O iginal A icle
P edic ing clinical and unc ional ajec o ies in
indi iduals wi h i s -episode psychosis by
baseline de ia ions in g ey ma e olume
Manuel Muñoz-Ca acuel, Claudio Alemán-Mo illo, Na alia Ga cía-San-Ma ín, Na halia Ga ido-To es,
Ma ía Alemany-Na a o, Richa d A. I. Be hlehem, Lena Do schmid , Jakob Seidli z, Rosa Ayesa-A iola,
Ja ie Vázquez-Bou gon, Miguel Ruiz-Veguilla, Li espan B ain Cha Conso ium,
Benedic o C espo-Faco o*and Ra ael Rome o-Ga cía*
Backg ound
P edic ing long- e m ou come ajec o ies in psychosis emains
a c ucial and challenging goal in clinical p ac ice. The
iden i ica ion o eliable neu oimaging ma ke s has o en been
hinde ed by he clinical and biological he e ogenei y o psycho ic
diso de s and he limi a ions o adi ional case-con ol
me hodologies, which o en mask indi idual a iabili y. Recen ly,
no ma i e b ain cha s de i ed om ex ensi e magne ic
esonance imaging (MRI) da a-se s co e ing he human li espan
ha e eme ged as a p omising biologically d i en solu ion,
o e ing a mo e indi idualised app oach.
Aims
To examine how de ia ions om no ma i e co ical and
subco ical g ey ma e olume (GMV) a i s -episode psychosis
(FEP) onse ela e o symp om and unc ional ajec o ies.
Me hod
We le e aged he la ges a ailable b ain no ma i e model
(N>100 000) o explo e no ma i e de ia ions in a sample o
o e 240 pa ien s wi h schizoph enia spec um diso de s who
unde wen MRI scans a he onse o FEP and ecei ed clinical
ollow-up a 1, 3 and 10 yea s.
Resul s
Ou indings e eal ha de ia ions in egional no ma i e GMV a
FEP onse a e signi ican ly linked o o e all long- e m clinical
ajec o ies, modula ing he e ec o ime on bo h symp om and
unc ional ou come. Speci ically, nega i e de ia ions in he le
supe io empo al gy us and B oca’s a ea a FEP onse we e
no ably associa ed wi h a mo e se e e p og ession o posi i e and
nega i e symp oms, as well as wi h unc ioning ajec o ies o e
ime.
Conclusions
These esul s unde sco e he po en ial o b ain de elopmen al
no ma i e app oaches o he ea ly p edic ion o diso de
p og ession, and p o ide aluable insigh s o he
de elopmen o p e en i e and pe sonalised he apeu ic
s a egies.
Keywo ds
Schizoph enia; neu oimaging; no ma i e models; p edic ion;
ou comes.
Copy igh and usage
© The Au ho (s), 2025. Published by Camb idge Uni e si y P ess
on behal o Royal College o Psychia is s. This is an Open
Access a icle, dis ibu ed unde he e ms o he C ea i e
Commons A ibu ion licence (h ps://c ea i ecommons.o g/lice
nses/by/4.0/), which pe mi s un es ic ed e-use, dis ibu ion
and ep oduc ion, p o ided he o iginal a icle is p ope ly ci ed.
Schizoph enia spec um diso de s a e se e e men al heal h
condi ions ha may p o oundly dis up daily li e and cause in ense
su e ing o hose a ec ed and hei amilies. Thei cou se ypically
spans a wide spec um o symp om and unc ional ajec o ies,
anging om comple e eco e y o occasional elapses wi h
sus ained unc ionali y and symp om- ee pe iods, o o pe sis en
symp oms leading o signi ican psychological, social and occupa-
ional impai men .1–6
No ma i e b ain cha s
Resea ch on i s -episode psychosis (FEP) has ex ensi ely u ilised
magne ic esonance imaging (MRI) o in es iga e b ain mo phologi-
cal al e a ions as po en ial p edic o s o symp om and unc ional
ou comes.7Al hough signi ican clinical associa ions o b ain
imaging-de i ed pheno ypes ha e been iden i ied, he obse ed
esul s ha e p o en o be sub le and he e ogeneous, which has
hinde ed hei ansla ion in o clinical p ac ice.8A po en ial key
limi a ion is ha mos s udies ha e employed case-con ol
app oaches, which o e look po en ially di e se biological and
clinical p o iles by conside ing wi hin-g oup a ia ions as noise,
he eby igno ing indi idual di e ences and unde mining he
eplicabili y o he esul s.9To add ess hese limi a ions, no ma i e
b ain cha s de i ed om la ge MRI da ase s spanning he human
li espan ha e ecen ly eme ged as a biologically d i en solu ion,10
enabling he assessmen o indi idual b ain mo phology de ia ions
in he psychosis spec um agains no ma i e neu ode elopmen al
anges o sex and age.11 Recen longi udinal s udies using no ma i e
modelling ha e shown signi ican associa ions be ween no ma i e
co ical and subco ical de ia ions a FEP and symp om p og ession
o e ime.9,12–14 Howe e , s udies using hese app oaches emain
sca ce, and ele an aspec s such as unc ioning and speci ic
symp oms emain unexplo ed. Fu he esea ch is needed o iden i y
eliable biological p edic o s ha can de e mine, om he ea ly s ages
o psychosis, which pa ien s a e a isk o poo o a ou able
long- e m symp om and unc ional ajec o ies.15,16 Such ad ances
would enhance ea ly clinical decision-making, op imise he apeu ic
esou ces and p omo e p e en i e, pe sonalised and comp ehensi e
men al heal h ca e aimed a maximising ea men ou comes and
quali y o li e o his popula ion. Wi hin his amewo k, he cu en
s udy in es iga es po en ial p edic o s o clinical and unc ional
ajec o ies, ocusing on he ela ionship be ween co ical and
subco ical g ey ma e olume (GMV) de ia ions a i s b eak o he
illness, and on long- e m symp om and unc ional ajec o ies. We
hypo hesise ha de ia ions in co ical and subco ical egions a
baseline will p edic long- e m clinical and unc ioning ajec o ies.
*Join senio au ho s.
The B i ish Jou nal o Psychia y (2025)
1–9. doi: 10.1192/bjp.2025.105
1
h ps://doi.o g/10.1192/bjp.2025.105 Published online by Camb idge Uni e si y P ess
Me hod
Pa icipan s
Pa icipan s we e d ug-naï e pa ien s expe iencing a FEP and
ec ui ed h ough an epidemiological Ea ly In e en ion P og amme
o Ini ial Phases o Psychoses (PAFIP) a he Uni e si y Hospi al
Ma qués de Valdecilla, San ande , Spain.17 This coho comp ised a
o al o 352 FEP pa ien s (138 emales, mean age 29.71 ±8.77 yea s)
who p o ided w i en consen o pa icipa e in a longi udinal s udy.
The s udy was app o ed by he e hical commi ee o he Uni e si y o
Se ille ( e e ence no. 2024-2534). Pa ien s we e ea ed wi h he
minimum e ec i e an ipsycho ic dosage, in acco dance wi h clinical
s anda ds.
Pa ien s we e eligible o he s udy i hey: (a) we e aged
15–60 yea s; (b) me DSM-IV c i e ia wi hin he Schizoph enia
Spec um and O he Psycho ic Diso de s (schizoph enia, schizo-
ph eni o m diso de , schizoa ec i e diso de , b ie psycho ic
diso de , schizo ypal pe sonali y diso de o psychosis no
o he wise speci ied), as assessed by he S uc u ed Clinical
In e iew o DSM Diso de s; and (c) had no p io ea men
wi h an ipsycho ic medica ion o , i p e iously ea ed, had a o al
li e ime o adequa e an ipsycho ic ea men o <6 weeks. Exclusion
c i e ia we e: (a) mee ing DSM-IV c i e ia o d ug dependence
(excluding nico ine dependence); (b) mee ing DSM-IV c i e ia o
in ellec ual disabili y; and (c) ha ing a his o y o neu ological
disease o head inju y. The ini ial diagnosis was e iewed a e a
6-mon h pe iod and upda ed, i necessa y, a ha ime o du ing
ollow-up.
