Serum cytokines and T-helper-related immune axes in adolescent major depressive disorder with and without psychotic features.

BACKGROUND: Whether adolescent major depressive disorder (MDD) with psychotic features has a distinct peripheral immune signature is unknown. Prior work in this cohort identified tryptophan-kynurenine alterations without measuring inflammatory markers. METHODS: We enrolled 280 adolescents (128 healthy controls (HC), 78 MDD without psychotic features (MDD-noPsy), 74 MDD with psychotic features (MDD-psy)) and quantified 11 serum cytokines by enzyme-linked immunosorbent assay (ELISA). Covariate-adjusted linear models tested group effects on cytokines and on five prespecified T-helper-related immune-axis scores, with false-discovery-rate control. Inter-cytokine correlation structure was compared edge-wise. Partial redundancy analysis tested immune-axis association with four preselected tryptophan-kynurenine anchors. Elastic-net classification was exploratory. RESULTS: IL-6 and IL-10 were lower and IL-17 A higher in both MDD groups than in HC (q < 0.05); TGF-β1 showed an overall group effect driven mainly by higher levels in MDD-noPsy than HC. Th17/regulatory polarization showed the largest effect (partial η2 = 0.158), higher in both patient groups. Correlation structure differed from HC at eight edges for MDD-psy and one for MDD-noPsy; no cytokine, axis, or edge distinguished the two subgroups. The immune-axis block was not significantly associated with the anchor block (permutation P = 0.577). Adding immune features to kynurenine anchors improved HC-versus-MDD discrimination (AUC 0.806 to 0.868; ΔAUC 0.064) but not subtype discrimination. CONCLUSIONS: Adolescent MDD was associated with a shift in effector-regulatory balance rather than uniform inflammation. Psychotic and non-psychotic MDD were not distinguished by the peripheral immune markers measured here, and immune and kynurenine markers appeared complementary rather than coupled.