Psychosis is a devastating condition, and there is growing interest in identifying biomarkers that can predict its occurrence during the prodromal stage, preceding actual onset (Chan et al., 2015; Couttas et al., 2022). For instance, plasma proteomics have been found to predict psychosis with greater accuracy than clinical features (Mongan et al., 2021). Early diagnosis is particularly important for implementing interventions that result in better recovery from the first psychotic episode (McGorry et al., 2007). Based on earlier studies, prodromal, first-episode, and chronic stages of psychosis share similar alterations in lipid and glucose metabolism, which may thus form promising biomarkers (Leppik et al., 2020; O’Gorman et al., 2017).
The physiological phenomena associated with psychotic-like experiences and prodromal psychotic symptoms have been extensively studied, yet a clear consensus remains to be established. For example, lysophosphatidylcholines, lipids found to be altered in the prodromal stages of psychosis, have been implicated in promoting inflammation, which is another system associated with a higher risk of psychosis in adolescence (O’Gorman et al., 2017). Additionally, cAMP signaling, important in the integration of information from neurotransmitter receptors (e.g., glutamatergic, dopaminergic, and GABAergic receptors), has been implicated in the pathophysiology of psychosis (Funk et al., 2012). Recognizing the interconnectedness of various systems and their impact on each other’s functions may unveil layers of development of disorder within the nervous system and at the systemic level. Conversely, subtypes such as autoimmune-related psychosis have been proposed, suggesting diverse mechanisms underlying psychoses (Najjar et al., 2018).
Alterations in the plasma lipidome in children have been observed to precede psychotic-like experiences (PLE) (Madrid-Gambin et al., 2019) and psychosis (O’Gorman et al., 2017) in adolescence, and dysregulated lipid metabolites have been found to predict psychosis in young adults (Dickens et al., 2021; Li et al., 2022). Furthermore, psychotic experiences in early adulthood have been associated with disturbances in lipid metabolism (Yin et al., 2022). In addition, altered lipid levels in red blood cell membranes have been associated with an increased risk of psychosis (Frajerman et al., 2023). Specifically, a disturbed biosynthesis of unsaturated fatty acid pathway and altered triacylglycerol levels have been found in both serum and plasma samples in patients clinically at high risk of psychosis (Dickens et al., 2021; Li et al., 2022; Yin et al., 2022).
Other common findings related to the prodromal stages of psychosis are altered serum and plasma levels of phosphatidylcholines, lysophosphatidylcholines, and sphingomyelins (Dickens et al., 2021; O’Gorman et al., 2017; Yin et al., 2022). In particular, alterations in phosphatidylcholines and lysophosphatidylcholines during childhood have been observed to precede the manifestation of PLE in adolescence (Madrid-Gambin et al., 2019). Similarly, phospho- and sphingolipids have been found to be altered in first-episode psychosis (FEP) when compared to healthy controls (Leppik et al., 2020). Apolipoprotein E, an important protein in cholesterol metabolism, has been present at greater levels in adolescents undergoing persistent psychotic experiences compared to those whose experiences did not persist (Sabherwal et al., 2019). Findings in individuals with an interview-assessed clinical high risk of psychosis (CHR) include altered catecholamine dopamine and noradrenaline metabolite alterations in saliva samples (Cui et al., 2021). Additionally, PLEs have been associated with changes in gene expression, observed as altered DNA methylation (Roberts et al., 2019), as well as alterations in the proteome (Föcking et al., 2021; Sabherwal et al., 2019) in children and adolescents when compared to healthy age-matched controls. To our knowledge, no metabolomic research, except for lipidomic studies, has been conducted in relation to PLE.
While metabolomic changes appear to be associated with the physiological process of psychosis, the chronicity of a disease and medications can also impact the metabolome (Kriisa et al., 2017; Leppik et al., 2020). Therefore, the investigation of unmedicated patients at risk of or in the early stages of a disorder is crucial for a better understanding of the early disease etiology. In order to identify metabolomic changes linked to the initial stages of the psychotic process, we conducted an exploratory study to investigate the associations between PLEs and the metabolome in a cohort of 14–20-year-old, mainly unmedicated depressed psychiatric outpatients.