MRI acquisi ion and olume ex ac ion a FEP
B ain s uc u al MRI scans we e conduc ed a FEP baseline using a
1.5-T Gene al Elec ic SIGNA Sys em (GE Medical Sys ems,
Milwaukee, WI, USA) and a 3T Philips Medical Sys ems MRI
scanne (Achie a, Bes , The Ne he lands) equipped wi h an eigh -
channel head coil. A h ee-dimensional T
1
-weigh ed sequence was
employed. The pa ame e s o 1.5 T we e: echo ime 5 ms, epe i ion
ime 24 ms, numbe o exci a ions 2, o a ion angle 45°, ield o iew
26 ×19.5 cm, slice hickness 1.5 mm and a ma ix o 256 ×192; and
o 3T, echo ime 3.7 ms, epe i ion ime 8.2 ms, lip angle 8°,
acquisi ion ma ix 256 ×256, oxel size 0.94 ×0.94 ×1 mm and 160
con iguous slices. P io o analysis, all images unde wen isual
inspec ion o de ec ion o po en ial mo ion a e ac s and g oss
ana omical abno mali ies. F eeSu e 6.0.0 econ-all was used o
ex ac egional olumes o each o he 68 co ical egions and 14
subco ical egions o he Desikan–Killiany pa cella ion (h p://su e
.nm .mgh.ha a d.edu). None o he pa icipan s had a o al GMV
mo e han i e s anda d de ia ions abo e o below he mean, so none
was iden i ied as an ou lie .
GMV cen ile es ima ion
To benchma k he GMVs o ou coho , we employed p io
no ma i e ajec o ies ac oss he li espan GMV de elopmen om
o e 100 000 neu o ypical indi iduals, which we e es ima ed using
a gene alised addi i e model o loca ion, scale and shape
(GAMLSS;10 code and no ma i e ajec o ies a ailable a
h ps://gi hub.com/b aincha /Li espan). This me hod allows he
es ima ion o cen iles ( anging om 0 o 1), indica ing
pa icipan s’de ia ions om he median no ma i e ajec o y
a e adjus ing o age, sex and scanning si e. Addi ionally, we
assessed ha cen ile es ima ion in ou sample would no a y
be ween scanne s (see Supplemen a y Ma e ials a ailable a
h ps://doi.o g/10.1192/bjp.2025.105, Supplemen a y Fig. 1).
Cen iles below 0.5 indica e a GMV educ ion compa ed wi h
he no ma i e alue, while hose abo e 0.5 e eal olume inc ease
(Fig. 1a). Thus, cen iles o 352 FEP pa icipan s a baseline
Age
L
Amygdala Accumbens Hippocampus
Pallidum Cauda e
Pu amen Thalamus
R
GMV
Cen iles
FEP
Time
0.7
0.5
0.3
Gene alised mixed modelling
Cen ilesB
L
Nega i e symp oms
Posi i e symp oms
Global unc ioning
baseline
cen iles
1
0
(a)
(b)
(c)
Fig. 1 (a) Illus a ion o cen ile-based no ma i e modelling, showing he anges ha de ine neu o ypical b ain g ey ma e olume (GMV)
p og ession wi h age.10 (b) GMV cen iles a e aged ac oss pa icipan s a baseline (BL). (c) Schema ic ep esen a ion o he s udy pipeline
explo ing he associa ions be ween baseline i s -episode psychosis (FEP) cen iles and symp om and unc ional ou comes o e ime.
Muñoz-Ca acuel e al
2
h ps://doi.o g/10.1192/bjp.2025.105 Published online by Camb idge Uni e si y P ess
(cen iles
BL
) we e compu ed o each co ical and subco ical
egion (Fig. 1b). All b ain igu es we e gene a ed using he
ENIGMA oolbox.18
Longi udinal assessmen o symp om se e i y and
unc ioning
Assessmen s o symp om and unc ional ou comes we e conduc ed
o e ime by expe ienced clinicians a ou ime poin s (baseline,
1 yea , 3 yea s and 8–15 yea s), wi h inal assessmen ex ending up
o 15 yea s. Symp om se e i y was assessed using he Scale o he
Assessmen o Nega i e Symp oms (SANS) and he Scale o he
Assessmen o Posi i e Symp oms (SAPS).19,20 Func ioning was
assessed using he Global Assessmen o Func ioning (GAF) scale
(Axis V o DSM-IV), a 0–100 a ing sys em ha e alua es o e all
psychological, social and occupa ional unc ioning.
S a is ical analyses
We used mixed modelling o e alua e he impac o co ical and
subco ical baseline cen iles on symp om and unc ional ajec o ies
o e ime (Fig. 1c). We op ed o a gene alised mixed model wi h
a Poisson dis ibu ion a he han a linea mixed model (LMM) o
wo main easons: (a) he dependen a iables (symp oms
and unc ioning) exhibi ed exponen ial dis ibu ions (see
Supplemen a y Ma e ials, Supplemen a y Fig. 2); and (b) symp om
and unc ioning sco es a e disc e e o coun da a, which a e mo e
accu a ely modelled using a Poisson dis ibu ion. To ensu e
uni o mi y, all independen a iables we e z-sco ed p io o
eg ession analysis. Only pa icipan s wi h a leas wo assessmen s
o SAPS, SANS and GAF du ing he ollow-up we e included in he
mixed modelling analyses.
The i s model conside ed sepa a ely he 82 egional GMV
cen iles a FEP baseline (68 co ical egions and 14 subco ical
egions), which we e speci ied as ollows:
Ou comes 1an ipsycho ic ea men egional cen iles
ime egional cen iles × ime age o onse
sex in ac anial olume 1jpa icipan
(1)
The second model conside ed he o al co ical and subco ical
GMV cen iles a FEP baseline, which we e speci ied as ollows:
Ou comes 1an ipsycho ic ea men o al co ical cen ile
o al subco ical cen ile ime
o al co ical cen ile × ime
o al subco ical cen ile × ime age o onse
sex in ac anial olume 1jpa icipan
(2)
In bo h models, Ou comes encompasses sepa a e dependen
a iables o o al sco es on SANS and SAPS, hei speci ic
symp oms and he GAF. An ipsycho ic ea men e e s o
medica ion le els exp essed in chlo p omazine (CPZ) equi alen s.21
Time measu es he du a ion since he i s con ac wi h Men al
Heal h Se ices ollowing FEP. Age o onse e e s o he
pa icipan ’s age a he ime o i s con ac wi h Men al Heal h
Se ices ollowing FEP. Sex is ea ed as a bina y a iable (male 1).
In ac anial olume (E i ) ep esen s he esidual alues o he
es ima ed o al in ac anial olume a e adjus ing o age and sex
e ec s. Finally, he e m 1 | pa icipan deno es a andom in e cep
o each pa icipan .
To es o he po en ial impac o non- andom a i ion on ou
indings, we applied - es s (o a non-pa ame ic equi alen es
when app op ia e) o assess di e ences in o al GMV baseline
cen iles, symp oms and unc ioning. Di e ences we e compu ed
be ween pa ien s who comple ed a leas wo symp om and
unc ional assessmen s and hose who d opped ou a e he ini ial
baseline assessmen .
In all ins ances whe e mul iple compa ison co ec ion was
equi ed, alse disco e y a e (FDR) was con olled using he
Benjamini–Hochbe g p ocedu e.
Resul s
O he ini ial 352 FEP pa ien s included, only hose wi h a leas wo
ou come assessmen s du ing he ollow-up we e analysed. This
esul ed in a inal sample o 241 pa ien s o clinical assessmen
(SANS and SAPS) and 244 o unc ioning assessmen (GAF).
