Impact of atypical antipsychotics on cytokine profiles in schizophrenia: An 8-week longitudinal study

Abstract

BACKGROUND:

Schizophrenia has traditionally been conceptualized through dopaminergic and glutamatergic frameworks. In recent years, attention has increasingly focused on the role of immune dysregulation in its pathophysiology, although conclusive evidence remains limited. In this context, the present study aimed to investigate alterations in cytokine profiles and their association with clinical symptomatology following treatment with atypical antipsychotics for 8 weeks.

MATERIALS AND METHODS:

Thirty patients aged 18–65 years, diagnosed with schizophrenia as per Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition criteria, were enrolled. Participants received atypical antipsychotic therapy as clinically indicated. Symptom severity was assessed using the Scale for the Assessment of Positive Symptoms (SAPS) and the Scale for the Assessment of Negative Symptoms (SANS). Serum levels of tumor necrosis factor-alpha (TNF-α), interleukin-8 (IL-8), and regulated upon activation, normal T-cell expressed and secreted (RANTES) were measured at baseline and after 8 weeks of therapy.

RESULTS:

Median SAPS and SANS scores significantly decreased from 39 to 30.5 (P < 0.001) and from 30 to 28 (P < 0.001), respectively. Correspondingly, median serum TNF-α levels declined from 28.55 (6.51–227) to 19.98 (4.56–145) pg/mL (P = 0.033). Reductions were also observed for IL-8 (62.2–50.25 pg/mL; P = 0.171) and RANTES (520–500 pg/mL; P = 0.758).

CONCLUSION:

Treatment with atypical antipsychotics significantly attenuated both positive and negative symptoms and reduced TNF-α levels in patients with schizophrenia. The concomitant decline in cytokine concentrations suggests a potential immunomodulatory effect of treatment, highlighting TNF-α, IL-8, and RANTES as possible prognostic biomarkers.

Introduction

Schizophrenia has long been associated with dysregulation of dopaminergic and glutamatergic neurotransmission.[1,2] In recent years, accumulating evidence has highlighted the role of subclinical inflammation and immune dysregulation as important pathophysiological mechanisms in the disorder. Patients with schizophrenia exhibit increased morbidity and mortality from various comorbid conditions, particularly cardiovascular and autoimmune diseases.[3] Individuals with psychosis are also predisposed to weight gain and cardiometabolic abnormalities, which adversely affect physical health and life expectancy.[4] While these risks are partly influenced by lifestyle factors, antipsychotic medications themselves contribute significantly to these metabolic disturbances.[5] Notably, individuals with first-episode psychosis are especially vulnerable to the adverse metabolic consequences of antipsychotic therapy, with unhealthy metabolic changes reported within 6–8 weeks of treatment initiation.[6]

Antipsychotics remain the cornerstone of schizophrenia management, offering robust symptomatic relief; however, their disease-modifying potential continues to be uncertain.[3] Increasingly, cytokines, key mediators of inflammation and immune signaling, have been implicated in the neurobiology of schizophrenia. Cytokines are broadly categorized into pro-and anti-inflammatory mediators based on their structural and functional properties, whereas chemokines represent a specific subgroup responsible for the regulation of cell migration.[7] Cytokines facilitate communication between the immune and nervous systems, shaping neuronal responses to various stimuli and stressors.[8] Elevated levels of C-reactive protein, interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-α), and soluble IL-2 receptor have been documented in patients with schizophrenia, while treatment with cyclooxygenase-2 inhibitors has been shown to alleviate psychotic symptoms.[9] Despite these findings, studies examining proinflammatory cytokines in schizophrenia have yielded inconsistent results.[10,11] Increased secretion of such cytokines may potentiate immune activation, disrupt hemodynamic regulation, and impair metabolic control, linking inflammation to broader systemic dysregulation.

In this context, the present study was designed to examine the influence of antipsychotic therapy on peripheral cytokine levels in patients with schizophrenia. The cytokines selected for evaluation were chosen based on prior evidence suggesting their involvement in the illness. TNF-α, a prototypical proinflammatory cytokine widely studied across various diseases, is primarily secreted by antigen-presenting cells and plays a central role in immune activation.[12] Chemokines, another key group within the cytokine network, are structurally related molecules that bind to G protein-coupled receptors and orchestrate leukocyte trafficking. Over 40 chemokines have been identified in humans, each playing a critical role in host defence and immune regulation.[13] Among them, CCL5, which is also known as regulated upon activation, normal T-cell expressed and secreted (RANTES), is a potent chemoattractant for T cells, eosinophils, and basophils, mediating leukocyte recruitment to inflammatory sites and promoting natural killer cell activation through cytokine-dependent mechanisms.[4] IL-8 or CXCL8, another cytokine evaluated in this study, is produced by macrophages and various structural cells, including epithelial and endothelial cells. It serves dual functions, i.e. inducing chemotaxis of neutrophils and granulocytes, and directing them toward sites of infection or inflammation.[5]

