Longitudinal Brain Imaging and Investigation of Etiological and Outcome Factors in Adolescents with Early Onset Psychosis:continuation study

Adolescent schizophrenia (EOS) and affective spectrum psychosis disorders (early-onset psychosis disorders; EOP) are severe mental disorders, and one of the main causes of disease burden in adolescents. Lifetime prevalence of EOP is 0.05-0.5%, steeply increasing after age 14 years. Treatment is suboptimal due to lack of knowledge of illness mechanisms. The clinical heterogeneity results in lower diagnostic stability and high rates of comorbidity. Disruptions of perception, thought, mood and functioning present in all age groups and genders. Similarities across schizophrenia and affective psychosis appear greater than in their adult counterparts. Adolescence is a critical period for development of psychiatric illness with 50% of lifetime psychopathological burden emerging by the mid-teens, and coincides with major structural brain changes. We have established a comprehensive database of systematically collected research information on EOP and healthy adolescent controls (HC) for current and future scientific investigations. The database contains detailed data on clinical symptoms, demographics, family history, physical and mental symptoms, brain structure and function from magnetic resonance imaging (MRI), cognitive test results, global functioning, immune and lipid markers, and gene variants from DNA. We established a biobank. We follow the participants who are aged between 12 and 18 years over time, initially over 3 years. The database connects with a similar database of adult-onset schizophrenia and bipolar disorders enabling comparison across all ages and disease onsets. We established a similar cohort of EOP at Karolinska Institutet. We coordinate two international multisite EOP networks for brain imaging and cognition. Using MRI, we describe, in 296 EOP patients and 360 HC, an EOP profile with small intracranial volumes, lower hippocampal volumes, and higher caudate and pallidum volumes relative to HC. Intracranial volume was lower in both EOS and affective psychosis, and showed an effect of antipsychotics use with lower intracranial volume in EOS. The findings demonstrate a similar pattern of brain alterations as reported in adult psychosis, but with notably low intracranial volume and larger caudate. We have formed hypotheses about obstetric complications as putative antecedents to brain development in childhood that might lead up to early psychosis development, since, in adult severe mental illness, we find association between birth asphyxia and lower IQ, smaller brain area, and white matter changes. Using the hitherto largest EOP sample (321 EOP patients and 265 HC) pooled from 9 international research sites) for a diffusion imaging study, we revealed widespread disruption of white matter microstructure in EOP relative to HC, most prominently in EOS, in the superior longitudinal fasciculus. We detected effects of sex and illness duration, but not medication status. We used fMRI to compare within-network brain connectivity in the default mode network between 68 Scandinavian EOP patients and 95 HCs. This functional brain network showed weaker connectivity in widespread brain areas in EOP. Affective psychosis patients deviated the most. Antipsychotics use showed no association but Lithium and anticonvulsants use predicted lower connectivity scores. We have obtained normalized cognitive test results in healthy adolescents using the MATRICS Consensus Cognitive Battery to enable standardized comparison of cognitive performance across different clinical populations and ages. The EOP patients performed significantly worse than the healthy reference group in all cognitive domains. Impairments were strongest for processing speed. The study provides a cognitive performance profile in EOP, stratified by age and sex, relative to adolescent standardized scores. A factor analysis of scored negative symptoms corroborated previously described factors of apathy and diminished expression in adult-onset schizophrenia. Our results support the presence of apathy and diminished expression in EOP. For both domains, negative symptom scores were inversely associated with verbal learning. Activation of inflammatory pathways occurs in EOP and we have shown interactions between stress, inflammation and depressive symptoms. Tau protein is necessary for microtubule cytoskeleton and we have demonstrated low plasma tau levels in EOP. Neuron-specific enolase is upregulated after neuronal injury and concentrations are lower in both adult (1000 patients and 900 HC) and adolescents (100 EOP patients and HC). Antibodies against the NMDA-receptor could point to a propensity toward a disrupted glutamate system leading to early psychosis development, but presence of antibodies in EOP patients or HC was not associated with any clinical or radiological features and had no diagnostic implication. A positive NMDAR antibody test in youth should therefore be interpreted with care, and reviewed within the individual clinical context.

The project has not had any direct effects for improving the lives of adolescents with psychosis, but we expect to have increased visibility, awareness and knowledge of this vulnerable patient group. We have contributed to improved diagnostic skills in clinical practice by sharing clinical research protocols and interacting with the clinic. We hope to have helped destigmatizing psychosis in the young and to having made society more aware of the needs of this patient group. We have not solved the question of what causes psychosis, but we have findings that are of interest for future studies. We are part in international research collaborations and expect the scientific community to consider how child and adolescent research can integrate better with adult mental health research. There is a need to increase biological research in child psychiatry and our broad and interdisciplinary approaches can help raise the interest to study this important age group.

Early-onset psychosis (EOP) and particularly early-onset schizophrenia (EOS) are long term devastating mental diseases with unknown cause and suboptimal treatment. EOS is considered to involve more premorbid neurodevelopmental abnormalities, worse anatomic brain abnormalities, higher rate of cytogenetic abnormalities, more rare copy number variants, and worse long-term outcome than what is found in adult-onset disease. It is important to use the improved methodologies of today to acquire new knowledge about the etiology and to find better treatments and prevention. We base the study on adolescent patients (ages 12-18 years) with schizophrenia or bipolar spectrum EOP and healthy controls of the same ages, both genders. The patients are characterized with clinical, neurocognitive, and genetic methods (whole genome scanning and deep sequencing of selected gene regions), magnetic resonance imaging of functional connectivity and brain structure development at different time points during adolescence. We obtain information on family history and early environmental risk factors; e.g. obstetric complications from medical records and birth registry, biomarkers for cardiovascular disease risk, autoimmunity and inflammation, as independent or related factors to brain abnormalities. In collaboration with partners in Bergen we will test smartphone app technology for symptom monitoring, foremost auditory verbal hallucinations, and with partners at Karolinska Institutet we have novel biomarker results. The project is part of the long-term research project agenda of NORMENT: Norwegian Centre for Mental Disorders Research, and will prospectively follow the cohort for study of etiology and outcome prediction. EOP is relatively rare and large samples are needed. The present proposal is a continuation proposal to increase inclusion rate and number of patients, conduct a two-year follow-up and introduce novel methods through collaborative partners in the Scandinavian EOP network.