De ailed in o ma ion abou sociodemog aphic, clinical and unc-
ional cha ac e is ics o he inal sample is shown in Supplemen a y
Ma e ial, Supplemen a y Table 1. The e we e no signi ican
di e ences in GMV cen iles, clinical o sociodemog aphic measu es
a baseline be ween he inal sample (pa ien s wi h ≥2 assessmen s)
and hose who d opped ou a e he ini ial assessmen (all
P- alues >0.146; see Supplemen a y Ma e ial, Supplemen a y
Table 2). Simila ly, no signi ican di e ences in GMV cen iles we e
ound a baseline be ween pa ien s diagnosed wi h schizoph enia and
hose diagnosed wi h o he schizoph enia spec um diso de s (see
Supplemen a y Ma e ial, Supplemen a y Fig. 3).
Regional baseline cen iles and ou come ajec o ies
Ou i s gene alised mixed model analysis examined he
in e ac ion e ec s be ween egional cen iles a FEP and ime on
SANS, SAPS and GAF o al sco es (Fig. 2a), while accoun ing o
he con ibu ion o clinical and sociodemog aphic co a ia es. Fo
in o ma ion on he e ec s o egional cen iles alone, please e e o
Supplemen a y Ma e ials, Supplemen a y Fig. 4.
The egional analysis o SANS o al sco es e ealed signi ican
nega i e in e ac ions be ween ime and cen iles ac oss a ious
co ical egions ha we e pa icula ly p ominen in he le supe io
empo al co ex (z=−4.10, P<0.001), on al pole (z=−4.74,
P<0.001), igh cuneus (z=−4.34, P<0.001), la e al occipi al
(z=−4.19, P<0.001), pa ahippocampal (z=−4.47, P<0.001)
and pe icalca ine (z=−4.51, P<0.001). This indica es ha highe
cen iles in hese egions a FEP a e associa ed wi h a mo e
a ou able p og ession o nega i e symp oms o e he long e m. In
con as , posi i e in e ac ions we e obse ed in subco ical egions
including he bila e al pallidum (le , z=2.65, P=0.045; igh ,
z=3.45, P=0.006) and le halamus (z=2.71, P=0.040),
e ealing ha highe cen iles in hese egions a FEP a e linked o a
wo sening p og ession o nega i e symp oms o e ime.
The egional analysis o SAPS o al sco es showed signi ican
in e ac ions be ween ime and cen iles in a ious co ical egions
ha we e bo h posi i e: le caudal middle on al (z=11.23, P<
0.001), supe io pa ie al (z=5.40, P<0.001), hippocampus
(z=3.05, P=0.018), bila e al pos cen al (le , z=7.86; igh ,
z=6.55, P<0.001), pa acen al (le , z=4.81; igh , z=4.89, P<
0.001) and igh cauda e (z=2.77, P=0.035); and nega i e: le
supe io empo al (z=−4.12, P<0.001), bila e al insula (le ,
z=−6.97; igh , z=−5.52, P<0.001), pa s iangula is (le ,
z=−5.30; igh , z=−6.10, P<0.001) and pa s ope cula is (le ,
z=−6.24; igh , z=−6.65, P<0.001).
The egional analysis o unc ioning (GAF o al sco e) e ealed
posi i e in e ac ions be ween ime and cen iles ac oss widesp ead
Volume ic b ain de ia ions in FEP p edic ou come ajec o ies
3
h ps://doi.o g/10.1192/bjp.2025.105 Published online by Camb idge Uni e si y P ess
co ical and subco ical egions, wi h no able impac s on he le
supe io empo al (z=5.12, P<0.001), sup ama ginal (z=4.63,
P<0.001), medial o bi o on al (z=6.53, P<0.001), hippocam-
pus (z=4.74, P<0.001), igh cuneus (z=6.98, P<0.001) and
pe icalca ine egions (z=6.61, P<0.001). This indica es ha
highe cen iles in hese egions a FEP a e associa ed wi h g ea e
imp o emen in unc ioning o e ime.
G aphical ep esen a ions o hese ime–ou come in e ac ions
a e depic ed in Fig. 2b. These illus a e how ha ing a high (>0.75)
o low (<0.25) egional cen ile in a ep esen a i e egion, such as
he le supe io empo al co ex, a FEP impac s he empo al
p og ession o he clinical a iables (SANS, SAPS, GAF). Fo a
comple e desc ip ion o model ou comes, please e e o
Supplemen a y Da a 1.
Regional baseline cen iles and ou comes ajec o ies
ac oss speci ic symp oms
As a subsequen explo a ion, we analysed he in e ac ion e ec s
be ween ime and egional baseline cen iles on speci ic posi i e and
nega i e symp oms, as depic ed in Fig. 3. Rega ding he in e ac ion
esul s o speci ic nega i e symp oms, signi ican nega i e e ec s
we e obse ed in empo al, in e io on al, p ecen al and occipi al
co ical egions o alogia, ina en ion and a oli ion se e i y, bu
no o anhedonia o a ec i e la ening. In con as , he subco ical
esul s e ealed a posi i e in e ac ion o anhedonia in he le
pallidum and o a oli ion in he igh pallidum, while a nega i e
in e ac ion was ound o ina en ion in he igh amygdala (all
|z|>2.70, all P<0.041). This sugges s ha highe cen iles a FEP in
speci ic co ical egions a e associa ed wi h a mo e a ou able
p og ession o nega i e symp oms, whe eas highe cen iles in he
pallidum gene ally lead o a wo sening p og ession o nega i e
symp oms o e ime.
Conce ning he in e ac ion esul s o speci ic posi i e
symp oms, signi ican posi i e e ec s we e obse ed in a ious
co ical egions, pa icula ly he le caudal middle on al egion
and, o a lesse ex en , he bila e al pos cen al and pa acen al
egions, o all posi i e symp oms (biza e beha iou , delusions,
o mal hough diso de and hallucina ions). Nega i e in e ac ion
e ec s, howe e , we e ound mainly in he insula, pa s ope cula is,
pa s iangula is and occipi al egions. On he o he hand, he
subco ical esul s e ealed posi i e in e ac ions o delusions and
hallucina ions, and nega i e in e ac ions o o mal hough
diso de and biza e beha iou (all |z|>2.60, all P<0.049; o
comple e model esul s, please e e o Supplemen a y Da a 2).
Sociodemog aphic and clinical p edic o s o symp om
and unc ional ou comes
In ou inal explo a ion, we analysed he con ibu ion o
sociodemog aphic and clinical ac o s o he se e i y o symp oms
and unc ional ou comes o e ime in he s udy sample, as well as
0
Cen ileBL > 0.75
Cen ileBL < 0.25
Cen ileBL > 0.75
Cen ileBL < 0.25
Cen ileBL > 0.75
Cen ileBL < 0.25
00 20 40 60 80 100 2 4 6
SANS o al sco e
0246
SAPS o al sco e
GAF o al sco e
51015
Time (yea s)
0 5 10 15
Time (yea s)
0 5 10 15
Time (yea s)
GAF
SAPS
SANS
LR
LR
LR
Z- alues
5
12
8
–8
–5
–12
Time × cen ilesBL
(a) (b)
Fig. 2 (a) Co ical and subco ical egional in e ac ions be ween ime and baseline (BL) cen iles a FEP o SANS, SAPS and GAF o al sco es.
S a is ically signi ican alues (P<0.05, alse disco e y a e-co ec ed) a e ep esen ed by colou ed egions, while non-signi ican egions a e
shown in whi e. (b) G aphical ep esen a ion o in e ac ion e ec s be ween ime and SANS, SAPS and GAF o al sco es, using he le supe io
empo al co ex cen ile o di ide he sample in o wo g oups: high cen ile (>0.75, yellow) and low cen ile (<0.25, g ey). The g aphs show he
e olu ion o SANS, SAPS and GAF sco es o e he yea s, wi h eg ession lines o each g oup. FEP, i s -line psychosis; SANS, Scale o he
Assessmen o Nega i e Symp oms; SAPS, Scale o he Assessmen o Posi i e Symp oms; GAF, Global Assessmen o Func ioning scale.
