Prolonged childhood and adolescent loneliness in first-episode psychosis: synergistic polygenic effects and functional outcomes

Andreu-Bernabeu, Álvaro, González-Peñas, Javier, Bernardo, Miguel, Amoretti, Silvia, Bobes, Julio, Arrojo, Manuel, Parellada, Mara, O'Donovan, Michael ORCID: https://orcid.org/0000-0001-7073-2379, van Os, Jim, Rutten, Bart P. F., Murray, Robin M, Di Forti, Marta, Vassos, Evangelos, Arango, Celso and Díaz-Caneja, Covadonga M. 2026. Prolonged childhood and adolescent loneliness in first-episode psychosis: synergistic polygenic effects and functional outcomes. Schizophrenia Bulletin 52 (5) , sbag107. 10.1093/schbul/sbag107

Full text not available from this repository.

Abstract

Background Prolonged childhood and adolescent loneliness (CAL) may mark an early socio-emotional vulnerability relevant to psychosis and interact with genetic liability. We examined whether CAL increases odds of first-episode psychosis (FEP), relates to diagnostic subtype and functioning, and interacts with polygenic risk scores (PGS) for schizophrenia (SCZ), bipolar disorder (BIP), and major depression (MDD). Study Methods We analyzed 2565 participants (1084 FEP, 1481 controls) from European Gene–Environment Interactions in Schizophrenia (EU-GEI) study. CAL was assessed retrospectively before age 12 (LNL12) and at 12-16 years (LNL12-16). Functioning was rated with a modified Global Assessment of Functioning (GAF). Genome-wide PGS for SCZ, BIP, and MDD were calculated using polygenic risk score–continuous shrinkage. Mixed-effects logistic and linear regression models tested associations of CAL with FEP, subtype, and functioning, including sex-stratified analyses. Additive gene–environment interaction was examined using dichotomized PGS and the Relative Excess Risk due to Interaction. Study Results Both LNL12 (OR = 2.90, 95% CI, 2.30-3.66) and LNL12-16 (OR = 2.34, 95% CI, 1.94-2.84) were associated with FEP, independent of premorbid social isolation. CAL increased FEP odds in both sexes. LNL12-16, but not LNL12, was associated with affective versus non-affective psychosis (OR = 1.37, 95% CI, 1.02-1.83), more clearly in females (OR = 1.79, 95% CI, 1.16-2.77). CAL was also associated with poorer functioning, with 3-4-point lower GAF Disability scores. Significant additive interactions were observed between LNL12-16 and SCZ-PGS, and between LNL12 and MDD-PGS, although the latter did not replicate in European-ancestry sensitivity analyses. Conclusions Prolonged CAL is linked to higher risk of FEP, preferentially affective psychosis in females, poorer functioning, and developmental potentiation of polygenic liability.

Item Type:	Article
Date Type:	Publication
Status:	In Press
Schools:	Schools > Medicine
Publisher:	Oxford University Press (OUP)
ISSN:	0586-7614
Date of Acceptance:	22 May 2026
Last Modified:	08 Oct 2026 11:16
URI:	https://orca.cardiff.ac.uk/id/eprint/189266

Actions (repository staff only)

Edit Item