Biomarkers/Biotypes, Course of Early Psychosis and Specialty Services

This proposal seeks to characterize the early course of psychotic disorders and to identify clinical and biological predictors of outcome. The large sample (=320) required will necessitate a multi-site study. All B-SNIP sites have coordinated specialty services for early course psychosis and have harmonized study procedures for uniform data collection. Each site will recruit 1/5 of the participants; The project will be managed by an all-site steering committee meeting weekly. Boston will be the coordinating site.

The outcome of early course psychosis (EP) is heterogeneous, ranging from early full recovery to treatment resistance and functional decline from the onset of illness. The ability to predict individual level outcomes would be highly valuable for treatment planning and for tailoring the duration and intensity of psychosocial and pharmacological interventions. Clinical features related to early psychosis onset are poor predictors of remission and recovery. At the same time, the field has not yet established the clinical utility of promising findings from decades of biomarker development/neuroscience research, especially in EP patients for whom empirically guided treatment planning may have the greatest impact. For example, neurocognitive, electroencephalographic (EEG) and imaging biomarkers early in the illness may predict outcome, but used individually, their prognostic value is limited. Multivariate approaches to clinical and neurobiological features may offer better value for outcome prediction.

Toward this goal, we will leverage the biomarker (EEG, eye movement testing, and neurocognition) based categorization of a cross-diagnostic sample of psychosis (biotypes) developed, replicated and validated by our Bipolar-Schizophrenia Network for Intermediate Phenotypes (B-SNIP) consortium. We have shown that Biotype-1, characterized by impaired cognition and decreased sensorimotor responses to salient stimuli is associated with the most severe impairments in functioning. Biotype-2 have the same level of cognitive impairment as Biotype-1 but increased sensorimotor reactivity. Biotype-3 cases are relatively normal on these measures and have the least impairments. These subgroups are not synonymous with DSM diagnosis and are not captured by any individual variable. We do not know if this categorization is predictive of symptomatic and functional outcome, given the cross-sectional design of B-SNIP. We will use the B-SNIP battery with EP participants in the context of Coordinated Specialty Care (CSC) treatment programs for EP available at established B-SNIP sites. Our overall goal is to identify biomarkers/Biotypes that predict clinical and functional outcome to CSC in EP. This success could guide targeted interventions in CSC programs. For those unlikely to have a successful CSC outcome, other targeted interventions could be developed and implemented. We have active EP Clinics in the consortium; each will enter about16 patients/year, 80/year overall, providing 320 EP (schizophrenia, schizoaffective disorder, psychotic bipolar disorder and schizophreniform disorder participants with approximately 240 completer cases, assuming 25% attrition) enrolled during the first 4 years of the proposed funding period. All EP cases will be tested with B-SNIP biomarker assessments at baseline, and the clinical and cognition assessments will be repeated at 1, 6, and 12 months. We will use an adapted version of the B-SNIP biomarker battery (called ADEPT) which can be implemented in under-resourced, community settings.

SIGNIFICANCE AND BACKGROUND FOR THE STUDY Schizophrenia and related psychoses cause enormous suffering for affected individual and their families and caregivers. They have a lifetime prevalence of around 3%, and cost ~ $155 billion per year in direct healthcare and indirect societal costs in the USA [5]. Relapse and poor response to treatment contribute to a significant illness burden due to high rates of hospitalization and poor social and occupational functioning. About 20% of early course psychosis (EP) cases relapse in year 1, and about 40% by year 2 [6]. Mortality is increased four-fold in EP compared to the general population [7]. EP outcomes are highly variable [8] with multiple trajectories, i.e. episodic vs persistent, initial remission vs treatment resistance [9-11]. Given this heterogeneity, actionable outcome predictors are needed to reliably individualize care for EP [12].