Polygenic associations with phenotypic classes across the psychosis-affective spectrum

Area of Science:

• Psychiatry
• Genetics
• Data Science

Background:

• Current classifications of schizophrenia, schizoaffective, and bipolar disorders exhibit overlapping symptoms, etiologies, treatments, and outcomes.
• This diagnostic overlap hinders novel treatment discovery and accurate prognostication.
• Alternative conceptualizations are needed to improve nosological validity and align diagnosis with etiology.

Purpose of the Study:

• To identify latent classes across the psychosis spectrum based on premorbid functioning and outcomes.
• To assess these latent classes in relation to genetic liability and symptom dimensions.
• To evaluate if current diagnostic categories explain the associations between polygenic scores and latent classes.

Main Methods:

• Latent class analysis was performed on a dataset of 5,043 participants from four UK clinical cohorts diagnosed with schizophrenia, schizoaffective disorder, or bipolar I disorder.
• Phenotypes used included premorbid functioning, age at illness onset, and measures of illness severity and course.
• Polygenic scores (PGS) for psychiatric disorders and behavioral traits were analyzed for associations with latent classes, and the explanatory role of diagnosis was tested.

Main Results:

• A three-class model best fit the data, differentiating individuals based on premorbid functioning and outcomes.
• Class one exhibited poorer functioning and higher PGS for schizophrenia and ADHD; Class three showed higher functioning and higher intelligence PGS.
• Class two was intermediate in functioning but characterized by high involuntary hospital admissions and high bipolar disorder PGS; diagnosis partially explained PGS-class associations.

Conclusions:

• Latent classes across the psychosis spectrum, defined by premorbid functioning and outcomes, cut across diagnostic boundaries.
• These classes capture genetic liability not fully explained by current diagnoses.
• Findings support alternative conceptualizations of psychotic disorders for advancing precision psychiatry and understanding etiology.