Longitudinal Inference of Multiscale Markers in Psychosis: From Hippocampal Centrality to Functional Outcome

Abstract

Psychosis represents a heterogeneous collection of biological and behavioural alterations that evolve over time. We propose a multiscale disease progression model of psychosis, in which hippocampal-cortical dysconnectivity precedes impaired episodic memory and social cognition, worsening negative symptoms and lowering functional outcome. In two cross-sectional datasets of first- and multi-episode psychosis (163 patients; 117 controls), we applied a recently developed machine-learning algorithm, SuStaIn, which uniquely integrates clustering and disease progression modeling. SuStaIn identified three patient subtypes, with Subtype 0 showing normal-range performance on all variables. In comparison, Subtype 1 showed lower episodic memory, social cognition, functional outcome, and higher negative symptoms, while Subtype 2 showed lower hippocampal-cortical connectivity. Subtype 1 deteriorated from (social) cognition to symptoms, functioning and hippocampal-cortical dysconnectivity, while Subtype 2 deteriorated from hippocampal-cortical dysconnectivity to (social) cognition, functioning and symptoms. This first application of SuStaIn in a multiscale model of psychiatry provides distinguishable disease trajectories of hippocampal-cortical connectivity, which might drive heterogeneous behavioural alterations in psychosis.