polygenic risk scores predict attention problems in a population-based sample of childrenGroen-Blokhuis,Maria M.Middeldorp,Christel M.Kan,Kees JanAbdellaoui,Abdelvan Beijsterveldt,Catharina EMEhli,Erik A.Davies,Gareth E.Scheet,Paul A.Xiao,XiangjunHudziak,James J.2014Not in File11231129Journal of the American Academy of Child & Adolescent Psychiatry5310Elsevier0890-8567Journal of the American Academy of Child & Adolescent Psychiatry1Martin201494Genetic risk for attention-deficit/hyperactivity disorder contributes to neurodevelopmental traits in the general populationJournal94Genetic risk for attention-deficit/hyperactivity disorder contributes to neurodevelopmental traits in the general populationMartin,JoannaHamshere,Marian L.Stergiakouli,EvangeliaOGÇÖDonovan,Michael C.Thapar,Anita2014Not in File664671Biological psychiatry768Elsevier0006-3223Biological psychiatry1Stergiakouli201595Shared genetic influences between attention-deficit/hyperactivity disorder (ADHD) traits in children and clinical ADHDJournal95Shared genetic influences between attention-deficit/hyperactivity disorder (ADHD) traits in children and clinical ADHDStergiakouli,EvieMartin,JoannaHamshere,Marian L.Langley,KateEvans,David M.St Pourcain,BeateTimpson,Nicholas J.Owen,Michael J.O'Donovan,MichaelThapar,Anita2015Not in File322327Journal of the American Academy of Child & Adolescent Psychiatry544Elsevier0890-8567Journal of the American Academy of Child & Adolescent Psychiatry151-53 These results suggest that combining several measures is an appropriate way to increase sample size and thus statistical power. Moreover, in the current study, we mitigated the effects of heterogeneity induced by the different instruments by performing a meta-regression analyses which provided the opportunity to test and control for between cohort mean differences in the effect of PRS on outcome. In this way, the current study strikes the optimal balance between the risk of bias and maximizing statistical power.
Another limitation that concerns longitudinal studies is the dropout over the years. We analyzed whether the schizophrenia PRS predicted non-participation and observed significant associations between schizophrenia PRS and non-participation at age 15 in both cohorts and at all ages in ALSPAC (Supplementary Table 3). We further observed that non-participation was related to a higher score on psychopathology scales at an earlier age in ALSPAC, especially for ODD/CD and ADHD at age 13 and 15. In NTR non-participation at age 15 was related to ODD/CD and depression at age 12 (Supplementary Table 4). As those with higher PRS and higher psychopathology scores at earlier ages are more likely to drop out, we expect that the dropout introduces downward bias in the estimated relationship between schizophrenia and childhood psychopathology. Thus, the magnitude of genetic associations may be underestimated. Note that only longitudinal analyses can provide insight into the influence of dropout on the estimate genetic relationship between traits, while in univariate studies a failure to participate results in the absence of genetic data and thus the influence of failure to participate cannot be quantified. For more comprehensive genetically informed dropout analysis of the ALSPAC data see: ADDIN REFMGR.CITE Martin201697Association of Genetic Risk for Schizophrenia With Nonparticipation Over Time in a Population-Based Cohort StudyJournal97Association of Genetic Risk for Schizophrenia With Nonparticipation Over Time in a Population-Based Cohort StudyMartin,JoannaTilling,KateHubbard,LeonStergiakouli,EvieThapar,AnitaSmith,George DaveyO'Donovan,MichaelZammit,S.2016Not in FileAOPAmerican Journal of EpidemiologyAmerican Journal of Epidemiology154.
Results from previous research focusing on the genetic overlap between schizophrenia and (childhood) psychopathology are largely in line with ours. ADDIN REFMGR.CITE Cross-Disorder Group of the Psychiatric Genomics Consortium201314Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsJournal14Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsCross-Disorder Group of the Psychiatric Genomics Consortium2013Not in File984994Nature genetics459Nature Publishing Group1061-4036Nature genetics1Cross-Disorder Group of the Psychiatric Genomics Consortium201346Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysisJournal46Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysisCross-Disorder Group of the Psychiatric Genomics Consortium2013Not in File13711379The Lancet3819875Elsevier0140-6736The Lancet1Krapohl201569Phenome-wide analysis of genome-wide polygenic scoresJournal69Phenome-wide analysis of genome-wide polygenic scoresKrapohl,E.Euesden,J.Zabaneh,D.Pingault,J.B.Rimfeld,K.von Stumm,S.Dale,P.S.Breen,G.O'Reilly,P.F.Plomin,R.2015Not in FileMolecular psychiatryAOPNature Publishing Group1359-4184Molecular psychiatry1Jones201682Phenotypic manifestation of genetic risk for schizophrenia
during adolescence in the general population.Journal82Phenotypic manifestation of genetic risk for schizophrenia