Despite these established roles, the extent to which cytokine levels are modulated by atypical antipsychotic therapy remains insufficiently explored. This study, therefore, aimed to assess the effects of 8 weeks of atypical antipsychotic treatment on serum levels of TNF-α, IL-8, and RANTES in patients with schizophrenia. The 8-week study duration is well-justified based on established clinical, metabolic, and immunological timelines in schizophrenia treatment. This timeframe captures early treatment response, with approximately 40% of first-episode patients demonstrating clinical improvement by week 8.[14] Critically, first-episode psychosis patients exhibit metabolic changes within 6–8 weeks of antipsychotic initiation, making this duration optimal for detecting both immunomodulatory and metabolic effects.[15] The timeline is consistent with existing cytokine research in schizophrenia, where meta-analyses typically include studies with mean treatment durations of 53–60 days, and specific investigations of TNF-α modulation have used 40–60 day protocols. Studies examining cytokines at 3 weeks found no significant changes, suggesting 8 weeks provides sufficient antipsychotic exposure to detect immunomodulation while avoiding confounding long-term metabolic complications that emerge after several months. Operationally, this duration represented a pragmatic balance that successfully minimized attrition during the COVID-19 pandemic constraints.

Materials and Methods

Study design

The present study was a prospective longitudinal study conducted in a single tertiary care center in the All India Institute of Medical Sciences, New Delhi, India. Written approval from the Institute Ethics Committee (IECPG-674/19.12.2019, RT-04/30.01.2020) was obtained before commencement of the study. The study was carried out in accordance with the National ethical guidelines for biomedical and health research involving human participants (2017). Written informed consent was obtained from the participants before enrollment. For those patients who were not able to provide consent due to the nature of their illness, informed consent was sought from the legally authorized representative of the patient.

Study population

Patients aged 18–65 years of either sex with a confirmed diagnosis of schizophrenia, as per the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria, were screened for enrollment. Individuals receiving a stable dose of atypical antipsychotics for at least 4 weeks before recruitment were included. Patients were excluded if they had severe neurological illness or major psychiatric comorbidities that could confound biomarker assessment, including schizoaffective disorder, mood disorders, or substance dependence syndrome. Those unwilling to provide informed consent, pregnant or lactating women, and individuals with deranged renal or hepatic function tests, abnormal lipid profiles, or altered blood counts were also excluded. Additional exclusion criteria included the presence of any significant medical condition, such as endocrine, immune, or metabolic disorders (e.g., diabetes mellitus, autoimmune diseases, inflammatory bowel disease, or acquired immunodeficiency syndrome), as well as any recent history of infectious, allergic, or inflammatory illness.

Study procedure

After enrollment, a detailed clinical history and comprehensive psychopathological evaluation were conducted using the Scale for the Assessment of Positive Symptoms (SAPS) and the Scale for the Assessment of Negative Symptoms (SANS) to quantify symptom severity. Venous blood samples were collected at baseline to estimate serum concentrations of TNF-α, IL-8, and RANTES. All participants continued to receive standard atypical antipsychotic therapy as prescribed by their treating psychiatrists throughout the study period. Following an 8-week treatment period, participants were reassessed clinically and biochemically, with repeat SAPS and SANS evaluations and biomarker measurements performed under identical conditions.

Outcome measures

Clinical outcome measures

The SAPS was utilized for comprehensive evaluation and global rating of hallucinations, delusions, positive formal thought disorder, and bizarre behavior. The SANS was employed to assess five domains of negative symptomatology, namely alogia, affective blunting, avolition/apathy, anhedonia/asociality, and attentional impairment. SAPS comprises 34 items and SANS 25 items, each rated on a 6-point scale (0–5). Composite scores for SAPS (range: 0–170) and SANS (range: 0–125) were computed separately, along with individual domain scores to quantify symptom severity.

Biochemical outcome measures

Serum concentrations of TNF-α, IL-8, and RANTES were estimated. Five milliliters of venous blood were collected and allowed to clot at room temperature before centrifugation to separate serum, which was aliquoted into sterile microcentrifuge tubes. The serum samples were stored at −80°C until analysis. Biomarker estimation was carried out using commercially available enzyme-linked immunosorbent assay (ELISA) kits (Elabscience, USA) based on the standard sandwich ELISA principle, following the manufacturer’s protocol. All samples and standards were assayed in duplicate, and the mean of the two readings was used for statistical analysis.

