Applied pharmacogenetics to predict response to treatment of first psychotic episode: study protocol

Applied pha macogene ics o p edic esponse o ea men o fi s psycho ic episode: s udy p o ocol Se gi Mas 1,2,3 *, Lau a Julià 2 , Manuel J. Cues a 3,4,5 , Benedic o C espo-Faco o 3,6,7,8 , Ja ie Va ´zquez-Bou gon 3,9,10 , Ca los Spuch 3,11 , Ana Gonzalez-Pin o 3,12 , Angela Ibañez 3,13 , Judi h Usall 14,15 , C is ina Rome o-Lo ´pez-Albe ca 3,16 , Ana Ca alan 3,17,18,19,20 , Anna Mane ´ 3,21,22,23 and Miquel Be na do 2,3,24,25 1 Depa men o Clinical Founda ions, Pha macology Uni , Uni e si y o Ba celona, Ba celona, Spain, 2 Ins i u d’in es igacions Biomèdiques Augus Pi i Sunye (IDIBAPs), Ba celona, Spain, 3 Cen o de In es igación Biomédica en Red en Salud Men al (CIBERSAM), Mad id, Spain, 4 Depa amen o de Psiquia ı ´a, Hospi al Uni e si a io de Na a a, Pamplona, Spain, 5 Ins i u o de In es igacio ´n Sani a ia de Na a a (IdiSNA), Pamplona, Spain, 6 Unidad de Ges ión Clínica de Salud Men al, Hospi al Uni e si a io Vi gen del Rocı ´o, Se illa, Spain, 7 Ins i u o de Biomedicina de Se illa (IBiS), Hospi al Uni e si a io VI gen del Rocio/Cen o Supe io de In es igaciones Cine íficas (HUVR/CSIC)/Uni e sidad de Se illa, Se ille, Spain, 8 Depa men o Psychia y, Uni e sidad de Se illa, Se ille, Spain, 9 Depa amen o de Psiquai ia, Ma que ´s de Valdecilla Uni e si y Hospi al –IDIVAL, San ande , Spain, 10 Depa amen o de Medicina y Psiquia ia, Uni e sidad de Can ab ia, San ande , Spain, 11 T ansla ional Neu oscience Resea ch G oup, Galicia Su Heal h Resea ch Ins i u e (IIS-Galicia Su ), Se izo Galego de Saúde/ Uni e sidad de Vigo (SERGAS-UVIGO), Vigo, Spain, 12 BIOARABA, Depa men Psychia y, Hospi al Uni e si a io Ala a, Uni e sidad del Paı ´s Vasco/Euskal He iko Unibe si a ea Vi o ia (UPV/EHU), Vi o ia, Spain, 13 Depa men o Psychia y, Hospi al Uni e si a io Ramo ´n y Cajal, Uni e sidad de Alcala ´, Ins i u o Ramón y Cajal de In es igación Sani a ia (IRYCIS), Mad id, Spain, 14 Ins i u de Rece ca San Joan de De ´u, Esplugues de Llob ega , Ba celona, Spain, 15 Cen e de Salu Men al, Pa c Sani a i San Joan de De ´u, San Boi de Llob ega , Ba celona, Spain, 16 Depa amen de Ciències Expe imen als i de la Salu , Depa men o Psychology, Uni e si y o Cadiz, Cádiz, Spain, 17 Psychia y Depa men , Basu o Uni e si y Hospi al, Osakide za, Basque Heal h Se ice, Bilbao, Spain, 18 Biobizkaia Heal h Resea ch Ins i u e, OSI Bilbao-Basu o, Bilbao, Spain, 19 Neu oscience Depa men , Uni e si y o he Basque Coun y, Leioa, Spain, 20 Depa men o Psychosis S udies, Ins i u e o Psychia y, Psychology & Neu oscience, King’s College London, London, Uni ed Kingdom, 21 Ins i u de Salud Men al, Hospi al del Ma , Ba celona, Spain, 22 Hospi al del Ma Resea ch Ins i u e, Ba celona, Spain, 23 Uni e si a Pompeu Fab a (UPF), Ba celona, Spain, 24 Ba celona Clinic Schizoph enia Uni , Hospi al Clı ´nic de Ba celona, Ba celona, Spain, 25 Depa amen de Medicina, Ins i u de Neu ociències (UBNeu o), Uni e si a de Ba celona (UB), Ba celona, Spain The applica ion o pe sonalized medicine in pa ien s wi h fi s -episode psychosis (FEP) equi es ools o classi yingpa ien sacco ding o hei esponse o ea men , conside ing bo h ea men e ficacy and oxici y. Howe e , se e al limi a ions ha e hinde ed i s ansla ion in o clinical p ac ice. He e, we desc ibe he a ionale, aims and me hodology o Applied Pha macogene ics o P edic Response o T ea men o Fi s Psycho ic Episode ( he Fa maPRED-PEP p ojec ), which aims o de elop and alida e p edic i e algo i hms o classi y FEP pa ien s acco ding o hei esponse o an ipsycho ics, he eby allowing he mos app op ia e ea men s a egy o be selec ed. These p edic o s will in eg a e, h ough machine lea ning echniques, pha macogene ic (measu ed as polygenic isk sco es) and epigene ic da a oge he wi h clinical, sociodemog aphic, en i onmen al, and neu oana omical da a. To do his, he Fa maPRED-PEP p ojec will use da a om wo al eady ec ui ed coho s: he PEPS coho om he “Geno ype-Pheno ype In e ac ion and En i onmen . Applica ion o a F on ie s in Psychia y on ie sin.o g01 OPEN ACCESS EDITED BY Michal Assa , Olin Neu opsychia y Resea ch Cen e (ONRC), Uni ed S a es REVIEWED BY Je ey Bishop, Uni e si y o Minneso a Twin Ci ies, Uni ed S a es Jacopo Sapienza, San Ra aele Scien ific Ins i u e (IRCCS), I aly *CORRESPONDENCE Se gi Mas [email p o ec ed] RECEIVED 17 Sep embe 2024 ACCEPTED 10 Decembe 2024 PUBLISHED 07 Janua y 2025 CITATION Mas S, Julià L, Cues a MJ, C espo-Faco o B, Va ´zquez-Bou gon J, Spuch C, Gonzalez-Pin o A, Ibañez A, Usall J, Rome o-Lo ´pez-Albe ca C, Ca alan A, Mane ´A and Be na do M (2025) Applied pha macogene ics o p edic esponse o ea men o fi s psycho ic episode: s udy p o ocol. F on . Psychia y 15:1497565. doi: 10.3389/ psy .2024.1497565 COPYRIGHT ©2025Mas,Julià,Cues a,C espo-Faco o, Va ´zquez-Bou gon, Spuch, Gonzalez-Pin o, Ibañez, Usall, Rome o-Lo ´pez-Albe ca, Ca alan, Mane ´andBe na do.Thisisanopen-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (CC BY). The use, dis ibu ion o ep oduc ion in o he o ums is pe mi ed, p o ided he o iginal au ho (s) and he copy igh owne (s) a e c edi ed and ha he o iginal publica ion in his jou nal is ci ed, in acco dance wi h accep ed academic p ac ice. No use, dis ibu ion o ep oduc ion is pe mi ed which does no comply wi h hese e ms. TYPE S udy P o ocol PUBLISHED 07 Janua y 2025 DOI 10.3389/ psy .2024.1497565 P edic i e Model in Fi s Psycho ic Episodes”s udy ( he PEPs s udy om he Spanish abb e ia ion) (N=335) and he PAFIP coho om “Clinical P og am on Ea ly Phases o Psychosis”(PAFIP om he Spanish