during adolescence in the general population.Jones,H.J.Stergiakouli,E.Tansey,K.E.Hubbard,L.Heron,J.Cannon,M.Holmans,P.Lewis,G.Linden,D.E.J.Jones,P.B.Smith,G.D.O'Donnovan,M.Owen,M.J.Walter,J.T.Zammit,S.2016Not in FileJAMA Psychiatry.733JAMA Psychiatry.1Bulik-Sullivan2015102An atlas of genetic correlations across human diseases and traitsJournal102An atlas of genetic correlations across human diseases and traitsBulik-Sullivan,BrendanFinucane,Hilary K.Anttila,VerneriGusev,AlexanderDay,Felix R.Loh,Po RuDuncan,LaramiePerry,John RBPatterson,NickRobinson,Elise B.2015Not in FileNature geneticsNature Publishing Group1061-4036Nature genetics1Hamshere201381Shared polygenic contribution between childhood attention-deficit hyperactivity disorder and adult schizophreniaJournal81Shared polygenic contribution between childhood attention-deficit hyperactivity disorder and adult schizophreniaHamshere,Marian L.Stergiakouli,EvangeliaLangley,KateMartin,JoannaHolmans,PeterKent,LindseyOwen,Michael J.Gill,MichaelThapar,AnitaO Donovan,Mick2013Not in File107111The British Journal of Psychiatry2032RCP0007-1250The British Journal of Psychiatry124-28, 48 Three differences are noteworthy. Another study in the ALSPAC sample focused on psychiatric symptoms at age 15 and found that schizophrenia PRS predicted anxiety disorder and negative symptoms, but not depressive disorder and psychotic experiences. The difference with the current results for depression may be explained by improvements in the method used to compute the polygenic scores and in the definition of the phenotype. ADDIN REFMGR.CITE Jones201682Phenotypic manifestation of genetic risk for schizophrenia
during adolescence in the general population.Journal82Phenotypic manifestation of genetic risk for schizophrenia
during adolescence in the general population.Jones,H.J.Stergiakouli,E.Tansey,K.E.Hubbard,L.Heron,J.Cannon,M.Holmans,P.Lewis,G.Linden,D.E.J.Jones,P.B.Smith,G.D.O'Donnovan,M.Owen,M.J.Walter,J.T.Zammit,S.2016Not in FileJAMA Psychiatry.733JAMA Psychiatry.127 Two studies by the Psychiatric Genomics Consortium Cross Disorder Group detected strong correlations between major depressive disorder, bipolar disorder and schizophrenia, but no genetic correlation between ADHD and schizophrenia. ADDIN REFMGR.CITE Cross-Disorder Group of the Psychiatric Genomics Consortium201314Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsJournal14Genetic relationship between five psychiatric disorders estimated from genome-wide SNPsCross-Disorder Group of the Psychiatric Genomics Consortium2013Not in File984994Nature genetics459Nature Publishing Group1061-4036Nature genetics1Cross-Disorder Group of the Psychiatric Genomics Consortium201346Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysisJournal46Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysisCross-Disorder Group of the Psychiatric Genomics Consortium2013Not in File13711379The Lancet3819875Elsevier0140-6736The Lancet124, 25 The latter is probably explained by the smaller sample size for ADHD at the time, as is confirmed by recent analyses which detected a significant association between ADHD and schizophrenia. ADDIN REFMGR.CITE Anttila2016103Analysis of shared heritability in common disorders of the brainJournal103Analysis of shared heritability in common disorders of the brainAnttila,VerneriBulik-Sullivan,BrendanFinucane,Hilary KiyoBras,JoseDuncan,LaramieEscott-Price,ValentinaFalcone,GuidoGormley,PadhraigMalik,RainerPatsopoulos,NikolaosRipke,StephanWalters,RaymondWei,ZhiYu,DongmeiLee,Phil,,,,,,,,,,,,,Breen,GeromeBulik,CynthiaDaly,MarkDichgans,MartinFaraone,StephenGuerreiro,RitaHolmans,PeterKendler,KennethKoeleman,BobbyMathews,CarolScharf,JeremiahSklar,PamelaWilliams,JulieWood,NickCotsapas,ChrisPalotie,AarnoSmoller,JordanSullivan,PatrickRosand,JonathanCorvin,AidenNeale,Benjamin2016/4/16Not in FilebioRxiv10.1101/048991http://biorxiv.org/content/early/2016/04/16/048991.abstractbioRxiv155 Finally, a study analyzing the association between schizophrenia and attention problems and impulsivity, anxiety and a general tendency for psychopathology yielded no significant associations. ADDIN REFMGR.CITE Krapohl201569Phenome-wide analysis of genome-wide polygenic scoresJournal69Phenome-wide analysis of genome-wide polygenic scoresKrapohl,E.Euesden,J.Zabaneh,D.Pingault,J.B.Rimfeld,K.von Stumm,S.Dale,P.S.Breen,G.O'Reilly,P.F.Plomin,R.2015Not in FileMolecular psychiatryAOPNature Publishing Group1359-4184Molecular psychiatry128 Yet, the direction of the effects was positive (i.e., higher schizophrenia risk predicted higher scores). As that study comprised polygenic risk scores for 13 traits as well as 50 outcome variables, the multiple-testing burden was considerably higher than in the current study resulting in lower power to detect an effect. Overall, the evidence so far suggests the presence of a positive genetic correlation between schizophrenia and psychopathology in childhood.
These previous studies24-28 also analyzed multiple outcomes, but our study provides additional insights into differences of genetic effects across diagnostic boundaries and over ages by adopting a longitudinal and multivariate approach. Our findings suggest that there are sets of SNPs broadly influencing psychopathology across ages in the general population, and that there are sets of SNPs of which the effect is either limited to or increases in puberty. This signifies that age-sensitive genome-wide meta-analysis of repeated measures, in either case-control or population based samples could well identify genetic variants. Some of these genetic variants will increase an individuals vulnerability for psychopathology and may be associated with persistence of symptoms from childhood into adolescence and adulthood, while other variants can be identified that have an age or disorder dependent effect on psychopathology. Identifying not only which variants influence psychopathology but also at what age can aid to focus translational studies on developmental processes.