Muñoz-Ca acuel e al
4
h ps://doi.o g/10.1192/bjp.2025.105 Published online by Camb idge Uni e si y P ess
SANS
A ec i e la ening Biza e beha iou
Alogia Delusions
Fo mal hough diso de
SAPS
Time × cen ilesBL
RL RL
RL
RL
RLRL
R
Z- alues
–5 5
Z- alues
–8 8
L
RL
RL
Anhedonia
Ina en ion
A oli ion
Hallucina ions
Fig. 3 Co ical and subco ical egional in e ac ion e ec s be ween ime and baseline (BL) cen iles a i s -episode psychosis on speci ic
nega i e (SANS) and posi i e (SAPS) symp oms. S a is ically signi ican alues (P<0.05, alse disco e y a e-co ec ed) a e ep esen ed by
colou ed egions, while non-signi ican egions a e in whi e. SANS, Scale o he Assessmen o Nega i e Symp oms; SAPS, Scale o he
Assessmen o Posi i e Symp oms.
Volume ic b ain de ia ions in FEP p edic ou come ajec o ies
5
h ps://doi.o g/10.1192/bjp.2025.105 Published online by Camb idge Uni e si y P ess
he impac o o al co ical and subco ical cen iles a baseline and
hei in e ac ion wi h ime (Fig. 4).
Age o onse showed a nega i e associa ion wi h ina en ion,
indica ing ha la e FEP onse is associa ed wi h less a en ional
impai men . Sex was associa ed wi h ou comes, whe e being male
was linked o wo se sco es on SANS, SAPS and GAF o al sco es, as
well as on a ec i e la ening, a oli ion, anhedonia and delusions.
An ipsycho ic ea men , exp essed in CPZ equi alen s,21 in lu-
enced SANS, SAPS and GAF o al sco es, as well as hei speci ic
symp oms. Speci ically, i was associa ed wi h mo e se e e nega i e
symp oms (excep o ina en ion) bu educed posi i e symp oms
and imp o ed unc ioning as CPZ equi alen s inc eased. Time was
posi i ely associa ed wi h GAF and nega i ely associa ed wi h SAPS
o al sco e, all posi i e symp oms and, o a lesse ex en , SANS o al
sco e, a oli ion and ina en ion, demons a ing a gene alised
imp o emen o e ime ollowing a FEP. To al co ical cen ile a
FEP baseline showed a posi i e associa ion wi h GAF, indica ing
ha highe o al co ical cen iles a FEP we e linked o be e
unc ioning ou comes. In con as , o al subco ical cen ile a FEP
baseline did no show any signi ican associa ions wi h he
measu ed a iables. The in e ac ion be ween ime and o al co ical
cen ile a baseline showed a nega i e associa ion wi h SANS o al
sco e, alogia and a oli ion, as well as a posi i e associa ion wi h
o mal hough diso de . This indica es ha highe o al co ical
cen iles a FEP a e linked o g ea e long- e m imp o emen in
mos nega i e symp oms and lowe long- e m imp o emen in
o mal hough diso de . Con e sely, he in e ac ion be ween ime
and o al subco ical cen ile a baseline exhibi ed he opposi e end,
showing a posi i e associa ion wi h SANS o al sco e, alogia,
a oli ion and anhedonia, along wi h a nega i e associa ion wi h
o mal hough diso de (all |z|>2.17, all P<0.041; o comple e
model esul s, please e e o Supplemen a y Da a 3).
Discussion
In his s udy, we in es iga ed po en ial p edic o s o long- e m
symp om and unc ional ou comes in schizoph enia spec um
diso de s h ough b ain no ma i e models, along wi h sociodemo-
g aphic a iables and clinical ac o s. We ound ha (a) co ical and
subco ical GMV no ma i e de ia ions a he onse o psychosis
modula ed he e ec o ime on bo h symp om and unc ional
ou comes, and (b) an ipsycho ic ea men , ime and sex we e
signi ican ly associa ed wi h symp om and unc ional ou comes
o e ime.
The in e ac ion e ec s be ween ime and cen iles sugges ha
di e en b ain egions con ibu e dis inc i ely o posi i e symp-
oms, nega i e symp oms and unc ioning ajec o ies, unde -
sco ing he complexi y o he b ain mac os uc u al subs a es
unde lying each o hese clinical mani es a ions. An excep ion was
obse ed in he le supe io empo al egion, whe e lowe co ical
olume was associa ed wi h mo e se e e ajec o ies ac oss all
measu ed ou comes, ex ending ecen indings ela ing lowe
no ma i e olumes in his egion wi h g ea e nega i e symp oms.12
The nega i e associa ion be ween supe io empo al gy us and
posi i e symp oms, along wi h he nega i e associa ions be ween
ce ain p e on al co ex (PFC) egions (e.g. on al pole, pa s
o bi alis) and nega i e symp oms, align wi h indings om
An ipsycho ic ea men Time Age o onse
Sex (male)
E i
To al co ical cen ileBL Time × o al co ical cen ileBL
Time × o al subco ical cen ileBL
To al subco ical cen ileBL
Nega i e symp oms Posi i e symp oms Global unc ioning
–20
Z- alues
20
–10
10
0
–20
Z- alues
20
–10
10
0
–5.0
Z- alues
5.0
–2.5
2.5
0.0
–5.0
Z- alues
5.0
–2.5
2.5
0.0
–5.0
Z- alues
5.0
–2.5
2.5
0.0
–5.0
SANS
SAPS
GAF
A ec i e la ening
Alogia
A oli ion
Anhedonia
Ina en ion
Hallucina ions
Delusions
Biza e beha iou
Fo mal hough
SANS
SAPS
GAF
A ec i e la ening
Alogia
A oli ion
Anhedonia
Ina en ion
Hallucina ions
Delusions
Biza e beha iou
Fo mal hough
SANS
SAPS
GAF
A ec i e la ening
Alogia
A oli ion
Anhedonia
Ina en ion
Hallucina ions
Delusions
Biza e beha iou
Fo mal hough
Z- alues
5.0
–2.5
2.5
0.0
–5.0
Z- alues
5.0
–2.5
2.5
0.0
–5.0
Z- alues
5.0
–2.5
2.5
0.0
–5.0
Z- alues
5.0
–2.5
2.5
0.0
Fig. 4 Z- alues o he gene alised mixed model o SANS, SAPS and GAF o al sco es, as well as o SANS and SAPS speci ic symp oms ac oss
he a iables o in e es . S a is ically signi ican alues (P<0.05, FDR-co ec ed) a e ep esen ed by solid-colou ed ba s, while non-signi ican
alues a e shown wi h aded colou ing. BL, baseline; SANS, Scale o he Assessmen o Nega i e Symp oms; SAPS, Scale o he Assessmen o
Posi i e Symp oms; GAF, Global Assessmen o Func ioning scale; FDR, alse disco e y a e.
Muñoz-Ca acuel e al
6
h ps://doi.o g/10.1192/bjp.2025.105 Published online by Camb idge Uni e si y P ess
ENIGMA s udies.22 Addi ionally, lowe olume in B oca’s a ea
(Desikan–Killiany: pa s ope cula is, pa s iangula is), ecen ly
p oposed as a po en ial neu opa hological o igin o schizoph e-
nia,23,24 was linked o wo sening p og ession in bo h posi i e
symp oms and unc ioning, as well as some nega i e symp oms,
u he suppo ing he po en ial ole o b ain egions in ol ed in
language ci cui y in he p ognosis o psychosis.24
No ably, a signi ican posi i e in e ac ion e ec o o al
posi i e symp oms was obse ed in he le caudal middle on al
egion, also known as he do sola e al PFC (DLPFC), sugges ing
ha highe co ical olume in his egion a he onse o psychosis
migh be associa ed wi h a less a ou able p og ession o posi i e
symp oms o e ime. This inding con as s wi h he hinne
olume ypically epo ed among schizoph enia pa ien s in his
egion,25 and i s associa ion wi h cogni i e dys unc ion26 and
poo e symp oma ic and unc ional ou comes.7Howe e , consid-
e ing ha simila indings ha e p e iously been epo ed,27 and ha
one o he mos obse ed e ec s o an ipsycho ic ea men is he
no malisa ion o p e on al ac i a ion,28 his esul may sugges ha
pa ien s wi h lowe olume in he DLPFC migh bene i o a g ea e
ex en om an ipsycho ic ea men s ha amelio a e DLPFC
hypo unc ionali y associa ed wi h posi i e symp oma ology.