abb e ia ion) (N = 350). These coho s will be used o c ea e he p edic o , which will hen be alida ed in a new coho , he Fa maPRED coho (N = 300). The Fa maPRED-PEP p ojec has been designed o o e come se e al o he limi a ions iden ified in pha macogene ic s udies in psychia y: (1) he sample size; (2) he pheno ype he e ogenei y and i s defini ion; (3) he complexi y o he pheno ype and (4) he gende pe spec i e. The global each o he Fa maPRED-PEP p ojec is o acili a e he e ec i e deploymen o p ecision medicine in na ional heal h sys ems. KEYWORDS pe sonalized medicine, an ipsycho ic, p edic ion, psychosis, Pha macogene ics In oduc ion Schizoph enia (SZ) is he mos pa adigma ic psycho ic diso de . The cou se o he illness is o en ch onic and highly a iable, causing a significan loss o quali y o li e o he pa ien and hei amily membe s (1). I also has a high cos o socie y, accoun ing o 10% o he global bu den o men al diso de s in Eu ope (2). The e is g ea a iabili y in he e ficacy o an ipsycho ic d ugs (APs) in he ea men o SZ as well as in he suscep ibili y o pa ien s o he side e ec s o hese d ugs. On a e age, be ween 20% and 30% o pa ien s do no espond app op ia ely o AP ea men and less han 40% achie e symp om emission (3), wi h a 70% ea men discon inua ion a e. This leads o elapses ha en ail new admissions, he eby wo sening he p ognosis, nega i ely a ec ing he pa ien ’s quali y o li e, and educing li e expec ancy (4). Fu he mo e, app oxima ely 30% o pa ien s de elop ea men - esis an SZ (TRS). I is essen ial o elucida e he unde lying pa hophysiology o TRS o iden i y bioma ke s o i s ea ly de ec ion and ea men . As an example, plasma le els o me aboli es in ol ed in he Kynu enine pa hway, a he c oss oad be ween neu oinflamma ion and glu ama e gic neu o ansmission, has been p oposed as bioma ke o TRS (5). Howe e , he a es o long- e m eco e y o fi s -episode psychosis (FEP) a e mo e a o able since abou 51% o pa ien s may achie e ei he symp oma ic, unc ional, and pe sonal eco e y (6). Selec ing he bes compound o each pa ien is a challenging p ocedu e (7), and i s selec ion, un o una ely, la gely elies on clinical expe ience and a ial-and-e o s a egy ha exposes he pa ien s o a highe isk o ad e se eac ions, p olongs he eco e y ime, and wo sens he long- e m esponse (8). In his con ex , he sea ch o bioma ke s o selec he mos sui able AP o each pa ien in he ea ly s ages is a p io i y a ea in psychia y (9). Pha macogene ics (PGx) has become one o he main ools in his sea ch o bioma ke s (3). Howe e , PGx esul s a e cha ac e ized by a lack o eplicabili y, hus limi ing hei ansla ion in o clinical p ac ice. These s udies ha e elied on he knowledge o pha macokine ic and/o pha macodynamic p ocesses. The pha macokine ic s udies ha e shown he mos p omising esul s o clinical implemen a ion as he gene ic a iabili y in me abolism- ela ed genes explains a significan pe cen age o he a iabili y in he plasma le els o some APs. Geno yping hese genes can be e y use ul in imp o ing APs e ficacy and ole abili y (10). In his ega d, in e na ional guideline g oups, such as he Du ch Pha macogene ics Wo king G oup (DPWG), ha e published pha macogene ic guidelines o gene-d ug in e ac ions in ol ing CYP2D6, CYP3A4, and CYP1A2 wi h an ipsycho ics (11), while he Clinical Pha macogene ics Implemen a ion Conso ium (CPIC) is de eloping simila guidelines (h ps://cpicpgx.o g/p io i iza ion-o -cpic-guidelines/). Addi ionally, egula o y agencies like he Food and D ug Adminis a ion (FDA) ha e iden ified 41 pha macogenomic bioma ke s ele an o psychia y in d ug labeling (h ps:// www. da.go /d ugs/science-and- esea ch-d ugs/ able- pha macogenomic-bioma ke s-d ug-labeling). By con as , he pha macodynamic s udies ha e been a ely eplica ed in independen popula ions, o en explaining a small pe cen age o he obse ed a iabili y ha limi s hei p edic i e capaci y (12). The candida e gene app oach is u he limi ed by he lack o unde s anding o he mechanism o ac ion o he APs. The pha macological esponse o APs can be defined as a complex pheno ype wi h polygenic inhe i ance in ol ing mul iple loci wi h small e ec s. Polygenic isk sco es (PRS) a e cons uc ed by summing he mul iple isk alleles associa ed wi h a pheno ype and a e weigh ed by he magni ude o hei es ima ed e ec in genome-wide associa ion s udies (GWAS) using la ge coho s wi h su ficien s a is ical powe (13). In psychia y, hese PRS ha e been calcula ed o es ima e he gene ic isk o a ious condi ions, such as SZ, majo dep ession, and bipola diso de (14). Se e al s udies ha e demons a ed ha he PRS calcula ed o psychopa hologies can be use ul in pha macogenomic s udies (14–17) as p edic o s o Mas e al. 10.3389/ psy .2024.1497565 F on ie s in Psychia y on ie sin.o g02 he esponse o APs (18–22). Addi ionally, he PRS calcula ed o pheno ypes ela ed o he pha macological esponse o APs, such as cogni i e pe o mance o me abolic al e a ions, can also be associa ed wi h he e ficacy o oxici y o AP ea men s (14,23– 25). The clinical applica ion o PRS and hei inclusion in heal hca e sys ems a e some o he mos p omising aspec s in implemen ing PGx in clinical p ac ice (13). Howe e , PRS ha e limi ed p ecision in hei p edic i e capaci y, as he e a e many o he sociodemog aphic, en i onmen al, clinical, and pha macological ac o s in ol ed in his complex pheno ype. Despi e hese limi a ions, he e is e idence o he clinical applicabili y o PRS, mainly o pa ien iden ifica ion and s a ifica ion (13,17). In hese examples, PRS ha e been inco po a ed in o isk algo i hms and a e al eady used in clinical p ac ice in some cases, inc easing