To conclude, our study shows how genetic risk factors for schizophrenia are of increasing importance during childhood and adolescence and demonstrate the value of longitudinal studies across diagnostic boundaries to increase our insight into the etiology of severe psychiatric disorders.
Acknowledgments.
We are extremely grateful to all the families who took part in this study, 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. The UK Medical Research Council and the Wellcome Trust (Grant ref: 102215/2/13/2) and the University of Bristol provide core support for ALSPAC. This work was supported in part by the Medical Research Council Integrative Epidemiology Unit at the University of Bristol (MC_UU_12013/6). This publication is the work of the authors and Michel G. Nivard will serve as guarantor for the contents of this paper. ALSPAC GWAS data was generated by Sample Logistics and Genotyping Facilities at the Wellcome Trust Sanger Institute and LabCorp (Laboratory Corporation of America) using support from 23andMe. We are equally grateful to the NTR twins, their parents and other family members who participate in the NTR research. The NTR gratefully acknowledges support from Genetic influences on stability and change in psychopathology from childhood to young adulthood (ZonMW 912-10-020) and Genetics of Mental Illness (ERC-230374). MGN is supported by Royal Netherlands Academy of Science Professor Award (PAH/6635). The Netherlands Organization for Scientific Research (NWO) and MagW/ZonMW grants Middelgroot-911-09-032, Center for Medical Systems Biology (CSMB, NWO Genomics), NBIC/BioAssist/RK(2008.024), Biobanking and Biomolecular Resources Research Infrastructure (BBMRINL, 184.021.007), VU Universitys Institute for Health and Care Research (EMGO+) and Neuroscience Campus Amsterdam (NCA). Part of the genotyping and analyses were funded by the Genetic Association Information Network (GAIN) of the Foundation for the National Institutes of Health, Rutgers University Cell and DNA Repository (NIMH U24 MH068457-06), the Avera Institute, Sioux Falls, South Dakota (USA) and the National Institutes of Health (NIH R01 HD042157-01A1, MH081802, Grand Opportunity grants 1RC2 MH089951 and 1RC2 MH089995).
Conflicts of interest
The authors declare no conflict of interest.
ADDIN REFMGR.REFLIST Reference List
(1) Kessler RC, Amminger GP, Aguilar-Gaxiola S, Alonso J, Lee S, Ustun TB. Age of onset of mental disorders: a review of recent literature. Current opinion in psychiatry 2007;20(4):359.
(2) Heinrichs RW, Zakzanis KK. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology 1998;12(3):426.
(3) Kahn RS, Sommer IE. The neurobiology and treatment of first-episode schizophrenia. Molecular psychiatry 2014.
(4) Fusar-Poli P, Bechdolf A, Taylor MJ et al. At risk for schizophrenic or affective psychoses? A meta-analysis of DSM/ICD diagnostic outcomes in individuals at high clinical risk. Schizophrenia bulletin 2013;39(4):923-32.
(5) Woodberry KA, Giuliano AJ, Seidman LJ. Premorbid IQ in schizophrenia: a meta-analytic review. The American journal of psychiatry 2008;165(5):579-87.
(6) Cunningham Owens DG, Johnstone EC. Precursors and prodromata of schizophrenia: findings from the Edinburgh High Risk Study and their literature context. Psychological medicine 2006;36(11):1501-14.
(7) Cannon M, Caspi A, Moffitt TE et al. Evidence for early-childhood, pan-developmental impairment specific to schizophreniform disorder: results from a longitudinal birth cohort. Archives of general psychiatry 2002;59(5):449-56.
(8) Kim-Cohen J, Caspi A, Moffitt TE, Harrington H, Milne BJ, Poulton R. Prior juvenile diagnoses in adults with mental disorder: developmental follow-back of a prospective-longitudinal cohort. Archives of general psychiatry 2003;60(7):709-17.
(9) Rossi A, Pollice R, Daneluzzo E, Marinangeli MG, Stratta P. Behavioral neurodevelopment abnormalities and schizophrenic disorder: a retrospective evaluation with the Childhood Behavior Checklist (CBCL). Schizophrenia research 2000;44(2):121-8.
(10) Miller PM, Byrne M, Hodges A, Lawrie SM, Johnstone EC. Childhood behaviour, psychotic symptoms and psychosis onset in young people at high risk of schizophrenia: early findings from the Edinburgh High Risk Study. Psychological medicine 2002;32(01):173-9.
(11) Muratori F, Salvadori F, D-Arcangelo G, Viglione V, Picchi L. Childhood psychopathological antecedents in early onset schizophrenia. European psychiatry 2005;20(4):309-14.
(12) Keshavan MS, Diwadkar VA, Montrose DM, Rajarethinam R, Sweeney JA. Premorbid indicators and risk for schizophrenia: a selective review and update. Schizophrenia research 2005;79(1):45-57.
(13) Meyer SE, Bearden CE, Lux SR et al. The psychosis prodrome in adolescent patients viewed through the lens of DSM-IV. Journal of Child & Adolescent Psychopharmacology 2005;15(3):434-51.
(14) Fusar-Poli P, Nelson B, Valmaggia L, Yung AR, McGuire PK. Comorbid depressive and anxiety disorders in 509 individuals with an at-risk mental state: impact on psychopathology and transition to psychosis. Schizophrenia bulletin 2014;40(1):120-31.