Safety assessment

Participants had unrestricted access to investigators throughout the study to report any adverse events or concerns. Safety monitoring was conducted through nondirective questioning at each study visit to document and address any treatment-related complications.

Statistical analysis

Continuous variables were expressed as median (range), whereas categorical variables were summarized as percentages. Differences in median values of continuous, nonparametric variables between baseline and follow-up were assessed using the Wilcoxon matched-pairs test. Kendall’s Tau-b correlation coefficient was computed to examine associations between clinical and biochemical outcome measures. All statistical analyses were performed using IBM SPSS Statistics for Windows, Version 26.0 (IBM Corp., Armonk, NY, USA). A two-tailed P < 0.05 was considered statistically significant.

The study sample was restricted to 30 participants owing to operational constraints imposed by the COVID-19 pandemic and subsequent surges associated with the Delta and Omicron variants. Post-hoc power analysis was conducted, and it showed that our study achieved a power of only 74% for the primary outcome measure of change in TNF-α levels and thus remains underpowered.

Results

Participant screening commenced in August 2021 following the decline of the Delta variant wave. The first participant was enrolled in October 2021, and the follow-up phase concluded in September 2022. A total of 37 patients were screened, of whom 4 did not meet the inclusion criteria, and 3 were lost to follow-up or declined further participation. The remaining 30 participants constituted the final study cohort. The flow of participants through the study is illustrated in Figure 1. The mean age of the study population was 39 years, and the male-to-female ratio was 3:2. Table 1 presents the baseline demographic and clinical characteristics of the enrolled participants.

Figure 1.

Study flow diagram

Table 1.

Baseline characteristics of patients

Parameter	Population
Sample size	30
Age (years)	38.87±11.45
Sex (male/female)	18/12
Marital status (married/never married)	11/19
Education (Illiterate/Literate/Primary/Matric/Graduate/Postgraduate)	0/4/2/14/9/1
Employment in the last 30 days (Employed/Unemployed)	21/9
Occupation (Professional/Skilled/Unskilled/Unemployed/Housewife/Retired)	1/10/3/9/6/1
Age of illness onset (years)	28±9.46
Total duration of illness (years)	9 (0–20)
SAPS	39 (9–95)
SANS	30 (2–90)
Serum TNF-α (pg/mL)	28.55 (6.51–227)
Serum interleukin-8 (pg/mL)	62.20 (8.37–303.13)
Serum RANTES (pg/mL)	520 (49–1930)

SAPS=Scale for the assessment of positive symptoms, SANS=Scale for the assessment of negative symptoms, TNF-α=Tumor necrosis factor-alpha, RANTES=Regulated upon activation, normal T-cell expressed and secreted

Outcome measures

Change in disease psychopathology

The median total SAPS scores decreased from 39 to 30.5 at follow-up after 8 weeks of atypical antipsychotic treatment, which was statistically significant (P < 0.001). Total SAPS showed a reduction by 21%. The median score for total SANS significantly (P < 0.001) decreased from 30 to 28 in the study group after 8 weeks of therapy with atypical antipsychotics. Total SANS showed a reduction of 6.67% [Figure 2a].

Figure 2.

Change in outcome parameters following antipsychotic therapy. (a) Clinical outcomes. (b) Biomarkers. SAPS = Scale for assessment of positive symptoms, SANS = Scale for assessment of negative symptoms, RANTES = Regulated upon activation = normal T-cell expressed and secreted, IL-8 = Interleukin-8, TNF-α = Tumor necrosis factor-alpha, pg/mL = picogram/millilitres

Change in tumor necrosis factor-alpha levels

The median serum level of TNF-α decreased from 28.55 pg/mL to 19.98 pg/mL after treatment with atypical antipsychotics for 8 weeks, and this decrease was statistically significant (P = 0.033) [Figure 2b].

Change in serum regulated upon activation, normal T-cell expressed and secreted levels

The median score of serum RANTES levels decreased from 520 pg/mL to 500 pg/mL over 8 weeks after treatment with atypical antipsychotics, but this was not statistically significant (P = 0.758) [Figure 2b].

Change in serum interleukin-8 levels

The median serum levels of serum IL-8 decreased from 62.2 pg/mL to 50.25 pg/mL after treatment with atypical antipsychotics for 8 weeks but was not statistically significant (P = 0.171) [Figure 2b].

