Biol Psychiatry. 2025 Sep 12:S0006-3223(25)01459-3. doi: 10.1016/j.biopsych.2025.08.022. Online ahead of print.

ABSTRACT

Cytokines are immune signaling molecules that also function as neuromodulators. Cytokines are elevated in the peripheral blood of some individuals with mental disorders, suggesting that inflammation may contribute to their illness. Furthermore, immune therapy trials for systemic inflammatory disorders have reported improvements in anxiety and depression as secondary endpoints. These findings bolster the salutary potential for anti-inflammatory treatment of primary psychiatric populations. However, immunotherapeutic trials in depression and other psychiatric disorders have largely yielded inconclusive or negative results. One possibility is the reliance of clinical trial designs on cross-sectional measurements of inflammatory markers in blood. Peripheral cytokine profiles may not reflect central inflammatory states and cannot disclose dynamic relationships between compartments. Because central cytokines directly modulate neural activity, mapping their dynamic relationships between the periphery and central nervous system may improve future clinical trial designs. Therefore, we performed a systematic search for studies that measured cytokines at multiple time points in paired blood and cerebrospinal fluid samples from humans and non-human primates. Our narrative synthesis of these studies found that peripheral and central cytokine fluctuations are uncorrelated in humans under basal conditions. Moreover, an evoked increase in a cytokine’s level peripherally may provoke a dramatic increase in a distinct cytokine centrally without eliciting a meaningful change in its own central level (cross-correlation in the absence of autocorrelation). Furthermore, physical and psychological stressors can induce compartment-specific cytokine changes and correlations. These initial observations highlight the need for a more complete map of cytokine dynamics in humans with and without mental illness.

PMID:40946890 | DOI:10.1016/j.biopsych.2025.08.022