(15) Gajwani R, Patterson P, Birchwood M. Attachment: Developmental pathways to affective dysregulation in young people at ultra-high risk of developing psychosis. British Journal of Clinical Psychology 2013;52(4):424-37.
(16) Maibing CF, Pedersen CBc, Benros ME, Mortensen PB, Dalsgaard S+, Nordentoft M. Risk of schizophrenia increases after all child and adolescent psychiatric disorders: a nationwide study. Schizophrenia bulletin 2014;sbu119.
(17) Perkins DO, Gu H, Boteva K, Lieberman JA. Relationship between duration of untreated psychosis and outcome in first-episode schizophrenia: a critical review and meta-analysis. American Journal of Psychiatry 2005;162(10):1785-804.
(18) Stafford MR, Jackson H, Mayo-Wilson E, Morrison AP, Kendall T. Early interventions to prevent psychosis: systematic review and meta-analysis. Bmj 2013;346:f185.
(19) Sullivan PF, Kendler KS, Neale MC. Schizophrenia as a complex trait: evidence from a meta-analysis of twin studies. Archives of general psychiatry 2003;60(12):1187-92.
(20) Sanchez-Gistau V, Romero S, Moreno D et al. Psychiatric disorders in child and adolescent offspring of patients with schizophrenia and bipolar disorder: A controlled study. Schizophrenia research 2015;168(1):197-203.
(21) Nivard MG, Dolan CV, Kendler KS et al. Stability in symptoms of anxiety and depression as a function of genotype and environment: a longitudinal twin study from ages 3 to 63 years. Psychological medicine 2014;1-11.
(22) Kan KJ, Dolan CV, Nivard MG et al. Genetic and environmental stability in attention problems across the lifespan: evidence from the Netherlands twin register. Journal of the American Academy of Child & Adolescent Psychiatry 2013;52(1):12-25.
(23) Benke KS, Nivard MG, Velders FP et al. A genome-wide association meta-analysis of preschool internalizing problems. Journal of the American Academy of Child & Adolescent Psychiatry 2014;53(6):667-76.
(24) Cross-Disorder Group of the Psychiatric Genomics Consortium. Genetic relationship between five psychiatric disorders estimated from genome-wide SNPs. Nature genetics 2013;45(9):984-94.
(25) Cross-Disorder Group of the Psychiatric Genomics Consortium. Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysis. The Lancet 2013;381(9875):1371-9.
(26) Hamshere ML, Stergiakouli E, Langley K et al. Shared polygenic contribution between childhood attention-deficit hyperactivity disorder and adult schizophrenia. The British Journal of Psychiatry 2013;203(2):107-11.
(27) Jones HJ, Stergiakouli E, Tansey KE et al. Phenotypic manifestation of genetic risk for schizophreniaduring adolescence in the general population. JAMA Psychiatry 2016;73(3).
(28) Krapohl E, Euesden J, Zabaneh D et al. Phenome-wide analysis of genome-wide polygenic scores. Molecular psychiatry 2015;(AOP).
(29) Schizophrenia Working Group of the Psychiatric Genomics Consortium. Biological insights from 108 schizophrenia-associated genetic loci. Nature 2014;511(7510):421-7.
(30) Wray NR, Lee SH, Mehta D, Vinkhuyzen AA, Dudbridge F, Middeldorp CM. Research Review: Polygenic methods and their application to psychiatric traits. Journal of child psychology and psychiatry 2014;55(10):1068-87.
(31) Purcell SM, Wray NR, Stone JL et al. Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature 2009;460(7256):748-52.
(32) Wray NR, Yang J, Hayes BJ, Price AL, Goddard ME, Visscher PM. Pitfalls of predicting complex traits from SNPs. Nature Reviews Genetics 2013;14(7):507-15.
(33) American Psychiatric Association. Diagnostic and statistical manual of mental diseases. DSM-IV 4th edn Washington (DC): American Psychiatric Association 1994.
(34) van Beijsterveldt CE, Groen-Blokhuis M, Hottenga JJ et al. The Young Netherlands Twin Register (YNTR): longitudinal twin and family studies in over 70,000 children. Twin Research and Human Genetics 2013;16(01):252-67.
(35) 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. International journal of epidemiology 2012;dys064.
(36) Ebesutani C, Bernstein A, Nakamura BJ et al. Concurrent validity of the Child Behavior Checklist DSM-oriented scales: Correspondence with DSM diagnoses and comparison to syndrome scales. Journal of Psychopathology and Behavioral Assessment 2010;32(3):373-84.
(37) Goodman R, Ford T, Richards H, Gatward R, Meltzer H. The Development and Well-Being Assessment: description and initial validation of an integrated assessment of child and adolescent psychopathology. Journal of child psychology and psychiatry 2000;41(05):645-55.
(38) Goodman A, Heiervang E, Collishaw S, Goodman R. The 'DAWBA bands` as an ordered-categorical measure of child mental health: description and validation in British and Norwegian samples. Social psychiatry and psychiatric epidemiology 2011;46(6):521-32.
(39) Vilhjalmsson B, Yang J, Finucane HK et al. Modeling Linkage Disequilibrium Increases Accuracy of Polygenic Risk Scores. bioRxiv 2015;015859.
(40) Minic CC, Dolan CV, Kampert MM, Boomsma DI, Vink JM. Sandwich corrected standard errors in family-based genome-wide association studies. European Journal of Human Genetics 2014.
(41) Viechtbauer W. Conducting meta-analyses in R with the metafor package. Journal of Statistical Software 2010;36(3):1-48.
