Schizophrenia (SCZ) shows marked biological heterogeneity, with negative symptoms linked to poor outcomes and hypothesised immune dysregulation. This study examined whether a peripheral cytokine–long non-coding RNA (lncRNA) panel could distinguish patients with SCZ and Brief Negative Symptom Scale (BNSS)-defined subgroups from healthy controls (HC).

Forty-one hospitalised patients with SCZ completed the BNSS and the Positive and Negative Syndrome Scale (PANSS). Twenty HCs, frequency-matched for age and sex, served as comparison samples. Severe negative-symptom subgroups were defined using two BNSS criteria: a broader (SNS1) and a more restrictive (SNS2) threshold. Serum cytokines – interleukin-6 (IL-6), tumour necrosis factor-α (TNF-α), interleukin-10 (IL-10) – and leukocyte lncRNAs (MALAT1, NEAT1, MEG3) were quantified by enzyme-linked immunosorbent assay and quantitative RT-PCR. Covariate-adjusted logistic and multinomial models (adjusting for age, sex, body mass index, and smoking) assessed discrimination using area under the receiver-operating-characteristic curve (AUC) and interquartile-range odds ratios (OR_IQR).

IL-6 correlated with PANSS Total (ρ = 0.48, p = 0.001) and Negative (ρ = 0.34, p = 0.032) scores and was higher in SCZ than HC (p = 0.033), with further increases in SNS subgroups. NEAT1 was significantly reduced only within BNSS-defined subgroups (p ≤ 0.025). The dual-marker pattern (IL-6 ↑, NEAT1 ↓) showed the strongest discrimination for SNS1 versus HC (AUC = 0.85) and the steepest multinomial contrasts for SNS2 (IL-6 OR_IQR = 4.98; NEAT1 OR_IQR = 0.11).

Elevated IL-6 and decreased NEAT1 define a peripheral signature linked to negative-symptom severity in SCZ and may support biologically informed stratification and longitudinal research.