hei p edic i e capaci y. The in eg a ion o PRS in o algo i hms ha include o he isk ac o s (i.e., sociodemog aphic, en i onmen al, clinical, and pha macological) ep esen s he u u e o pe sonalized medicine. Applied pha macogene ics o p edic esponse o ea men o fi s psycho ic episode PGx s udies ha e been hinde ed by limi a ions such as di ficul ies in ec ui ing la ge coho s, as hey equi e de ailed in o ma ion and longi udinal ollow-up o assess he esponse o pha macological ea men . Addi ionally, PGx s udies in SZ a e cons ained by disease-specificlimi a ions(14) such as diagnos ic he e ogenei y (mul iple and a iable symp oms accompanied by neu opsychological and unc ional impai men s) (26), mul iple como bidi ies, and pha macological he e ogenei y (mul iple APs wi h di e en p ope ies, a iable doses, AP combina ions, and concomi an medica ions) (27,28). Fu he mo e, he e is he e ogenei y in he defini ion o esponse pheno ypes ha is usually defined as a pe cen age o imp o emen in o e all symp oma ology o a h eshold alue measu ed in c oss-sec ional s udies, wi hou he conside a ion o he di e en symp oms and dimensions o psychia ic diso de s, hei longi udinal e olu ion, o yea s o ea men . Acco dingly, se e al au ho s p oposed adding unc ioning and pe sonal eco e y measu es o he lis o ou come indica o s o FEP, expanding i beyond symp oma ologic emission (29). The “Applied Pha macogene ics o P edic Response o T ea men o Fi s Psycho ic Episode”(Fa maPRED-PEP om he Spanish abb e ia ion) p ojec is a mul icen e s udy designed o allow he de elopmen and alida ion o a p edic i e algo i hm o he applica ion o pe sonalized medicine in pa ien s wi h fi s - episode psychosis (FEP). The pu pose o his p edic o will be o classi y FEP pa ien s acco ding o hei esponse pheno ype o APs, he eby allowing he mos app op ia e ea men s a egy o be selec ed. This p edic o will in eg a e, h ough machine lea ning echniques, pha macogene ic (measu ed as polygenic isk sco es) and epigene ic da a wi h clinical, sociodemog aphic, en i onmen al, and neu oana omical da a. To do his, he Fa maPRED-PEP p ojec will use he da a om wo al eady ec ui ed coho s o pa ien s wi h FEP and a longi udinal ollow- up: he PEPS coho om “Geno ype-Pheno ype In e ac ion and En i onmen . Applica ion o a P edic i e Model in Fi s Psycho ic Episodes”( he PEPs s udy om he Spanish abb e ia ion; PI08/ 0208) (N = 335) (30) and he PAFIP coho om “Clinical P og am on Ea ly Phases o Psychosis”(PAFIP om he Spanish abb e ia ion)”(N = 350) (31). These coho s will allow esponse pheno ypes o be defined using longi udinal da a, aking in o accoun no only he symp oma ological dimensions o he pa hology, bu also he neu ocogni i e dimensions and ad e se e ec s. These coho s will be used o c ea e as well as in e nally alida e he p edic o . This p edic o will be ex e nally alida ed in a new p ospec i e coho o pa ien s wi h FEP and a longi udinal ollow-up, he Fa maPRED coho (N = 300). S udy design S udy design Fa maPRED-PEP is an obse a ional, na u alis ic, and longi udinal s udy examining clinical ajec o ies and he p edic o s o clinical esponse o APs in FEP coho s (Figu e 1). P ojec aims The specific aims o he Fa maPRED-PEP s udy a e o: 1. Define ea men esponse pheno ypes o an ipsycho ics in wo coho s o pa ien s wi h fi s -episode psychosis (N = 700) using s a is ical echniques applied o longi udinal da a on symp oma ology, neu ocogni ion, and ad e se e ec s, be o e alida ing hese pheno ypes in a new p ospec i e coho (N = 300). 2. Reach consensus on clinical ecommenda ions o he defined ea men esponse pheno ypes. 3. De elop and pe o m in e nal, ex e nal, and p ospec i e alida ion o he p edic i e algo i hms o he defined esponse pheno ypes using machine lea ning echniques ha in eg a e pha macogene ic a iables (measu ed as PRS) and epigene ic da a along wi h clinical, sociodemog aphic, en i onmen al, and neu oana omical da a. 4. De elop p edic i e algo i hms o an ipsycho ic esponses specifically adap ed o each gende . 5. De elop a compu e applica ion ha con ains he p edic i e algo i hms o he ea men esponse pheno ypes and he clinical ecommenda ions o each. 6. S udy he easibili y o he clinical applicabili y o he p edic i e algo i hms in coo dina ion wi h heal hca e sys ems. 7. P omo e educa ional p og ams on pe sonalized and p ecision medicine in psychia y. 8. Explo e s a egies o p omo e access o genomic and heal h da a, and hei po en ial isks and benefi s in psychia ic pa ien s. Mas e al. 10.3389/ psy .2024.1497565 F on ie s in Psychia y on ie sin.o g03 To achie e hese objec i es, he Fa maPRED-PEP s udy has es ablished a na ional esea ch ne wo k ha is ocused on ec ui ing FEP pa ien s om 12 na ional s udy si es (Figu e 2), wi h eigh wo k packages ha ing al eady been designed (Table 1). The eams om hese si es a e membe s o he Biomedical Resea ch Ne wo king Cen e in Men al Heal h (CIBERSAM om he Spanish abb e ia ion), a Spanish ne wo k ocusing on ansla ional esea ch on he neu oscien ific aspec s ela ed o heal h and men al illness (www.cibe sam.es)(32). FIGURE 1 Fa maPRED-PEP s udy design and wo kflow. FIGURE 2 Fa maPRED-PEP esea ch ne wo k and s udy si es. Mas e al. 10.3389/ psy .2024.1497565 F on ie s in Psychia y on ie sin.o g04 Rec ui ed samples The Fa maPRED-PEP s udy will ake ad ance o wo al eady ec ui ed Spanish coho s o FEP pa ien s wi h longi udinal assessmen and is cu en ly ec ui ing a hi d coho . The wo coho s p e iously ec ui ed a e: he PEPS coho (N = 335) (30) and he PAFIP coho (N = 350) (31). A comple e desc ip ion o hese coho s can be ound elsewhe e (30,33). P ospec i e sample