(42) Higgins J, Thompson SG. Controlling the risk of spurious findings from meta-regression. Statistics in medicine 2004;23(11):1663-82.
(43) Loh PR, Bhatia G, Gusev A et al. Contrasting genetic architectures of schizophrenia and other complex diseases using fast variance-components analysis. Nature genetics 2015.
(44) Dudbridge F. Power and predictive accuracy of polygenic risk scores. PLoS Genet 2013;9(3):e1003348.
(45) Pappa I, Fedko IO, Mileva-Seitz VR et al. Single nucleotide polymorphism heritability of behavior problems in childhood: Genome-wide complex trait analysis. Journal of the American Academy of Child & Adolescent Psychiatry 2015;54(9):737-44.
(46) Trzaskowski M, Eley TC, Davis OS et al. First genome-wide association study on anxiety-related behaviours in childhood. PloS one 2013;8(4):e58676.
(47) Golan D, Lander ES, Rosset S. Measuring missing heritability: Inferring the contribution of common variants. Proceedings of the National Academy of Sciences 2014;111(49):E5272-E5281.
(48) Bulik-Sullivan B, Finucane HK, Anttila V et al. An atlas of genetic correlations across human diseases and traits. Nature genetics 2015.
(49) Copeland WE, Wolke D, Shanahan L, Costello EJ. Adult functional outcomes of common childhood psychiatric problems: a prospective, longitudinal study. JAMA psychiatry 2015;72(9):892-9.
(50) Bellina M, Brambilla P, Garzitto M, Negri GA, Molteni M, Nobile M. The ability of CBCL DSM-oriented scales to predict DSM-IV diagnoses in a referred sample of children and adolescents. European child & adolescent psychiatry 2013;22(4):235-46.
(51) Groen-Blokhuis MM, Middeldorp CM, Kan KJ et al. Attention-deficit/hyperactivity disorder polygenic risk scores predict attention problems in a population-based sample of children. Journal of the American Academy of Child & Adolescent Psychiatry 2014;53(10):1123-9.
(52) Martin J, Hamshere ML, Stergiakouli E, OGDonovan MC, Thapar A. Genetic risk for attention-deficit/hyperactivity disorder contributes to neurodevelopmental traits in the general population. Biological psychiatry 2014;76(8):664-71.
(53) Stergiakouli E, Martin J, Hamshere ML et al. Shared genetic influences between attention-deficit/hyperactivity disorder (ADHD) traits in children and clinical ADHD. Journal of the American Academy of Child & Adolescent Psychiatry 2015;54(4):322-7.
(54) Martin J, Tilling K, Hubbard L et al. Association of Genetic Risk for Schizophrenia With Nonparticipation Over Time in a Population-Based Cohort Study. American Journal of Epidemiology 2016;AOP.
(55) Anttila V, Bulik-Sullivan B, Finucane HK et al. Analysis of shared heritability in common disorders of the brain. bioRxiv 2016 April 16.
Figure legends
Figure 1: A forest plot of the observed associations between schizophrenia PRS (obtained at prior proportion of causal SNPs = 0.50) and of the model predicted associations. The blue polygons indicate the association as predicted from the meta-regression model, while the black square indicates the association as observed in the empirical data. The whiskers indicate the 95% confidence regions around the empirical PRS associations. The results are ordered by increasing age for each disorder, with in the top halve the results in the ALSPAC cohort and in the bottom halve the results for the NTR cohort.
Figure 2: Bubble plot showing the effect of age on the association between schizophrenia PRS and childhood psychopathology, split per disorder. Circles indicate the observed effect sizes in the univariate regression analyses (ALSPAC in red, NTR in blue). The size of the circles is proportional to the inverse of the variance, and thus larger circles reflect more accurate estimates. The solid line reflects the meta-regression fitted effect size and the dashed lines indicate the upper and lower 95% confidence interval around the meta-regression line.
Figure 3: Bubble plot of the approximated genetic correlations between schizophrenia and childhood psychopathology per disorder. We assume the variance in childhood psychopathology explained by all markers used to construct the PRS is 15% and constant over disorders and age. We further assume the PRS captures 200.000 independent genetic effects. Circles indicate the transformed observed regression coefficients to genetic correlations (ALSPAC in red, NTR in blue). The size of the circles is proportional to the inverse of the variance, and thus larger circles reflect more accurate estimates. The solid line reflects the genetic correlation and the dashed lines indicate the upper and lower 95% confidence interval around the genetic correlation, quantifying the uncertainty in the meta-regression.
Table legends
Table 1 Model fit criteria for the meta-regression models. This table contains the predictors, likelihood ratio test (LRT), AIC and residual heterogeneity for the 4 meta-regression models considered. The LRT tests the relative performance of adjacent models, i.e, model 2 is tested against model 1, model 3 is tested against model 2, and model 4 is tested against model 3. The test of residual heterogeneity considers whether the residual variance in the effects observed in the original studies, after conditioning on the predictors, still deviates from zero. Parametric resample reports on the percentage of resampled datasets in which each model fitted best according to the AIC.
Table 2: Parameter estimates and test statistics for meta-regression model 3. Depression serves as a reference disorder. Therefore the intercept and age effect reflect the expected association between schizophrenia and depression at age 7 and the yearly increase in the expected association. se: standard error, zval: z-value, pval: p-value, ci.lb: lower bound of 95% confidence interval, ci.ub: upper bound of 95% confidence interval
Figure 1
Figure 2
Figure 3
Supplement
Figure notes and Table titles
Figure S1: Bubble plot of the relationship between the polygenicity prior and the effect size in the polygenic risk score analyses. Note the increase in effect size with the increase in polygenicity prior. The solid line indicates the best fit obtained from the meta-regression model (model 3). The dashed lines reflect the upper and lower confidence bounds.