Correlation of disease psychopathology with change in biomarker levels

At baseline, no significant correlations were observed between cytokine levels and psychopathology scores [Figure 3]. TNF-α, IL-8, and RANTES showed weak, nonsignificant positive correlations with both SAPS and SANS (p > 0.05).

Figure 3.

Correlation of inflammatory biomarkers (regulated upon activation, normal T-cell expressed and secreted, interleukin-8, and tumor necrosis factor-alpha with (a) Scale for assessment of positive symptoms. (b) Scale for assessment of negative symptoms. SAPS = Scale for assessment of positive symptoms, SANS = Scale for assessment of negative symptoms, RANTES = Regulated upon activation, normal T-cell expressed and secreted, IL-8 = Interleukin-8, TNF-α = Tumor necrosis factor-alpha, pg/mL = picogram/millilitres

Following 8 weeks of treatment, changes in cytokine concentrations did not show significant associations with changes in SAPS or SANS scores [Figure 4]. TNF-α exhibited a weak negative correlation with both SAPS (r = –0.026, P = 0.843) and SANS (r = –0.198, P = 0.146), whereas IL-8 and RANTES demonstrated similarly weak, nonsignificant relationships with clinical improvement.

Figure 4.

Correlation of change from baseline in inflammatory biomarkers (regulated upon activation, normal T-cell expressed and secreted, interleukin-8, and tumor necrosis factor-alpha with change in (a) Scale for assessment of positive symptoms. (b) Scale for assessment of negative symptoms. SAPS = Scale for assessment of positive symptoms, SANS = Scale for assessment of negative symptoms, RANTES = Regulated upon activation, normal T-cell expressed and secreted, IL-8 = Interleukin-8, TNF-α = Tumor necrosis factor-alpha, pg/mL = picogram/milliliters

Discussion

This study was conducted at a tertiary care center in India. The mean age of participants was 39 years, with mean ages of 41 years among females and 37 years among males. The mean age at illness onset was 28 years. Our baseline median TNF-α level (28.55 pg/mL) is consistent with published reports of elevated TNF-α in untreated schizophrenia patients.[11,12] This suggests our cohort had the anticipated inflammatory phenotype. Following 8 weeks of treatment with atypical antipsychotics, SAPS and SANS scores decreased by 21% and 6.7%, respectively. Chaudhury and Kumar examined the efficacy of amisulpride and olanzapine for negative symptoms and cognitive impairment, reporting SAPS improvements of 13% and 30%, and SANS improvements of 74% and 68%, respectively.[6] Approximately half of our participants received olanzapine, demonstrating improvements of 20% (SAPS) and 13% (SANS). Discrepancies between findings may reflect our smaller sample size and differing inclusion criteria, notably the exclusion of treatment-naïve patients in their study.

Another study by Erbaǧci et al. assessing psychopathology at baseline and after 60 days of risperidone treatment reported no significant differences in IL-8 and TNF-α levels between patients and controls, although baseline TNF-α concentrations were associated with Brief Psychiatric Rating Scale and SAPS scores. To our knowledge, no prior study has correlated SAPS and SANS scores with changes in circulating cytokine levels. In our cohort, changes in TNF-α showed trends toward an inverse correlation with improvements in SAPS and SANS, whereas decreases in IL-8 showed a weak positive correlation. These findings partially align with Momtazmanesh et al., who observed associations between negative symptoms and increased TNF-α and IL-8 concentrations.[15]

Frydecka et al. identified elevated IL-8, TNF-α, and RANTES levels in multiepisode schizophrenia compared with healthy controls.[12] Consistent with this, our study demonstrated a post-treatment decline in TNF-α levels after 8 weeks of therapy. Similar results were reported by Ajami et al., where TNF-α levels significantly decreased following 40 days of risperidone or clozapine therapy.[16] In contrast, Song et al. monitored cytokine variations during 6-month risperidone treatment in drug-naïve first-episode patients and observed progressive increases in TNF-α over time.[11] Strzelecki et al. found no TNF-α changes after adjunctive sarcosine therapy despite clinical improvement, suggesting complex cytokine–symptom relationships.[17] While TNF-α levels decreased significantly following treatment, changes in TNF-α concentrations were not significantly correlated with symptom improvement (r = –0.026, P = 0.843), suggesting that TNF-α reduction may reflect a mechanistic effect of antipsychotics independent of immediate clinical symptomatic benefit. Longitudinal biomarker studies with longer follow-up periods are required to determine utility in predicting sustained treatment response.