The Fa maPRED-PEP esea ch ne wo k is ec ui ing he Fa maPRED coho . The inclusion c i e ia a e: aged be ween 16 and 35 yea s a he ime o fi s e alua ion; a du a ion o posi i e symp oma ology ha does no exceed 12 mon hs; a du a ion o AP ea men ha does no exceed 3 mon hs; fluency in Spanish; and signed in o med consen o he s udy ( o mino s, by a legal gua dian i he pa ien ag ees o pa icipa e). The exclusion c i e ia a e: a neu ological diso de ; auma ic b ain inju y wi h a loss o consciousness; in ellec ual disabili y, no only an IQ < 70, bu also poo unc ioning; soma ic pa hology wi h a men al impac ; oxic psychosis; o a e usal o unde go gene ic es ing. The s udy was app o ed by he esea ch e hics commi ees o all he pa icipa ing clinical cen e s (HCB/2022/0079). In o med consen is o be ob ained om all he pa icipan s. In o med consen has been designed acco ding o he ecommenda ions o he Spanish P ecision Medicine In as uc u e Associa ed wi h Science and Technology (IMPaCT) (h ps://impac .isciii.es/) and he Ca los III Heal h Ins i u e (Spain) o ensu e p ope da a sha ing and o p omo e open science. P ima y clinical endpoin Clinical ajec o ies, de i ed om longi udinal da a ha cha ac e ize AP esponse pheno ypes, will se e as he p ima y clinical endpoin . These ajec o ies will u ilize da a collec ed a baseline, 3, 6, and 12 mon hs and will be iden ified ia clus e ing and la en a iable analyses. Response pheno ypes will be based on TABLE 1 Fa maPRED-PEP wo k packages. Wo k package Func ion WP1. Fa maPRED coho ec ui men Task 1. Decide clinical assessmen s and ime-poin s Task 2. Design and de elop a web-based EDC Task 3. Rec ui 300 pa ien s expe iencing FEP a 12 clinical cen e s. Longi udinal ollow-up will consis o ou isi s (baseline, a 3, 6 and 12 mon hs a e s udy inclusion) du ing which clinical and neu opsychological assessmen s will be conduc ed and he occu ence o ad e se e ec s and plasma le els o APs will be assessed. A he baseline isi , sociodemog aphic da a and amily his o y will be collec ed, a biological sample o DNA ex ac ion will be ob ained, and magne ic esonance imaging (MRI) will be pe o med. WP2. Defini ion o esponse pheno ypes and clinical ecommenda ions Task 1. De elop da a p ep ocessing p o ocols, including da a ans o ma ion, da a impu a ion, and da a ha moniza ion, among he h ee coho s included in he p esen s udy. Task 2. Define esponse pheno ypes o APs in FEP using 12-mon h longi udinal da a on symp oma ology, neu ocogni ion, and ad e se e ec s h ough clus e ing and la en a iable analyses. Task 3. De elop a guide o clinical ecommenda ions o each o he esponse pheno ypes. WP3. Pha macogene ic s udy Task 1. De elop and apply quali y con ol p o ocols o gene ic and epigene ic da a. Task 2. Calcula e selec ed polygenic isk sco es using whole-genome geno yping ollowing p e iously desc ibed s anda d p o ocols. Task 3. Calcula e epigene ic clocks and epigene ic isk sco es using whole-genome me hyla ion da a acco ding o s anda d p o ocols. Task 4. Pe o m a pha macokine ic s udy o he candida e genes in ol ed in he abso p ion, dis ibu ion, me abolism, and exc e ion o APs and i s ela ionship o AP plasma le els. WP4. De elopmen and alida ion o p edic i e algo i hms De elop p edic i e algo i hms o he iden ified AP esponse pheno ypes using baseline da a. The PEPS coho will be used o c ea e he p edic ion model and he PAFIP coho will be used as an independen alida ion sample. The mos obus model will be selec ed and fine- uned by me ging he PEPS and PAFIP coho s be o e being ex e nally alida ed using he Fa maPRED coho . Explainable a ificial in elligence (AI) me hods ha de i e in o ma i e p edic ions will be used. WP5. De elopmen o a clinical decision suppo sys em In coo dina ion wi h a specialized company, a compu e applica ion ha in eg a es he p edic i e algo i hms (WP4) and he clinical ecommenda ions (WP2) o he iden ified esponse pheno ypes will be de eloped. This applica ion mus in eg a e he a ious algo i hms and ecommenda ions sequen ially: (1) p edic ion o symp om ajec o y; (2) p edic ion o neu ocogni i e ajec o y; (3) p edic ion o ad e se e ec s; (4) he apeu ic s a egy ecommenda ions and APs; and (5) dosing ecommenda ions and he need o moni o ing plasma le els. WP6. S udy o clinical applicabili y and easibili y o in eg a ion in o he Spanish Na ional Heal h Sys em In coo dina ion wi h he e i o ial men al heal h manage s o each pa icipa ing au onomous communi y, bo h he clinical applicabili y and in eg a ion in o he Spanish Na ional Heal h Sys em o he findings om his s udy will be assessed. WP7. Educa ion and eaching o pe sonalized and p ecision medicine in psychia y A p og am will be de eloped o p omo e he eaching o pe sonalized and p ecision medicine in psychia y using he ne wo k o CIBERSAM cen e s as a p omo ion pla o m. WP8. Dissemina ion o esul s P epa a ion o publica ions and communica ions. Mas e al. 10.3389/ psy .2024.1497565 F on ie s in Psychia y on ie sin.o g05 longi udinal changes ac oss mul iple domains, including psycho ic symp oma ology (posi i e, nega i e, and gene al symp oms), a ec i e symp oms, and side e ec s. All measu es used o define hese ajec o ies a e de ailed below. Me hods Diagnosis The diagnosis o a psycho ic diso de will be es ablished using semi-s uc u ed in e iews based on he age o he pa ien : he Schedule o A ec i e Diso de s and Schizoph enia o School-Age Child en - P esen and Li e ime Ve sion (K-SADS-PL) (< 18 yea s old) (34) and he Mini-In e na ional Neu opsychia ic In e iew (MINI) (> 18 yea s old) (35). To e ospec i ely cha ac e ize and da e he ini ial symp oms