Figure S2 Bubble plot of the approximated genetic correlations between schizophrenia and childhood psychopathology per disorder given the assumptions described in Supplementary Note 1. In this figure, we assume the variance explained by all markers in childhood psychopathology is constant and 20%. Circles indicate the transformed observed regression coefficients to genetic correlations (ALSPAC in red, NTR in blue). The size of the circles is proportional to the inverse of the variance, and thus larger circles reflect more accurate estimates. The solid line reflects the genetic correlation and the dashed lines indicate the upper and lower 95% confidence interval around the genetic correlation, quantifying the uncertainty in the meta-regression but not in the variance in childhood psychopathology explained by all measured markers, or the estimate of the number of independent markers.
Figure S3 Bubble plot of the approximated genetic correlations between schizophrenia and childhood psychopathology per disorder given the assumptions described in Supplementary Note 1. In this figure, we assume the variance explained by all markers in childhood psychopathology is constant and 10%. Circles indicate the transformed observed regression coefficients to genetic correlations (ALSPAC in red, NTR in blue). The size of the circles is proportional to the inverse of the variance, and thus larger circles reflect more accurate estimates. The solid line reflects the genetic correlation and the dashed lines indicate the upper and lower 95% confidence interval around the genetic correlation, quantifying the uncertainty in the meta-regression but not in the variance in childhood psychopathology explained by all measured markers, or the estimate of the number of independent markers.
Table S1: Sample sizes per age group for the NTR, ALSPAC and combined
Table S2: Descriptives
Table S3: Prediction of non-participation based on the schizophrenia PRS
Table S4: Prediction of non-participation based on psychopathology scores at an earlier time point
Supplementary Note 1
This note accompanies the manuscript entitled: Genetic overlap between schizophrenia and developmental psychopathology: a longitudinal analysis of common childhood disorders between age 7 and 15. All data described here and analyses presented here serve to support the conclusions of the manuscript as published.
Phenotype descriptives
Table S2 presents the mean scores on the DSM-IV based scales of anxiety, depression, ADHD, ODD/CD for males and females in the NTR at different ages (left) as well as the percentages of male and female ALSPAC participants with these diagnoses, defined as a score of 4 or 5 on the DAWBA (right). (Note that in the analyses, the 6-category DAWBA band was used as outcome variable since this is a more informative measure than the dichotomous DAWBA diagnosis).
Genotyping and genotype quality control:
The NTR participants were genotyped on Affymetrix 6.0, Affymetrix-perlegen 5.0, Illumina 660 and Omni express (1M) platforms. Array specific calls and cleaning were performed before data from different platforms were combined. Data from different platforms were strand aligned, SNPs with a minor allele frequency below 1%, a HWE p-value < 1*10-5 and with a genotype missingness rate > 10% or call rate < 95% were removed. Individuals with an excessive or low heterozygosity were removed (F > .10 or F < .10). After QC, genotypes were imputed to a common set of SNPs based on the goNL reference set. ADDIN REFMGR.CITE Genome of the Netherlands Consortium201489Whole-genome sequence variation, population structure and demographic history of the Dutch populationJournal89Whole-genome sequence variation, population structure and demographic history of the Dutch populationGenome of the Netherlands Consortium2014Not in File818825Nature genetics468Nature Publishing Group1061-4036Nature genetics11 SNPs were imputed that were not directly measured on each platform. Samples were excluded when reported gender did not match biological gender or when individuals were of non-European ancestry based on principle component analysis. ADDIN REFMGR.CITE Abdellaoui201321Population structure, migration, and diversifying selection in the NetherlandsJournal21Population structure, migration, and diversifying selection in the NetherlandsAbdellaoui,AbdelHottenga,Jouke Jande Knijff,PeterNivard,Michel G.Xiao,XiangjunScheet,PaulBrooks,AndrewEhli,Erik A.Hu,YueshanDavies,Gareth E.2013Not in File12771285European journal of human genetics2111Nature Publishing Group1018-4813European journal of human genetics12 In the NTR, sex, call rate, F (inbreeding coefficient), five principle components based on global ancestry and five principal components correcting for local ancestry differences within the Netherlands were included as covariates in all analyses . ADDIN REFMGR.CITE Abdellaoui201321Population structure, migration, and diversifying selection in the NetherlandsJournal21Population structure, migration, and diversifying selection in the NetherlandsAbdellaoui,AbdelHottenga,Jouke Jande Knijff,PeterNivard,Michel G.Xiao,XiangjunScheet,PaulBrooks,AndrewEhli,Erik A.Hu,YueshanDavies,Gareth E.2013Not in File12771285European journal of human genetics2111Nature Publishing Group1018-4813European journal of human genetics12
In ALSPAC, children were genotyped on the Illumina HumanHap550 quad chip genotyping platforms. The raw genome-wide data were subjected to standard quality control methods. Individuals were excluded on the basis of gender mismatches, minimal or excessive heterozygosity, disproportionate levels of individual missingness (>3%), and insufficient sample replication (IBD < 0.8). Population stratification was assessed by multidimensional scaling analysis, and compared with Hapmap II (release 22); all individuals of non-European ancestry were removed. SNPs with a minor allele frequency of < 1%, a call rate of < 95%, or evidence for violations of Hardy-Weinberg equilibrium (p < 5E-7) were removed. Cryptic relatedness was measured as proportion of identity by descent (IBD > 0.1). Related subjects that passed all other quality control thresholds were retained during subsequent phasing and imputation, though not association. Imputation of the target data was performed using Impute V2.2.2 ADDIN REFMGR.CITE Howie200977A flexible and accurate genotype imputation method for the next generation of genome-wide association studiesJournal77A flexible and accurate genotype imputation method for the next generation of genome-wide association studiesHowie,Bryan N.Donnelly,PeterMarchini,Jonathan2009Not in Filee1000529PLoS Genet56PLoS Genet13 against the 1000 genomes phase 1 version 3 reference panel, using all 2186 reference haplotypes (including non-Europeans). ADDIN REFMGR.CITE Genomes Project Consortium201275An integrated map of genetic variation from 1,092 human genomesJournal75An integrated map of genetic variation from 1,092 human genomesGenomes Project Consortium2012Not in File5665Nature4917422Nature Publishing Group0028-0836Nature14 As the ALSPAC sample, after QC, is assumed to be genetically homogeneous with respect to ancestry and local ancestry differences, no principal components were added as covariates, sex was included as covariate in all analyses.