Fu he esea ch is needed o elucida e his ques ion.
Rega ding he impac o no ma i e subco ical olume ic
de ia ions on ou comes, ecen s udies in psychosis ha e ound ha
smalle limbic olumes and la ge basal ganglia a e ela ed o wo se
ou comes.9,29 Al hough he in e ac ion e ec s in ou subco ical
esul s we e highly he e ogeneous, hey s ill aligned wi h some o
hese indings. Speci ically, a highe olume ic cen ile in he
cauda e egion was linked o wo se p og ession o posi i e
symp oms, while a lowe cen ile in he amygdala was associa ed
wi h less a ou able ajec o ies o posi i e symp oms and
unc ioning.
A de ailed analysis o egional in e ac ion e ec s be ween cen iles
and ime o speci ic posi i e and nega i e symp oms allowed o a
mo e nuanced clinical cha ac e isa ion o he obse ed egional
e ec s. Speci ically, he associa ion be ween lowe cen iles in he le
supe io empo al egion and wo sening ajec o ies o nega i e
symp oms o e ime appea ed o be p ima ily linked o alogia and
ina en ion, wo symp oms closely ela ed o language unc ion.
These indings, along wi h consis en esul s in B oca’s a ea o he
same symp oms, ex end p e ious esea ch epo ing dys unc ion in
co ical language egions as a po en ial p edic o o ou comes in
clinical high- isk you hs.30 Delusions and hallucina ions exhibi ed a
sha ed posi i e in e ac ion e ec in he le cauda e, while in e ac ion
e ec s spanning mos posi i e symp oms we e iden i ied in he
pos cen al gy us, B oca’s a ea, on o-insula egion and DLPFC,
albei wi h dis inc signs. These indings sugges ha common
neu obiological pa hways may unde lie mos posi i e symp oms a
he co ical le el, whe eas g ea e neu obiological he e ogenei y
could cha ac e ise di e en nega i e symp oms and subco ical
egions. Toge he , hese esul s p o ide insigh s in o po en ial
neu oana omical signa u es o speci ic symp om p og ession, which
migh con ibu e o mo e accu a e p ognos ic assessmen s and
in o m a ge ed p e en i e in e en ions.
The obse ed a ou able e ec o an ipsycho ic ea men on
posi i e symp oms and unc ioning indica es he e ec i eness o
cu en ea men s. Howe e , his con as s wi h hei de imen al
impac on nega i e symp oma ology, indica ing ha , while
an ipsycho ic ea men may educe acu e psycho ic s a es and
unc ional impai men , i migh also con ibu e o he in ensi ica-
ion o nega i e symp oms. Whe he his symp om accen ua ion is
a p ima y mani es a ion o he diso de (p ima y nega i e
symp om) o a seconda y ad e se e ec o he medica ion, medical
como bidi ies o en i onmen al ac o s (seconda y nega i e
symp om) emains a long-s anding and elusi e discussion in
psychia y,31,32 which is beyond he scope o his s udy. An
excep ion was obse ed wi h ina en ion which, unlike o he
nega i e symp oms, showed clinical imp o emen wi h an ipsy-
cho ic ea men . This inding aligns wi h cu en consensus
s a emen s on nega i e symp oms,33,34 which classi y ina en ion as
a cogni i e a he han a nega i e symp om.
The gene al bene icial impac o ime unde ea men on
posi i e symp oms and daily unc ioning ollowing FEP aligns wi h
clinical obse a ions and exis ing esea ch,3,4,13,35 challenging he
K aepelinian no ion o schizoph enia spec um diso de s as
ine i ably de e io a ing condi ions. This iew is u he con es ed
by ecen longi udinal s udies on no ma i e b ain mo phology
de elopmen , which ha e ound ha mo phological de ia ions
p esen a FEP end o no malise o e he long e m.13 This sugges s
ha , wi h adequa e ea men , he unc ional p ognosis can be
a ou able in mos cases. Howe e , as highligh ed by p e ious
esea ch,2a signi ican p opo ion o FEP pa ien s s ill expe ience
incomple e eco e y and un a ou able ajec o ies, unde sco ing
he need o iden i y clinically ele an p edic o s o be e
cha ac e ise his he e ogenei y. On he o he hand, he modes
e ec o ime unde ea men on nega i e symp oms indica es ha
hese symp oms may ollow a mo e s able cou se han posi i e
symp oms, wi h minimal imp o emen s o e ime, which is
consis en wi h p e ious indings4,36 and emphasises he p essing
need o mo e a ge ed pha macological and psychosocial
in e en ions in his a ea.
Being male was associa ed wi h wo se symp om and unc ional
ou comes ollowing a FEP, eplica ing well-s ablished sociodemo-
g aphic indings.37–39 Addi ionally, he subsequen analysis e ealed
ha males had highe a es o delusions, a ec i e la ening,
a oli ion and anhedonia, which may p o ide aluable in o ma ion
o a ge ing speci ic symp oms in his subg oup o pa ien s.
Howe e , al hough younge age a FEP has ypically been ound o
p edic poo e ou comes,37,38 age o onse did no show a signi ican
associa ion wi h symp om o unc ional ou comes in ou sample,
excep o ina en ion.
While a highe o al co ical olume ic cen ile a FEP was
ound o p edic be e o e all unc ioning o e se e al yea s, no
associa ion was obse ed wi h nega i e o posi i e symp oms, no
be ween o al subco ical cen ile and unc ioning. This isola ed
e ec is unexpec ed gi en he adi ional ela ionship be ween
unc ioning and symp oma ology, pa icula ly nega i e symp-
oms,40 and hus wa an s ca e ul conside a ion. P e ious longi u-
dinal esea ch using no ma i e models ound ha nega i e
de ia ions in mean co ical hickness we e associa ed wi h mo e
se e e nega i e symp oms,12 while nega i e de ia ions in he
olume o a ious subco ical egions we e linked o lowe posi i e
symp oms.9Al hough we did no ind such di ec associa ions in
ou sample, we did ind in e ac ion e ec s be ween ime and o al
cen iles, indica ing ha highe o al subco ical cen ile and lowe
o al co ical cen ile a FEP p omo ed a mo e a ou able ajec o y
o nega i e symp oms o e ime. This sugges s ha o al GMV
de ia ions migh be a use ul gene al measu e o p edic ion o
nega i e symp oms p og ession o e ime.
S eng hs and limi a ions
To ou knowledge, his s udy includes he la ges b ain no ma i e
modelling da a-se o da e, wi h o e 100 000 pa icipan s,
alongside a signi ican inal sample size (o e 240 pa ien s) and a
long ollow-up pe iod (o e 10 yea s), con e ing bo h s a is ical
powe and clinical ele ance o ou esul s. The s udy ex ends
ecen longi udinal esea ch using b ain no ma i e modelling in
psychosis9,12–14 by in es iga ing p e iously unexplo ed ac o s,
Volume ic b ain de ia ions in FEP p edic ou come ajec o ies
7
h ps://doi.o g/10.1192/bjp.2025.105 Published online by Camb idge Uni e si y P ess
including o e all unc ioning and speci ic posi i e and nega i e
symp oms. Addi ionally, a key s eng h o his s udy is i s ocus on
ou come ajec o ies, which ha e ecei ed li le a en ion in
longi udinal app oaches o neu oimaging p edic o s compa ed
wi h o e all ou comes (i.e. ac oss all ime poin s). Al hough he
dis inc ion be ween hese wo app oaches is sub le, i is also o
majo ele ance: he i s measu es he ex en o which he p edic o
is associa ed wi h ou comes ac oss all assessed imepoin s, while he
second e lec s how he p edic o modula es he e ec o ime unde
ea men on ou comes. We belie e his la e app oach may be
e en mo e clinically in o ma i e, as i e eals he expec ed a e o
change o e ime, a he han he o e all ou comes, based on
di e en neu omo phological p o iles, which could enable mo e
accu a e p edic ions o ea men e icacy o indi idual pa ien s.