o a psycho ic illness, he Symp om Onse in Schizoph enia (SOS) in en o y will be used (36). Demog aphic and en i onmen al ac o s A baseline, a comple e pe sonal and amily his o y will be aken, including a his o y o d ug use. En i onmen al ac o s and s esso s will also be eco ded using se e al measu es ha include he Lewis-Mu ay Obs e ic Complica ions Scale (37), he Lis o Th ea ening Expe iences Ques ionnai e (38) and he Childhood T auma Ques ionnai e (39), wi h in o ma ion on u banici y and he socioeconomic s a us also being collec ed using he Hollingshead- Redlich index. The P emo bid Adjus men Scale (PAS) (40) will be used o assess p emo bid adjus men , one o he mos s udied ac o s in ela ion o he p ognosis o psycho ic diso de s. In each e alua ion, weigh , heigh , he body mass index (BMI), blood p essu e, and he abdominal pe ime e will also be eco ded, wi h he aim o moni o ing he physical heal h indica o s. T ea men da a In each e alua ion, in o ma ion on he d ugs p esc ibed will be eco ded, including he ype o d ug, he du a ion o ea men , and dosage. In o ma ion on psychological ea men will also be collec ed. To assess ad e se d ug eac ions, se e al measu es will be included: he Ud alg ü Kliniske Unde sogelse (UKU) side e ec a ing scale (41), he Simpson-Angus scale, and gene al blood es s ( o al choles e ol, LDL, HDL, iglyce ides, glucose, and p olac in). AP plasma le els will be measu ed a each isi . The analysis o plasma le els oge he wi h he comple e geno yping o he cy och omes and anspo e s will allow he iden ifica ion o genuine non- esponde s, di e en ia ing hem om non-adhe e s o ea men . Adhe ence will be measu ed using he Mo isky G een Le ine Medica ion Adhe ence Scale (42). Clinical measu es A baseline and in each e alua ion, a comp ehensi e assessmen o psychopa hology will be pe o med ha will include well- es ablished measu es such as he Posi i e and Nega i e Synd ome Scale (PANSS) (43), he B ie Nega i e Symp om Scale (BNSS) (44), he Young Mania Ra ing Scale (YMRS) (45), and he Mon gome y —Asbe g Dep ession Ra ing Scale (MADRS) (46). The assessmen o global unc ioning will include he Clinical Global Imp ession (CGI) Scale (47), he Global Assessmen o Func ioning (GAF) Scale and Func ional Assessmen S aging (FAST) (48). Cogni ion The neu opsychological assessmen ba e y will be applied in he h ee-mon h e alua ion o ensu e ha he pa ien is clinically s able. The neu opsychological assessmen ba e y will be epea ed in he one-yea ollow-up isi . The s udy will use he MATRICS Consensus Cogni i e Ba e y (MCCB) (49) consis ing o 10 indi idually adminis e ed es s ha measu e cogni i e pe o mance in se en domains: speed o p ocessing (BACS: symbol coding; ca ego y fluency: animal naming; T ail Making Tes - Pa A), a en ion/ igilance (CPT-IP), wo king memo y (WMS®-III: Spa ial Span; Le e -Numbe Span), e bal lea ning (HVLT-R™), isual lea ning (BVMT-R™), easoning and p oblem sol ing (NAB®: Mazes), and social cogni ion (MSCEIT™: Managing Emo ions). The MCCB will be complemen ed by se e al es s o acili a e he ha moniza ion o he cogni i e da a among he coho s. These es s include: he Vocabula y and Ma ix Reasoning sub es s o he Wechsle Adul In elligence Scale – Fou h Edi ion (WAIS-IV) (50) o measu e cu en IQ; he Backwa d Digi Span sub es o he WAIS-IV (50) o measu e wo king memo y; and he T ail Making Tes - Pa B (51) o measu e execu i e unc ions. Addi ionally, a baseline, he Cogni i e Rese e Assessmen Scale in Heal h (CRASH) (52) will be used o measu e cogni i e ese e. Neu oimaging The mul imodal neu oimaging p o ocol is designed o a 3T MRI scanne and includes s uc u al imaging and es ing-s a e unc ional MRI ( MRI). Images will be collec ed a baseline o in he h ee-mon h e alua ion o ensu e ha he pa ien is clinically s able. The e a e cu en ly 11 MRI scanning si es wi h a a ie y o endo s and pla o ms. Da a will be collec ed om each cen e and p ocessed a one si e. Upon a i al a he da a analysis cen e , quali y con ol (QC) will check he consis ency o he da a and adhe ence o he specified imaging p o ocol. Following QC, he da a will be ma hema ically ha monized o elimina e esidual si e e ec s and will hen be p ocessed using obus analysis s eams. Mas e al. 10.3389/ psy .2024.1497565 F on ie s in Psychia y on ie sin.o g06 Gene ics and epigene ics In all e alua ions, blood samples will be collec ed in whole-blood EDTA ubes. The blood samples will be p ocessed using a s anda dized p o ocol o ob ain aliquo s o plasma ( o plasma AP assessmen ) and bu y coa ( o DNA ex ac ion). Samples will be s o ed in -80°C eeze s. The ele an me ada a will include he as ing du a ion, he ime since he las AP dosage, and ecen illnesses o inflamma o y condi ions. DNA will be ex ac ed om bu y coa samples ollowing es ablished s anda d p o ocols. No malized DNA concen a ions will be sen o he Spanish Na ional Geno yping Cen e (CeGen). Whole- genome geno yping will be unde aken using he Axiom™Spain Biobank A ay (de eloped in he Uni e si y o San iago de Compos ela, Spain) (The mo Fishe Scien ific) and da a will be analyzed o ob ain polygenic isk sco es (PRS) o schizoph enia and o he psychopa hologies as well as o he condi ions including cogni ion, pe sonali y ai s, and me abolic o inflamma o y ai s. GWAS summa y esul s will be downloaded om he Psychia ic Genomics Conso ium and he Science Gene ic Associa ion Conso ium. S anda d pipeline analysis (p e iously desc ibed (25)) will be pe o med, allowing e ficien quali y con ol, ela edness es ing, p incipal componen analysis, and geno ype impu a ion. The selec ed PRSs will be compu ed using he PRS