The correction for the presence of overlapping subjects at the different ages of measurement, and the correlation between the polygenic predictors.
In the meta-regression analysis performed we had a set of 6 predictors X, and 32 outcomes Y. We performed a series of 192 univariate regressions:
We constructed an approximate error correlation matrix (i.e. the correlation between the regression parameters B) for a series of univariate regressions of p equal to:
We specified the error covariance matrix as: QUOTE . Where se was a 192 x 192 diagonal matrix with the standard associated with each of the parameters B on the diagonal. The errors were assumed to be independent between cohorts and therefore correlations between cohorts were set to zero in matrix QUOTE . Based on the specified error correlation matrix, we performed the meta-analysis and meta-regression of the betas obtained from the univariate regression analyses. To test whether the proposed error correlation matrix accurately accounted for the dependence induced by correlated predictors and outcomes, we performed type 1 error simulations.
We simulated 3 traits (Y) (correlations between .3 and .5) and 3 polygenic scores (X) (correlations between .9 and .8) for 100 subjects. In each simulation there was no true association between PRS and traits. We regressed each trait Y on each polygenic score X, and meta-analyzed the 9 test statistics obtained from these regressions, correcting for the dependence between traits and risk scores as outlined above. Given a small sample in the univariate regressions (N=100) the following slightly liberal type 1 error rates were observed. The liberal type-1 error was likely induced by the fact that the test statistic obtained in each meta-analysis followed a t-distribution and not a normal distribution.
AlphaType 1 error0.100.1230.050.0650.010.015
Simulating data given a larger sample of 1000 subjects in the initial univariate regressions the following, accurate, type 1 error rates were observed:
AlphaType 1 error0.100.1020.050.0520.010.01
In our study, the sample size for the individual univariate regressions to be meta-analyzed ranged between 1200 and 6000 thus we were satisfied with the results of the type-1 error simulations.
A different limitation was that the error covariance as specified here assumed total sample overlap, and the absence of any covariates. However, we did include covariates to control for population stratification and mean differences between male and female participants. These effects were assumed to be sufficiently small to allow our approximation to be valid. Strong covariate effects and substantial dropout would likely reduce power to detect an overall or age effect, and possibly increase type 1 error in some situations. As a form of sensitivity analysis the off diagonal elements of the phenotypic correlation matrices in ALSPAC and NTR were shrunk by 50% or 33% and increased by up to 10% to simulate the effect of less than total sample overlap or the effect of covariates changing the error covariance matrix. The conclusions remained virtually unchanged. Model 3 as described in the main text, had the best model fit when the off-diagonal elements in the phenotypic covariance matrix were reduced 50% or 33% and model 4 performed best when the phenotypic covariance was increased 10%. Parameter estimates and test statistics in model 3, fitted on the increased or decreased error covariance matrix were virtually unchanged. To conclude, the sensitivity analyses revealed that the effects of misspecification of the error covariance matrix probably did not influence the conclusions.
Parametric resampling of the data to account for the influence of sampling fluctuation
To quantify the influence of sampling fluctuation on our model selection, we resampled the input for the meta-regression from a multivariate normal distribution with means equal to the observed regression coefficients in the univariate PRS analyses, and covariance equal to the above specified error covariance matrix. Unlike non-parametric bootstrapping this technique makes assumptions about the asymptotic distributions of test statistics. However parametric resampling does allow for quantification of sampling variance in the model selection procedure.
We resampled 1000 datasets, on each of these the model selection procedure was repeated, the percentages in Table 1 (main text) reflect the percentage of resample datasets for which each model best fitted the data. The results represent the expected sample fluctuation in the model selection procedure induced by sample fluctuation, assuming the estimated effect sizes and error covariance are representative of the true effects and error covariance.
Mixed effects meta-regression to account for residual heterogeneity
In some cases a random effect which is unjustly omitted from the model can induce false positive results in meta-regression. Given only two cohorts are in the current study, a full mixed meta-regression makes little sense. However as a robustness check we fit two meta-regression models which allow for random effects and determine their influence on the results.
