Prior meta-analyses have shown that patients with psychotic disorders have elevated levels of proinflammatory cytokines in the peripheral blood, which seem to be driven by samples taken when patients have acute psychotic episodes.^{1, 2} However, antipsychotic medication can affect immune markers. Thus, investigations of contrasts in cytokine levels among patients with first episode psychosis who are antipsychotic-naïve are of utmost importance to disentangle the contribution of cytokines to the pathophysiology of psychotic disorders. In their paper in Acta Psychiatrica Scandinavia, Dr. Dunleavy et al³ provide a systematic review and meta-analysis of cytokine levels in the peripheral blood of patients with first episode psychosis who are antipsychotic-naïve. The study finds that blood levels of proinflammatory cytokines are elevated also among patients with first episode psychosis who are antipsychotic-naïve. These elevated cytokine levels seem to be driven by proinflammatory cytokines where IL-6, IL-12, IL-17 and IFN-γ displayed significantly elevated levels in the blood compared to healthy controls. This indicates that contrasts in cytokine levels in patients with psychotic disorders are not only due to effects of anti-psychotic medication or chronicity of the psychotic disorder. These findings are further supported by a prior meta-analysis showing that patients with first episode schizophrenia who were drug-naïve had increased blood levels of IL-6, IL-8 and TNF-α compared to healthy controls.⁴ However, only IL-6 displayed consistent statistically significant increases across the two meta-analysis of drug-naïve patients with first-episode psychosis and schizophrenia, respectively,^{3, 4} and this discrepancy is likely due to power issues and variability between samples and methods. Moreover, increases in levels of proinflammatory cytokines are induced by underlying immune activation, and thus, the observed contrasts in specific cytokine levels depend on when in the inflammatory cascade patients were sampled for the given study.

The systematic review from Dr. Dunleavy et al³ further displays results from a narrative review on the associations between cytokine levels and negative symptoms among patients with first episode psychosis. However, unfortunately they did not have enough data to perform meta-analysis. Nonetheless, statistically significant associations were reported for at least one of the included studies on IL-1β, IL-2, IL-4, IL-6, and TNF-α, whereas decreased levels of the anti-inflammatory cytokine IL-10 were reported to be associated with negative symptoms. Interestingly, the inflammatory biomarker concentrations were also reported to be associated with the severity of negative symptoms. However, the results on associations between cytokine levels and negative symptoms in the study by Dunleavy et al³ should be interpreted with great caution as they were not based on meta-analyses but only a narrative review of the literature. Nonetheless, in support of the findings by Dr. Dunleavy et al,³ a prior study found that peripheral blood levels of IL-8, IL-10 and TNF-α were associated with the severity of negative symptoms observed among patients with psychotic disorders also after correction for multiple testing.⁵ Moreover, studies have found associations between peripheral immune cell populations with the severity of cognitive deficits and negative symptoms among patients with treatment-resistant schizophrenia.⁶ Interestingly, meta-analysis of RCTs with placebo-controlled treatment with anti-inflammatory drugs have shown that anti-inflammatory drugs can improve both psychotic and negative symptoms to some extent.⁷ Meta-analysis of RCTs with immunomodulating treatment for psychotic disorders and depression, respectively, indicates that the effect seem to be larger on depression and depressive symptoms than the effect sizes observed for psychotic disorders. Thus, immunomodulating treatments could provide particular potential for the negative symptoms observed in schizophrenia due to a considerable overlap with depressive symptoms.^{7, 8}

Although concentrations of cytokines in peripheral blood can provide insights into the potential immunological contribution to psychotic disorders, the central nervous system (CNS) is protected by the blood–brain-barrier. Thus, peripheral blood levels of cytokines might not translate directly to changes within the central nervous system. Immunological investigations of the cerebrospinal fluid (CSF) provide better insight into immune processes within the CNS and CSF analyses are the gold standard for investigations of neuroinflammation. Here, meta-analyses of studies investigating CSF of patients with psychotic disorders compared to healthy controls, have found elevated levels of both IL-6 and IL-8 in the CSF, particularly during acute phase of the psychotic disorder.⁹ For CSF analysis of contrasts in cytokine levels, meta-analyses have not yet been able to include analyses on patients with first episode psychosis who are antipsychotic-naïve or specific associations with negative symptoms and cytokine levels, which is needed to further advance the evidence base. Moreover, elevation in cytokine levels are often a product of underlying immune activation, with up- or down regulation of cytokines and chemokines possessing pleiotropic effects depending on the state of immune activation. Thus, longitudinal assessment of contrasts in multiple cytokine levels are of utmost importance together with assessment of other immune markers to disentangle the potential involvement of the immune system in the pathophysiology of psychosis and schizophrenia.

Methodological considerations regarding the studies included in the meta-analysis are of considerable importance for the interpretation of the results, as the meta-analysis will be largely influenced by the quality of the included studies. Moreover, meta-analyses would benefit from quality assessment of the included studies and of the level of evidence obtained by for instance the Newcastle-Ottawa criteria as suggested by Cochrane¹⁰ and by the Grading of Recommendations Assessment, Development and Evaluation (GRADE) instrument.¹¹ Current methods for analyses of cytokines are particularly prone to the influence of considerable differences between plates, which need to be handled analytically but also in the design where samples should be randomized to plates, to secure equal amount of cases and controls on each plate, and case/control status should be blinded to the analyst. Reporting of values above lower limit of detection (LLOD) and lower limit of quantification (LLOQ) in the studies are of further importance for evaluating the accuracy of the cytokine measurements and thus the expected variability. Moreover, confounding factors as the Body Mass Index (BMI), which Dunleavy et al³ also found to be associated with the levels of IFN-γ and IL-6, but also factors as smoking, alcohol use, age and comorbidities need to be accounted for in the analyses of cytokine levels. Combining individual level data from different studies have been suggested as a way forward for the field to account for these factors in meta-analysis. However, when investigating cytokines, ideally all the samples should be re-tested using the same assay and batch, in order to minimize the influence of variability between assays, batches, laboratories, technicians, etc.

Thus, future work should focus on performing large high-quality studies with longitudinal assessments of immune-related biomarkers as cytokine levels to inform on the importance of state and trait of psychotic disorders on the observed associations. Moreover, cytokine levels display considerable fluctuations, and thus might not be an ideal biomarker to base treatment trials on. Thus the research area should have a focus on multimodal analyses of immune markers including sampling from multiple tissues as the cerebrospinal fluid, blood and microbiome with longitudinal assessment of the biological markers and the relationship with clinical symptoms in order to identify more stable biomarker panels associated with psychosis that can be used for stratification.^{12, 13} Preferably, these analyses should be done on cross-disorder cohorts to investigate differences and similarities in the immunological profile between the mental disorders to determine specificity of the findings for the given disorder. Further, this would enable to study associations between immune markers and specific psychiatric symptoms across mental disorders to identify potential cross-disorders subgroups with common biological underpinnings. Examples of this could be specific negative symptoms of schizophrenia and specific depression symptoms as the neurovegative symptoms.^{12, 13} As psychotic disorders display substantial heterogeneity on symptomatology, clinical course and prognosis, biomarkers as immune markers could show potential for stratification of patients, which is currently used in an increasing number of RCTs testing the clinical utility of different stratification methods for psychotic disorders to improve clinical outcomes and treatment response.