con inuous sh inkage (PRS-CS) so wa e, a Bayesian-based me hod ha in e s an upda ed pos e io SNP e ec size by applying con inuous sh inkage o he disco e y GWAS summa y s a is ics. The ex e nal linkage disequilib ium (LD) e e ence panel will be cons uc ed using a publicly a ailable subsample o he UK Biobank. The de i ed PRSs will be s anda dized o z-sco es wi h mean 0 and a s anda d de ia ion o 1. The geno yping o specific pha macogene ic genes (including cy och omes and anspo e s) will be conduc ed wi h he Pha macoScan a ay (The mo Fishe Scien ific). Only genes wi h meaning ul impac s on AP me abolism, as pe CPIC le els A–B o gene-d ug in e ac ions (h ps://cpicpgx.o g/genes-d ugs/), will be included. Gene ic a ian s will be anno a ed using s a (*) alleles, whe e applicable, ia he Pha mVa da abase (h ps:// www.pha m a .o g/). Fo s a is ical analysis, geno ype- o- pheno ype ansla ions will ollow consensus ecommenda ions (e.g., CYP2D6) (53) o , i absen , CPIC-defined pheno ypes. DNA me hyla ion will be assessed using he Illumina Infinium Me hyla ionEPIC 2.0 ki (Illumina) and analyzed o calcula e di e en epigene ic clocks (54)andme hyla ionp ofile sco es (MPS) (55) acco ding o s anda d p o ocols ha include quali y con ol, be a- alue compu a ion, no maliza ion, and singula alue decomposi ion o iden i y and emo e unwan ed sou ces o a ia ion (56,57). We will use publicly a ailable da a om me hyla ion-wide associa ion s udies as a disco e y sample o calcula e me hyla ion p ofile sco es using simila app oaches o hose used o he cons uc ion o PRS. We de eloped a code o compu ing MPS accessible ia he ollowing public eposi o y on Gi Hub: h ps://gi hub.com/agonse/me hylsco e. Da a collec ion Pheno ypic da a will be collec ed and managed using Lib eClinica, an open-sou ce elec onic da a cap u e (EDC) sys em o clinical ials based on he OpenClinica(R) (OC) 3 communi y edi ion. A web-based EDC has been designed specifically o he collec ion o da a om all he measu es and ques ionnai es. Real- ime da a alida ion and quali y ules will be defined o ensu e ha he da a a e en e ed accu a ely and as comple ely as possible. Clinical a e s will di ec ly en e he ques ionnai e da a using he co esponding da a en y o ms. Each da a en y will hen be ca e ully double-checked by an ex e nal p ojec managemen eam ha will esol e any disc epancy o que y. S anda d ope a ing p ocedu es will be es ablished o ensu e p ope da a collec ion and compa abili y ac oss he s udy si es. Da a analysis The analyses will ocus on wo main goals: (1) he cha ac e iza ion o AP esponse pheno ypes using longi udinal da a and (2) he p edic ion o hese pheno ypes. Clinical ajec o ies u ilizing da a om baseline, 3-, 6-, and 12-mon h ollow-ups will iden i y AP esponse pheno ypes based on he longi udinal change in di e en psycho ic symp om domains (posi i e, nega i e, and gene al) as well as in a ec i e symp oms and side e ec s. T ajec o y analyses will include all he in o ma ion a ailable om all he ime-poin s and will apply s a e-o - he-a clus e ing and la en a iable analyses. The PEPS coho (N = 300) will be used o iden i y hese AP esponse pheno ypes and o pe o m hei clinical cha ac e iza ion. P edic ions will be based on he da a collec ed a baseline and he cogni ion in o ma ion collec ed in he 3-mon h ollow-up. To ensu e ou abili y o es he obus ness o he p edic ion models in an unbiased manne , he PEPS coho will be used o c ea e he p edic ion model, while he PAFIP coho (N = 350) will be used as an unseen independen alida ion sample. Explainable machine lea ning me hods will be used. The mos obus model will hen be selec ed and fine- uned by me ging he PEPS and PAFIP coho s be o e being ex e nally alida ed using he Fa maPRED coho . Sample size and powe We used a ecen ly-desc ibed me hod o de e mine sample size o de eloping a clinical p edic ion model using a adi ional likelihood-based app oach (e.g., logis ic eg ession) (58). Sample sizes we e de i ed o he scena ios o low, medium o high p edic ion pe o mance (Nagelke ke R-squa ed alue o 0.25/0.40/ 0.55), wi h he numbe o pa ame e s in he model being 10. The minimum expec ed equency o he AP esponse clus e was 15% acco ding o p e ious s udies wi h he PEPS coho (59). Using he R package ‘pmsampsize’, we es ima ed he sample sizes equi ed o bina y ou comes. Fo a bina y ou come, he e a e h ee c i e ia o de e mining sample size: (1) a small o e fi ing defined by an expec ed sh inkage o p edic o e ec s by 10% o less; (2) a small absolu e di e ence o 0.05 in he model’s appa en and adjus ed Nagelke ke R-squa ed alue; and (3) a p ecise es ima ion (wi hin +/- 0.05) o he a e age ou come isk in he popula ion o a key Mas e al. 10.3389/ psy .2024.1497565 F on ie s in Psychia y on ie sin.o g07 ime-poin o in e es o p edic ion. Each c i e ion may equi e a di e en sample size and he chosen sample size is he la ges o he h ee. Fo his s udy, when he ou come is conside ed bina y, he c i e ion wi h he la ges equi ed sample size will be c i e ion 1 o a low p edic i e pe o mance and 2 o medium and high pe o mances. Assuming a pheno ype equency o 15% and a maximum o 10 pa ame e s o be included in he p edic i e model, he minimum sample size o a model is es ima ed o be be ween 290 and 310. Discussion The Fa maPRED-PEP p ojec has been designed o o e come se e al o he limi a ions iden ified in PGx s udies in psychia y, including: 1. Sample size. As commen ed p e iously, he ec ui men o la ge coho s wi h longi udinal ollow-ups and deep pheno yping is challenging and di ficul o achie e. He e, he aim o he s udy is o collec da a om 1000 pa ien s h