Some limi a ions need o be conside ed. Fi s , he signi ican
d op-ou a e a yea 10 u he inc eases he isk o a i ion bias,
which may ha e a ec ed he eliabili y o ou indings. Second,
symp om and unc ional ou comes we e modelled using a limi ed
se o a iables, o p e en model o e i ing and o acili a e
in e p e a ion acco ding o he s udy’s objec i es. Howe e , his
app oach may ha e o e looked o he a iables ha could also ha e
in luenced he ou comes, such as he du a ion o un ea ed
psychosis, cogni i e unc ioning, como bid subs ance use and
p emo bid adjus men , among o he s, which may limi he abili y
o ully explain ou come ajec o ies. Thi d, since all pa ien s we e
medica ed and he e was no un ea ed compa ison g oup due o
e hical conside a ions, we canno disen angle he e ec s o medica ion
om he na u al cou se o he diso de . Fou h, an ipsycho ic
ea men was calcula ed as CPZ equi alen , a me hod ha educes
he numbe o a iables ela ed o medica ion and he isk o model
o e i ing bu o e looks he indi idualised impac o speci ic
medica ions. Fi h, unc ioning was assessed using a gene ic sco ing
sys em (GAF) ha lacks speci ici y. Fu u e esea ch could bene i om
mo e comp ehensi e ins umen s o assess social and occupa ional
unc ioning. Finally, he e is a need o eplica ion in independen
samples om di e se popula ions o enhance gene alisabili y.
Implica ions
Ou indings sugges ha de ia ions om no ma i e GMV
ma u a ion a he onse o psychosis may in luence long- e m
clinical and unc ional ajec o ies, highligh ing he po en ial o
b ain de elopmen al no ma i e app oaches o ea ly p edic ion o
diso de p og ession. Addi ionally, he associa ions wi h speci ic
symp oms p o ide aluable insigh s ha could guide he de elop-
men o p e en i e in e en ions, po en ially enabling pe sonalised
he apeu ic s a egies and mo e e icien use o esou ces.
Manuel Muñoz-Ca acuel , Men al Heal h Uni , Vi gen del Rocío Uni e si y Hospi al-
CIBERSAM, Se ille, Spain; and T ansla ional Psychia y G oup, Se ille Biomedical Resea ch
Ins i u e (IBiS)-CSIC, Founda ion o Heal h Resea ch Managemen in Se ille, Se ille, Spain;
Claudio Alemán-Mo illo, Depa men o Medical Physiology and Biophysics, Uni e si y
o Se ille, Se ille, Spain; Na alia Ga cía-San-Ma ín , Depa men o Medical
Physiology and Biophysics, Uni e si y o Se ille, Se ille, Spain; Na halia Ga ido-To es,
Men al Heal h Uni , Vi gen del Rocío Uni e si y Hospi al-CIBERSAM, Se ille, Spain; and
T ansla ional Psychia y G oup, Se ille Biomedical Resea ch Ins i u e (IBiS)-CSIC,
Founda ion o Heal h Resea ch Managemen in Se ille, Se ille, Spain;
Ma ía Alemany-Na a o, T ansla ional Psychia y G oup, Se ille Biomedical Resea ch
Ins i u e (IBiS)-CSIC, Founda ion o Heal h Resea ch Managemen in Se ille, Se ille, Spain;
Richa d A. I. Be hlehem, Depa men o Psychology, Uni e si y o Camb idge,
Camb idge, UK; Lena Do schmid , Depa men o Psychology, Uni e si y o Camb idge,
Camb idge, UK; Jakob Seidli z, Depa men o Child and Adolescen Psychia y and
Beha io al Science, The Child en’s Hospi al o Philadelphia, Philadelphia, Pennsyl ania,
USA; Li espan B ain Ins i u e, The Child en’s Hospi al o Philadelphia and Penn Medicine,
Philadelphia, Pennsyl ania, USA; Depa men o Psychia y, Uni e si y o Pennsyl ania,
Philadelphia, Pennsyl ania, USA; and Ins i u e o T ansla ional Medicine and The apeu ics,
Uni e si y o Pennsyl ania, Philadelphia, Pennsyl ania, USA; Rosa Ayesa-A iola,
Depa men o Psychia y, Ma qués de Valdecilla Uni e si y Hospi al, IDIVAL, CIBERSAM,
School o Medicine, Uni e si y o Can ab ia, San ande , Spain; Ja ie Vázquez-Bou gon,
Depa men o Psychia y, Ma qués de Valdecilla Uni e si y Hospi al, IDIVAL, CIBERSAM,
School o Medicine, Uni e si y o Can ab ia, San ande , Spain; Miguel Ruiz-Veguilla,
Men al Heal h Uni , Vi gen del Rocío Uni e si y Hospi al-CIBERSAM, Se ille, Spain; and
T ansla ional Psychia y G oup, Se ille Biomedical Resea ch Ins i u e (IBiS)-CSIC,
Founda ion o Heal h Resea ch Managemen in Se ille, Se ille, Spain; Li espan B ain
Cha Conso ium;Benedic o C espo-Faco o, Men al Heal h Uni , Vi gen del Rocío
Uni e si y Hospi al-CIBERSAM, Se ille, Spain; and T ansla ional Psychia y G oup, Se ille
Biomedical Resea ch Ins i u e (IBiS)-CSIC, Founda ion o Heal h Resea ch Managemen in
Se ille, Se ille, Spain; Ra ael Rome o-Ga cía , Ins i u o de Biomedicina de Se illa
(IBiS), HUVR/CSIC/Uni e si y o Se ille/CIBERSAM, ISCIII, Se ille, Spain; Depa men o
Psychia y, Uni e si y o Camb idge, Camb idge, UK; and Depa men o Medical
Physiology and Biophysics, Uni e si y o Se ille, Se ille, Spain
Co espondence: Ra ael Rome o-Ga cía. Email: [email p o ec ed]
Fi s ecei ed 30 Oc 2024, inal e ision 12 Feb 2025, accep ed 15 Ma 2025
Supplemen a y ma e ial
The supplemen a y ma e ial is a ailable online a h ps://doi.o g/10.1192/bjp.2025.105
Da a a ailabili y
All code and non-clinical da a used o pe o m he analyses can be ound a h ps://gi hub.com/
Ra aelRome oGa cia/Cen ilesP edic o sSCZOu comes. Da a om pa ien s a e a ailable upon
eques .
