The liability-threshold model of psychosis risk predicts stronger phenotypic manifestation of the polygenic risk score (PRS) in the healthy relatives of patients, as compared with healthy comparison subjects.
First-degree relatives of patients with psychotic disorder (871 siblings and 812 parents) and healthy comparison subjects (n = 523) were interviewed three times in 6 years. Repeated measures of two psychosis phenotypes, the Community Assessment of Psychic Experiences (CAPE; self-report - subscales of positive, negative and depressive symptoms) and the Structured Interview for Schizotypy - Revised (SIS-R; clinical interview - subscales of positive and negative schizotypy), were examined for association with PRS. Interview-based lifetime rate of depressive and manic episodes were also examined, as was association with repeated measures of intelligence quotient (IQ).
In the relatives, PRS was associated with CAPE/SIS-R total score (respectively, B = 0.12, 95% CI 0.02-0.22 and B = 0.11, 95% CI 0.02-0.20), the SIS-R positive subscale (B = 0.16, 95% CI 0.04-0.28), the CAPE depression subscale (B = 0.21, 95% CI 0.07-0.34), any lifetime affective episode (OR 3.1, 95% CI 1.04-9.3), but not with IQ (B = -1.8, 95% CI -8.0 to 4.4). In the controls, similar associations were apparent between PRS on the one hand and SIS-R total score, SIS-R positive, SIS-R negative, any lifetime affective episode and, in contrast, lower IQ (B = -8.5, 95% CI -15.5 to -1.6).
In non-ill people, polygenic risk for psychotic disorder is expressed pleiotropically in the domain of neurodevelopment, emotion regulation and attribution of salience. In subjects at elevated genetic risk, emerging expression of neurodevelopmental alterations may create floor effects, obscuring genetic associations.