ough he in eg a ion o h ee coho s o FEP pa ien s: he PEPS coho , he PAFIP coho , and he Fa maPRED coho . The assessmen ins umen s and measu emen ime-poin s o be used in he Fa maPRED coho we e decided ia a wo king g oup discussion o acili a e he ha moniza ion o clinical and neu ocogni i e da a among he coho s. 2. Pheno ype he e ogenei y and defini ion. One ac o hinde ing he iden ifica ion o p edic o s o AP esponse is he inconsis ency in how AP esponse pheno ypes a e defined ac oss a ious s udies (28,59). Many esea che s adop bina y classifica ions o measu e e ec i eness (ca ego izing subjec s as ei he esponde s o non- esponde s) o e alua e oxici y (de e mining whe he an ad e se e ec is p esen o absen ), al hough he e is no consensus ega ding he ideal cu -o alues o hese a iables. Such bina y classifica ions ail o cap u e he complex na u e o AP esponses, which a e influenced by bo h he e ec i eness o he ea men (encompassing mul iple aspec s such as clinical symp oms, neu ocogni i e pe o mance, and quali y o li e) and he side e ec s in ol ed. Addi ional challenges in p edic ing AP esponses include he di e se p og ession o he disease i sel . P edic ing esponse ou comes using pa ien s wi h ch onic condi ions in oduces mo e a iabili y and limi s he applicabili y o he findings due o di e ences in illness du a ion, diagnos ic c i e ia, and p e ious ea men s. Con e sely, pa ien s expe iencing hei fi s episode o schizoph enia gene ally show less a ia ion in hei p io use o an ipsycho ics, making hem mo e app op ia e subjec s o s udying p edic o s o ea men ou comes (60,61). The aim o he Fa maPRED-PEP s udy is o define he pheno ype o esponse o APs using longi udinal da a om o e he cou se o one yea a e he FEP and conside ing ha his esponse is a complex pheno ype encompassing no only he emission o he a ious symp oms ha cha ac e ize he pa hology, bu also he neu ocogni i e aspec s and he eme gence o ad e se e ec s. 3. The complexi y o he pheno ype. The AP esponse is a complex pheno ype wi h a polygenic basis ha could be pa ially cap u ed using PRS. Howe e , as is he case wi h o he isk ac o s used in heal hca e (e.g., choles e ol le els), PRS ha e a p edic i e capaci y wi h limi ed accu acy, meaning hey canno p edic a clinical a iable o in e es wi h su ficien p ecision a he indi idual le el. E en i he PRS cap u ed all he gene ic a iabili y in a pha macological esponse a ibu able o common gene ic a ian s, hei p edic i e capaci y would s ill be impe ec , mainly o wo easons. Fi s ly, gene ic ac o s a e no he only isk ac o s ha explain a iabili y in a pha macological esponse as he e a e many o he sociodemog aphic, en i onmen al, clinical, and pha macological ac o s in ol ed in his complex pheno ype (62,63). Secondly, PRS only accoun o he con ibu ion o common gene ic a ian s, each wi h a small e ec , wi hou conside ing he e ec s o a e gene ic a ian s, which a e less equen bu ha e a la ge impac , o he e ec s a ibu able o epigene ic modifica ions. Al hough he inclusion o a e gene ic a ian s, iden ifiable h ough massi e sequencing echniques, is no expec ed o inc ease he p edic i e capaci y o he PRS in he sho e m (13), he inclusion o epigene ics in pha macogenomic s udies is expec ed o help explain much o he missing he i abili y, ha is, he he i abili y o a pheno ype ha is no cap u ed by gene ic a ian s (3). In his ega d, a iables such as epigene ic clocks ha e ecen ly been shown o be ela ed no only o schizoph enia, bu also o he esponse o APs (56,64,65). 4. The gende pe spec i e. The s udy assumes a gende -based app oach as a c oss-cu ing axis o imp o e heal h in e en ions. The e o e, i s objec i e is o de elop p edic i e algo i hms o he esponse o AP, especially ailo ed o each gende . The di e ences be ween women and men in he incidence o FEP a e well known, bu less s udied in ela ion o he esponse o AP. In his sense, he s udy will no be limi ed o conside ing gende as a simple ca ego ical a iable o in oduc ion in o s a is ical models, bu aims o conduc an in-dep h s udy o he impac ha gende has on pha macological esponse in FEP and i s in e ac ion wi h o he heal h de e minan s (e.g., age, socioeconomic s a us, educa ional le el, e c.). The e o e, his s udy will p omo e he sea ch, de ec ion, and analysis o di e ences, as well as simila i ies, be ween men and women, bo h in e ms o e ficacy and in e ms o he equency and ype o ad e se e ec s. Thus, i aims o de e mine whe he he imp o emen in psycho ic symp oma ology, bu also in neu ocogni i e cha ac e is ics, as well as he equency and se e i y o ad e se e ec s, a e exclusi e o one o he wo sexes, mo e p e alen in one o he wo sexes, wi h di e en Mas e al. 10.3389/ psy .2024.1497565 F on ie s in Psychia y on ie sin.o g08 cha ac e is ics be ween bo h sexes, o e en i hey ecei e di e en esponses om he sys em depending on whe he hey a e men o women. To a oid some o he gende biases and common malp ac ices in heal h esea ch, he s udy da a will be collec ed and analyzed disagg ega ed by gende and will be p esen ed in epo s and publica ions de i ed in he same way. O e coming hese limi a ions, he aim o he Fa maPRED-PEP s udy is o de elop algo i hms o p edic AP esponse pheno ypes using gene ic and epigene ic da a (measu ed as isk sco es) oge he wi h clinical, sociodemog aphic, en i onmen al, and neu oana omical da a. The u u e o medicine is ocused on o e ing a comp ehensi e app oach o heal h, aking in o accoun all he de e mining ac o s o heal h o illness. The gene a ion o la ge amoun s o heal h da a has exceeded he capaci y o manage all a ailable in o ma ion in eal ime; he e o e, a ificial