The best fitting fixed effects meta-analysis model (model 3; Table 1) revealed a moderate amount of residual variation not accounted for by the meta-regression model (Qe = 226,49, df=181, p = 0.0122). We therefore performed two additional random effects meta-analyses. Our first random effect model allowed for (correlated) random effects for each observed effect size, where the correlations between the random effects were assumed to be equal to the error covariance. The first mixed effects model significantly improved the model fit (LRT=8.81, df = 1, p = 0.0009). The fixed effects age, ODD/CD, and agexADHD were all significant (p < 0.05) in this mixed effects model (as they were in the fixed effects meta-regression model), but the effect agexODD/CD no longer reached significance (p = 0.0681) . The omnibus test including all meta-regression parameters also remained significant (QM = 37.0610, df = 10, p < 0.0001). The second mixed effects model included a random intercept, i.e., in this model, the only dependence between effect sizes was introduced by the meta-regressors or the error covariance. This second random effects model also significantly improved model fit over the fixed effects model (LRT = 14.7982, df=1, p < 0.0001). The second mixed effects model revealed a significant age effect and agexADHD effect (p < 0.05), but no significant ODD/CD and agexODD/CD effects. The overall test of parameters remained significant (QM= 30.0032, df=10, p = 0.0009). To conclude, both random effects models retained the main conclusions as the fixed effects model, i.e., an increasing association between schizophrenia PRS and childhood psychopathology with age, and some differences between the disorders in their relationship with schizophrenia.
Estimating genetic correlations based on the results from the polygenic risk score analyses
The univariate polygenic risk analyses results were obtained from either an ordered logistic regression (ALSPAC) or general estimation equations (GEE in NTR). For explanatory purpose we consider an OLS regression:
where the trait (y) and the PRS are scaled to unit variance and centered. The regression contained a number of other covariates (such as sex and principal components). Assuming the effects of the principal components and sex on the phenotypes were small to negligible, the square of B1 (B12) is equal to the variance explained in the phenotype by the PRS. We further assume that the squared predicted outcome of the multivariate meta-analyses correspond to R2. Given these assumptions we used the previously derived relationship between R2 and genetic correlation ADDIN REFMGR.CITE Dudbridge201387Power and predictive accuracy of polygenic risk scoresJournal87Power and predictive accuracy of polygenic risk scoresDudbridge,Frank2013Not in Filee1003348PLoS Genet93Public Library of Science1553-7404PLoS Genet15 to approximate the genetic correlations between childhood psychopathology and schizophrenia:
The inverse relationship equals:
From which we can obtain:
where N equals sample size in the discovery sample, M equals the independent number of genetic effects in the set of SNPs, QUOTE is the genetic covariance between target and discovery trait and QUOTE equals the genetic variance explained by all measured markers in the target trait. As the discovery sample here was an ascertained case control sample (34241 cases and 45604 controls), we substituted the effective N using the effective sample size formula proposed by Wilier et al. ADDIN REFMGR.CITE Willer201096METAL: fast and efficient meta-analysis of genomewide association scansJournal96METAL: fast and efficient meta-analysis of genomewide association scansWiller,Cristen J.Li,YunAbecasis,Gon+ºalo R.2010Not in File21902191Bioinformatics2617Oxford Univ Press1367-4803Bioinformatics16
Note that N is approximate and R2 is estimated directly in the PRS analyses. Therefore, we needed to assume values for M and QUOTE . Any uncertainty in these values will not be reflected in the confidence bounds around the genetic covariance. We assumed M to equal 200.000. To explore the influence of the uncertainty in QUOTE on the estimate of rg we computed the genetic correlations assuming the heritability explained by all SNPs included in the score for childhood psychopathology to be 0.15 (Figure 3), 0.20 (Figure S1) or 0.10 ( Figure S2). Note that we did not account for differences in the variance explained by the SNPs for the different psychopathologies at the different ages. We further assumed that the equations remained valid for estimates of B obtained from GEE (to correct for the presence of related samples) or ordered logistic regression (to correct for the fact that the ALSPAC phenotype was an ordered categorical variable).
ADDIN REFMGR.REFLIST Reference List
(1) Genome of the Netherlands Consortium. Whole-genome sequence variation, population structure and demographic history of the Dutch population. Nature genetics 2014;46(8):818-25.
(2) Abdellaoui A, Hottenga JJ, de Knijff P et al. Population structure, migration, and diversifying selection in the Netherlands. European journal of human genetics 2013;21(11):1277-85.
(3) Howie BN, Donnelly P, Marchini J. A flexible and accurate genotype imputation method for the next generation of genome-wide association studies. PLoS Genet 2009;5(6):e1000529.
(4) Genomes Project Consortium. An integrated map of genetic variation from 1,092 human genomes. Nature 2012;491(7422):56-65.
(5) Dudbridge F. Power and predictive accuracy of polygenic risk scores. PLoS Genet 2013;9(3):e1003348.
(6) Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics 2010;26(17):2190-1.
Figure S1
Figure S2
Figure S3
PAGE \* MERGEFORMAT 26 Nivard et al.
) I $ ' ) A C D J K k l
>
@
ȽȽֳ֥ȽȽȽֽȽȽȽ֔q $h{ hgz B*H*^J mH ph sH h B*H*^J mH ph sH !h{ hgz B*^J mH ph sH h{ h@ H*^J mH sH h1_ ^J mH sH h H*^J mH sH h{ hgz H*^J mH sH h{ hgz ^J mH sH h| 5\^J mH sH h{ hgz 5\^J mH sH , E
| < 6
7
]
^
i
r
gdY d gdgz d @&