In the present study, serum IL-8 and RANTES levels also showed downward trends after 8 weeks of atypical antipsychotic treatment, though these changes did not reach statistical significance, likely reflecting the small sample size. Reale et al. reported elevated IL-8 levels in schizophrenia that predicted poorer antipsychotic response, whereas antipsychotic treatment did not significantly alter RANTES levels.[18] Their findings, like ours, indicate that IL-8 and RANTES may not exhibit robust early treatment-related modulation, but their trajectories merit investigation in larger cohorts.

Schizophrenia is increasingly recognized as a multisystem disorder, wherein immune and inflammatory pathways complement dopaminergic and glutamatergic abnormalities in shaping its pathophysiology.[19] Cytokines, as key mediators of immune signaling, are consistently elevated in patients with schizophrenia and contribute to disease mechanisms via several routes. Elevated cytokines increase kynurenic acid, a tryptophan metabolite that antagonizes NMDA receptors and thereby disrupting glutamatergic signaling.[20] They may also impair neurodevelopment and promote neurodegeneration.[21] These processes collectively influence monoaminergic transmission and may underlie negative symptom domains. Our observation of reduced TNF-α without parallel symptom changes suggests that antipsychotics may modulate immune signaling through mechanisms independent of dopamine antagonism. This is consistent with the hypothesis that TNF-α-mediated disruption of glutamatergic and kynurenergic pathways may persist beyond symptomatic improvement, requiring longer follow-up to assess pathway normalization.

While serum IL-8 levels decreased by 19.2% and RANTES by 3.8%, neither change reached statistical significance at the current sample size (n = 30). Post hoc power analysis revealed achieved power of 0.38 (IL-8) and 0.06 (RANTES), substantially below the conventional threshold of 0.80. These null findings reflect insufficient statistical power rather than the absence of a biological effect. IL-8 in particular warrants investigation in a larger prospective cohort (n ≈ 100-120) to characterize treatment-related modulation of this chemokine. Conversely, the negligible RANTES effect (3.8% reduction, r = 0.06) suggests this marker is unlikely to be clinically relevant in antipsychotic response monitoring.

We hypothesized that pro-inflammatory cytokine levels would decline alongside clinical improvement. Consistent with this, all measured cytokines demonstrated decreasing trends paralleling symptomatic improvement. Mechanistically, these reductions may reflect downstream modulation of glutamatergic, developmental, and neurodegenerative pathways. Given the growing therapeutic success of cytokine-modulating agents in autoimmune and oncologic disorders, exploration of similar immunomodulatory strategies in schizophrenia could provide valuable translational insights. Furthermore, cytokines hold promise as potential biomarkers for disease progression and treatment response, warranting further study in larger, longitudinal cohorts.

This study has several important limitations. The modest sample size, constrained by COVID-19 operational challenges, resulted in differential statistical power across outcomes, suggesting non-significant secondary findings reflect statistical limitations rather than the absence of biological effect. The absence of healthy control comparisons precludes the determination of baseline cytokine elevations and posttreatment normalization relative to nonpsychotic individuals. Heterogeneous antipsychotic medications (50% olanzapine and 50% other agents) and nonstandardized dosing preclude drug-specific inference. The brief 8-week follow-up duration may not capture sustained immunomodulation or long-term prognostic significance of cytokine changes. Finally, unmeasured confounders, including lifestyle factors, metabolic parameters, circadian sampling timing, and concurrent medications, may have influenced cytokine levels. Despite these limitations, this pilot study provides preliminary evidence for TNF-α immunomodulation and establishes a foundation for adequately powered longitudinal investigations.

Conclusion

Treatment with atypical antipsychotics significantly reduced SAPS and SANS scores and decreased TNF-α levels in patients with schizophrenia. Whereas changes in TNF-α concentrations were not significantly correlated with symptom improvement, the observed immunomodulation suggests a potential immunoregulatory mechanism independent of dopamine antagonism. Larger longitudinal studies with longer follow-up periods, standardized antipsychotics, and healthy control groups are required to validate TNF-α as a prognostic biomarker and elucidate its role in antipsychotic treatment response.

Author contribution

All authors made a substantial contribution to the conception of the study. A. H., A. M., and M. S. contributed to the acquisition of data. A. H., A. M., and K. R. contributed to data analysis and interpretation. A. H., A. M., and S. C. S. drafted the manuscript. K. R. and M. S. critically revised the manuscript for important intellectual content. All authors approved the final manuscript.

Data availability

The trial data are available on reasonable request from the first or corresponding author.

Ethics approval and consent to participate

The study was carried out according to good clinical practice and the ethical principles of the Declaration of Helsinki. Written informed consent was obtained from participants before enrolment. In patients not able to provide consent due to the nature of their illness, informed consent was sought from the primary caregiver of the patient, and assent was taken from the patient.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.