in elligence can become a suppo ool o heal hca e p o essionals, who, howe e , mus always ha e he final say in decision-making. Machine lea ning sys ems, and mo e specifically deep lea ning ones, a e capable o ex ac ing pa e ns and gene a ing conclusions om a la ge amoun o da a using complex o mulas and associa ions, making hem less in ui i e. This ac limi s he unde s anding and comp ehension o he sys ems by heal hca e p o essionals, making i di ficul o in eg a e hem in o wo kflows and gene a ing a lack o confidence in he esul s hey can p o ide. Ou p ojec will be ocus in explainable a ificial in elligence me hods ha de i e in o ma i e p edic ions ha can be mapped back o indi idual ea u es and bioma ke s, as opposed o some deep lea ning and o he “black-box” echniques ha may no be app op ia e o achie ing in e p e abili y. The applica ion o p ecision medicine in psychia y has been defined as a mul i-s age p ocess (66), whe e PRS play a significan ole in each s age, such as in: (1) p edic ing he isk o de eloping he diso de o design p e en i e s a egies; (2) s a i ying pa ien s based on hei own cha ac e is ics, bo h clinical and gene ic, o iden i y he mos e ec i e ea men , ea men - esis an pa ien s, and suscep ibili y o de eloping ad e se e ec s; and (3) op imizing dosage egimens o ensu e ea men e ficacy and ole abili y based on d ug kine ics and pa ien gene ic cha ac e is ics. The Fa maPRED-PEP s udy ocuses on he second and hi d s ages o his s a egy. Al hough he Fa maPRED s udy has been designed o add ess se e al limi a ions commonly ound in pha macogenomics (PGx) s udies, ce ain cons ain s should be conside ed when in e p e ing u u e findings. Fi s ly, as a na u alis ic s udy, ea men he e ogenei y and he non- andom selec ion o specific ea men s could ac as po en ial con ounde s. Secondly, while allowing p io an ipsycho ic (AP) ea men o less han h ee mon hs aims o acili a e sample ec ui men and ensu e adequa e s a is ical powe , i may also in oduce a iabili y ha could con ound clinical ajec o ies. Thi dly, al hough ea men adhe ence will be moni o ed, no in e en ions o enhance adhe ence a e planned; as a esul , adhe ence a iabili y may u he con ound he esponse pheno ypes based on clinical ajec o ies. Finally, despi e he implemen a ion o igo ous da a ha moniza ion p o ocols o ensu e compa abili y ac oss he h ee coho s in ol ed in he s udy, uncon olled di e ences be ween coho s may s ill influence he findings. The applica ion o hese s a egies in pe sonalized and p ecision medicine is especially ele an in pa ien s wi h FEP, as eco e y a e FEP has become he p ima y goal o any ea men s a egy (61), wi h hei p og ess du ing he fi s yea o ea men conside ed c ucial o disease p ognosis and elapse p e en ion. Al hough SZ is a po en ially disabling and se ious men al illness, an app op ia e mul idisciplina y app oach o FEP can con ibu e o comple e eco e y (67). A c i ical pe iod o 2 o 5 yea s a e FEP has been defined, du ing which he apeu ic in e en ions may be mo e success ul in p e en ing elapses and achie ing unc ional eco e y in pa ien s (68). P e en ing elapses in he ea ly s ages o he disease has become a majo challenge due o he c i ical impac ha a elapse can ha e on unc ional p ognosis (69). Non- adhe ence o APs and a lack o disease knowledge a e e y common in pa ien s wi h FEP and a e associa ed wi h an inc eased isk o elapse, leading o a wo se disease cou se and unc ionali y (70–72). Psycho ic diso de s encompass schizoph enia, schizoph eni o m diso de , schizoa ec i e diso de , b ie psycho ic diso de , and subs ance-induced psycho ic diso de . Addi ionally, o he men al heal h condi ions, such as majo dep essi e diso de , bipola diso de , obsessi e-compulsi e diso de (OCD), pos auma ic s ess diso de (PTSD), and au ism spec um diso de , may also p esen wi h sho - o medium- e m psycho ic symp oms. Gi en he gene ic pleio opy sha ed among hese condi ions -cap u ed h ough polygenic isk sco es- and he widesp ead use o APs o ea hem, he findings o he Fa maPRED-PEP s udy ha e he po en ial o o e ansdiagnos ic insigh s ha ex end beyond fi s -episode psychosis (FEP) pa ien s. Pha macogenomic s udies and pe sonalized and p ecision medicine in he field o psycho ic diso de s will need o add ess he challenges ha mus be aced o ensu e hei u u e applicabili y in clinical p ac ice. These challenges ha e been defined by he In e na ional Conso ium o Pe sonalized Medicine (ICPe Med) (h ps://www.icpe med.eu/index.php) and include, among o he s, he managemen o genomic and heal h da a in e ms o ensu ing confiden iali y as well as access o ci izens and esea che s, he in ol emen o heal h au ho i ies in p omo ing pe sonalized medicine o acili a e i s implemen a ion in heal hca e sys ems, he educa ion and in ol emen o heal hca e p o essionals o ensu e knowledge, access, and applica ion o pe sonalized and p ecision medicine s a egies, he mul idisciplina y in eg a ion o esea che s and clinicians o de elop pe sonalized medicine s a egies, and he collec ion o da a and hei use o a mo e e ficien pa ien -cen e ed heal hca e sys em. These challenges a e also he pilla s o IMPaCT (h ps://impac .isciii.es/) o he Ca los III Heal h Ins i u e (Spain), he ounde o he Fa maPRED-PEP s udy. The aims o ou s udy a e aligned wi h he mission o IMPaCT o acili a e he e ec i e deploymen o p ecision medicine in he Spanish Na ional Heal h Sys em, ensu ing scien ific and echnical quali y, equi y, and e ficiency in he use o a ailable scien ific esou ces o mee he needs o he ci izen y. Mas e al. 10.3389/ psy .2024.1497565 F on ie s in Psychia y on ie sin.o g09