Acknowledgemen s
We hank Vic o O iz o his con ibu ion. Membe s o he Li espan B ain Cha Conso ium:
C. Adamson, S. Adle , A.F. Alexande -Bloch, E. Anagnos ou, K.M. Ande son, A. A eces-
Gonzalez, D.E. As le, B. Auyeung, M. Ayub, J.B. Bae, G. Ball, S. Ba on-Cohen, R. Bea e,
S.A. Bed o d, V. Benegal, R.A.I. Be hlehem, F. Beye , J. Blange o, M. Blesa Cábez,
J.P. Boa dman, M. Bo zage, J.F. Bosch-Baya d, N. Bou ke, E.T. Bullmo e, V.D. Calhoun,
M.M. Chak a a y, C. Chen, C. Che a ian, G. Che ela , Y.S. Chong, A. Co in, M. Cos an ino,
E. Cou chesne, F. C i ello, V.L. C opley, J. C osbie, N. C ossley, M. Dela ue, R. Delo me,
S. Des i ie es, G. De enyi, M.A. Di Biase, R. Dolan, K.A. Donald, G. Donohoe, L. Do schmid ,
K. Dunlop, A.D. Edwa ds, J.T. Elison, C.T. Ellis, J.A. Elman, L. Eyle , D.A. Fai , P.C. Fle che ,
P. Fonagy, C.E. F anz, L. Galan-Ga cia, A. Gholipou , J. Giedd, J.H. Gilmo e, D. C. Glahn,
I.M. Goodye , P.E. G an , N.A. G oenewold, S. Gudapa i, F.M. Gunning, R.C. Gu , R.E. Gu ,
C.F. Hammill, O. Hansson, T. Hedden, A. Heinz, R.N. Henson, K. Heue , J. Hoa e, B. Holla,
A.J. Holmes, H. Huang, J. Ipse , C.R. Jack J , A.P. Jackowski, T. Jia, D.T. Jones, P.B. Jones,
R.S. Kahn, H. Ka lsson, L. Ka lsson, R. Kawashima, E.A. Kelley, S. Ke n, K. Kim, M.G. Ki zbichle ,
W.S. K emen, F. Lalonde, B. Landeau, J. Le ch, J.D. Lewis, J. Li, W. Liao, C. Lis on,
M.V. Lomba do, J. L , T.T. Malla d, M. Ma celis, S.R. Ma hias, B. Mazoye , P. McGui e,
M.J. Meaney, A. Mechelli, B. Misic, S.E. Mo gan, D. Mo he sill, C. O inau, R. Ossenkoppele,
M. Ouyang, L. Palaniyappan, L. Paly, P.M. Pan, C. Pan elis, M.T.M. Pa k, T. Paus, Z. Pauso a,
D. Paz-Lina es, A. Piche Bine e, K. Pie ce, X. Qian, A. Qiu, A. Raznahan, T. Ri man,
A. Rod igue, C.K. Rollins, R. Rome o-Ga cia, L. Ronan, M.D. Rosenbe g, D.H. Rowi ch,
G.A. Salum, T.D. Sa e hwai e, H.L. Schaa e, J. Schabdach, R.J. Schacha , M. Schöll,
A.P. Schul z, J. Seidli z, D. Sha p, R.T. Shinoha a, I. Skoog, C.D. Smyse , R.A. Spe ling,
D.J. S ein, A. S olicyn, J. Suckling, G. Sulli an, B. Thy eau, R. To o, N. T au , K.A. Ts e ano ,
N.B. Tu k-B owne, J.J. Tuula i, C. Tzou io, É. Vachon-P esseau, M.J. Valdes-Sosa, P.A. Valdes-
Sosa, S.L. Valk, T. an Amels oo , S.N. Vandeka , L. Vasung, P.E. Vé es, L.W. Vic o ia,
S. Villeneu e, A. Vill inge , J.W. Vogel, K. Wags yl, Y.S.S. Wang, S.K. Wa ield, V. Wa ie ,
E. Wes man, M.L. Wes wa e , H.C. Whalley, S.R. Whi e, A.V. Wi e, N. Yang, B.T.T. Yeo, H.J. Yun,
A. Zalesky, H.J. Za , A. Ze e g en, J.H. Zhou, H. Ziauddeen, D. Zimme man, A. Zugman,
X.N.N. Zuo.
Au ho con ibu ions
M.M.-C. pe o med da a cu a ion, me hodological design, da a analysis and d a ed he
manusc ip . C.A.-M., N.G.-S.-M., N.G.-T., M.A.-N., R.A.I.B., L.D., J.S., R.A.-A., J.V.-B., M.R.-V.,
LBCC, B.C.-F. and R.R.-G. con ibu ed o da a acquisi ion, p o ided ad ice on da a analysis and
pa icipa ed in w i ing and edi ing he manusc ip . R.R.-G. and B.C.-F. also con ibu ed o
concep ualisa ion and supe ision o he wo k.
Funding
M.M.-C. is unded by Ins i u o de Salud Ca los III (no. CM23/00263), Acci ´on Es a égica en
Salud, Spanish Go e nmen . R.R.-G. is unded by he EMERGIA Jun a de Andalucía p og amme
(no. EMERGIA20_00139), Plan P opio o he Uni e si y o Se ille and Plan de Consolidaci ´on (no.
CNS2023-143647). Bo h R.R.-G. and C.A.-M. a e unded by Plan de Gene aci ´on de
Conocimien o om Agencia Es a al de In es igaci´on (no. PID2021-122853OA-I00).
Decla a ion o in e es
R.A.I.B. and J.S. hold equi y in, and a e di ec o s o , Cen ile Bioscience. All au ho s decla e no
o he con lic s o in e es .
Re e ences
1Lally J, Ajnakina O, S ubbs B, Cullinane M, Mu phy KC, Gaugh an F, e al.
Remission and eco e y om i s -episode psychosis in adul s: sys ema ic
Muñoz-Ca acuel e al
8
h ps://doi.o g/10.1192/bjp.2025.105 Published online by Camb idge Uni e si y P ess
e iew and me a-analysis o long- e m ou come s udies. B J Psychia y 2017;
211: 350–8.
2Ca alan A, Rich e A, Salaza de Pablo G, Vaque izo-Se ano J, Mancebo G,
Ped uzo B, e al. P opo ion and p edic o s o emission and eco e y in i s -
episode psychosis: sys ema ic e iew and me a-analysis. Eu Psychia y 2021;
64: e69.
3Ayesa-A iola R, O íz-Ga cía de la Foz V, Ma ínez-Ga cía O, Se ién-Sue o E,
Ramí ez ML, Suá ez-Pinilla P, e al. Dissec ing he unc ional ou comes o i s
episode schizoph enia spec um diso de s: a 10-yea ollow-up s udy in he
PAFIP coho . Psychol Med 2021; 51: 264–77.
4Aus in SF, Mo s O, Bud z-Jø gensen E, Seche RG, Hjo høj CR, Be elsen M,
e al. Long- e m ajec o ies o posi i e and nega i e symp oms in i s
episode psychosis: a 10 yea ollow-up s udy in he OPUS coho . Schizoph
Res 2015; 168:84–91.
5Mayo al- an Son J, Juncal-Ruiz M, O iz-Ga cía de la Foz V, Vázquez-
Bou gon J, Se ién-Sue o E, To desillas-Gu ié ez D, e al. Long- e m clinical
and unc ional ou come a e an ipsycho ic discon inua ion in ea ly phases
o non-a ec i e psychosis: esul s om he PAFIP-10 coho . Schizoph Res
2021; 232:28–30.
6Canal-Ri e o M, Ruiz-Veguilla M, O iz-Ga cía de la Foz V, L ´opez-Díaz A,
Ga ido-To es N, Ayesa-A iola R, e al. Longi udinal ajec o ies in nega i e
symp oms and changes in b ain co ical hickness: 10-yea ollow-up s udy.
B J Psychia y 2023; 223: 309–18.
7Dazzan P, A ango C, Fleischacke W, Galde isi S, Glen hoj B, Leuch S, e al.
Magne ic esonance imaging and he p edic ion o ou come in i s -episode
schizoph enia: a e iew o cu en e idence and di ec ions o u u e
esea ch. Schizoph Bull 2015; 41: 574–83.
8Segal A, Pa kes L, Aquino K, Kia SM, Wol e s T, F anke B, e al. Regional, ci cui
and ne wo k he e ogenei y o b ain abno mali ies in psychia ic diso de s. Na
Neu osci 2023; 26: 1613–29.
9Remiszewski N, B yan JE, Ru he o d SE, Ma quand AF, Nelson E, Aska I, e al.
Con as ing case-con ol and no ma i e e e ence app oaches o cap u e
clinically ele an s uc u al b ain abno mali ies in pa ien s wi h i s -episode
psychosis who a e an ipsycho ic nai e. JAMA Psychia y 2022; 79: 11338.
10 Be hlehem RAI, Seidli z J, Whi e SR, Vogel JW, Ande son KM, Adamson C, e al.
B ain cha s o he human li espan. Na u e 2022; 604: 525–33.
11 Ga cía-San-Ma ín N, Be hlehem RAI, Mihalik A, Seidli z J, Sebenius I, Alemán-
Mo illo C, e al. Molecula and mic o-a chi ec u al mapping o g ay ma e
al e a ions in psychosis. Mol Psychia y 2025; 30: 1287–96.
12 Wo ke A, Be he P, Law ence AJ, Kia SM, A ango C, Dinga R, e al. Ex eme
de ia ions om he no ma i e model e eal co ical he e ogenei y and
associa ions wi h nega i e symp om se e i y in i s -episode psychosis om
he OPTiMiSE and GAP s udies. T ansl Psychia y 2023; 13: 373.
13 Be he P, Haa ei BC, Kjelkenes R, Wo ke A, Kia SM, Wol e s T, e al.
A 10-yea longi udinal s udy o b ain co ical hickness in people wi h i s -
episode psychosis using no ma i e models. Schizoph Bull 2024; 51:95–107.
14 Wol e s T, Doan NT, Kau mann T, Alnæs D, Mobe ge T, Aga z I, e al. Mapping
he he e ogeneous pheno ype o schizoph enia and bipola diso de using
no ma i e models. JAMA Psychia y 2018; 75: 114655.
15 Dazzan P. Disen angling psychosis: he challenges o in o ming p ecision
medicine o wha is no a single diso de . Psychia y Res 2023; 330: 115596.
16 Jollans L, Whelan R. The clinical added alue o imaging: a pe spec i e om
ou come p edic ion. Biol Psychia y Cogn Neu osci Neu oimaging 2016; 1:
423–32.
17 Pelayo‐Te án JM, Pé ez‐Iglesias R, Ramí ez‐Bonilla ML, González‐Blanch C,
Ma ínez‐Ga cía O, Pa do‐Ga cía G, e al. Epidemiological ac o s associa ed
wi h ea ed incidence o i s -episode non-a ec i e psychosis in Can ab ia:
insigh s om he Clinical P og amme on Ea ly Phases o Psychosis. Ea ly
In e Psychia y 2008; 2: 178–87.
18 La i iè e S, Paquola C, Pa k B-Y, Roye J, Wang Y, Benka im O, e al. The
ENIGMA oolbox: mul iscale neu al con ex ualiza ion o mul isi e neu oimag-
ing da ase s. Na Me hods 2021; 18: 698–700.
19 And easen NC. Scale o he assessmen o nega i e symp oms (SANS). APA
PsycNe 1982. A ailable om: h ps://doi.o g/10.1037/ 12696-000.
20 And easen NC. Scale o he assessmen o posi i e symp oms (SAPS). APA
PsycNe 1986. A ailable om: h ps://doi.o g/10.1037/ 48377-000.
21 Woods SW. Chlo p omazine equi alen doses o he newe a ypical
an ipsycho ics. J Clin Psychia y 2003; 64: 663–7.
22 Thompson PM, Jahanshad N, Ching CRK, Salminen LE, Thomopoulos SI, B igh
J, e al. ENIGMA and global neu oscience: a decade o la ge-scale s udies o
he b ain in heal h and disease ac oss mo e han 40 coun ies. T ansl
Psychia y 2020; 10: 100.
23 Jiang Y, Palaniyappan L, Luo C, Chang X, Zhang J, Tang Y, e al. Neu oimaging
epicen e s as po en ial si es o onse o he neu oana omical pa hology in
schizoph enia. Sci Ad 2024; 10: eadk6063.
24 Jiang Y, Wang J, Zhou E, Palaniyappan L, Luo C, Ji G, e al. Neu oimaging
bioma ke s de ine neu ophysiological sub ypes wi h dis inc ajec o ies in
schizoph enia. Na Men Heal h 2023; 13: 186–99.
25 Shephe d AM, Lau ens KR, Ma heson SL, Ca VJ, G een MJ. Sys ema ic me a-
e iew and quali y assessmen o he s uc u al b ain al e a ions in
schizoph enia. Neu osci Biobeha Re 2012; 36: 1342–56.
26 Smucny J, Dienel SJ, Lewis DA, Ca e CS. Mechanisms unde lying do sola e al
p e on al co ex con ibu ions o cogni i e dys unc ion in schizoph enia.
Neu opsychopha macology 2021; 471: 292–308.
27 Volpe U, Mucci A, Qua an elli M, Galde isi S, Maj M. Do sola e al p e on al
co ex olume in pa ien s wi h de ici o nonde ici schizoph enia. P og
Neu opsychopha macol Biol Psychia y 2012; 37: 264–9.
28 Kani AS, Shinn AK, Lewandowski KE, Öngü D. Con e ging e ec s o di e se
ea men modali ies on on al co ex in schizoph enia: a e iew o
longi udinal unc ional magne ic esonance imaging s udies. J Psychia Res
2017; 84: 256–76.
29 Okada N, Fukunaga M, Miu a K, Nemo o K, Ma sumo o J, Hashimo o N, e al.
Subco ical olume ic al e a ions in ou majo psychia ic diso de s: a mega-
analysis s udy o 5604 subjec s and a olume ic da a-d i en app oach o
classi ica ion. Mol Psychia y 2023; 28: 5206–16.
30 Sabb FW, an E p TGM, Ha d ME, Dap e o M, Caplan R, Cannon TD, e al.
Language ne wo k dys unc ion as a p edic o o ou come in you h a clinical
high isk o psychosis. Schizoph Res 2010; 116: 173–83.
31 Ki schne M, Aleman A, Kaise S. Seconda y nega i e symp oms: a e iew o
mechanisms, assessmen and ea men . Schizoph Res 2017; 186:29–38.
32 Fe aha G, Takeuchi H, Lee J, Foussias G, Fle che PJ, Agid O, e al.
An ipsycho ics and amo i a ion. Neu opsychopha macology 2015; 406:
1539–48.
33 Ki kpa ick B, Fen on WS, Ca pen e WT, Ma de SR. The NIMH-MATRICS
consensus s a emen on nega i e symp oms. Schizoph Bull 2006; 32: 214–9.
34 Ma de SR, Galde isi S. The cu en concep ualiza ion o nega i e symp oms in
schizoph enia. Wo ld Psychia y 2017; 16:14–24.
35 Zipu sky RB, Reilly TJ, Mu ay RM. The my h o schizoph enia as a p og essi e
b ain disease. Schizoph Bull 2013; 39: 1363–72.
36 Co ell CU, Schoole NR. Nega i e symp oms in schizoph enia: a e iew
and clinical guide o ecogni ion, assessmen , and ea men . Neu opsychia
Dis T ea 2020; 16: 519–34.
37 Lang FU, Kös e s M, Lang S, Becke T, Jäge M. Psychopa hological long- e m
ou come o schizoph enia: a e iew. Ac a Psychia Scand 2013; 127: 173–82.
38 Tandon R, Nas allah H, Akba ian S, Ca pen e WT, DeLisi LE, Gaebel W, e al.
The schizoph enia synd ome, ci ca 2024: wha we know and how ha in o ms
i s na u e. Schizoph Res 2024; 264:1–28.
39 Hä ne H. Gende di e ences in schizoph enia. Psychoneu oendoc inology
2003; 28:17–54.
40 Bes MW, Law H, Pyle M, Mo ison AP. Rela ionships be ween psychia ic
symp oms, unc ioning and pe sonal eco e y in psychosis. Schizoph Res
2020; 223: 112–8.
Volume ic b ain de ia ions in FEP p edic ou come ajec o ies
9
h ps://doi.o g/10.1192/bjp.2025.105 Published online by Camb idge Uni e si y P ess
