bioRxiv2025
BackgroundPsychiatric disorders are increasingly conceptualised as heterogeneous categories with transdiagnostic underlying mechanisms that cut across multiple diagnoses and vary within a single diagnosis. Inflammation induced by psychosocial stress is one particularly potent example: previous research suggests that inflammatory profiles may correspond to symptom subgroups rather than traditional diagnostic categories. However, robust identification of transdiagnostic symptoms linked to specific inflammatory profiles remains rare. In this study, we examined the relationship between inflammatory profiles (at baseline and after a stress induction) and transdiagnostic symptom dimensions in females, who show higher prevalence of stress-related disorders such as anxiety and depression. MethodsA modest but relatively homogenous healthy female sample, between the ages of 18 and 35, was recruited (N=26). We obtained venous blood samples, at baseline and after a combined physiological and social stress induction, to measure full blood counts, plasma cytokines and peripheral blood mononuclear cells (PBMCs; for cell stimulation and intracellular flow cytometry analysis). Participants completed a battery of psychiatric self-report questions, from which we modelled three transdiagnostic factor scores. Finally, we used Bayesian regressions to evaluate the predictive contribution of transdiagnostic factors to inflammatory markers (at baseline and stress-induced), as well as to the principal components of inflammatory measures (derived from a principal component analysis (PCA) of the inflammatory data). ResultsWe identified specific relationships between inflammatory profiles and transdiagnostic symptom dimensions. Higher scores on a social withdrawal factor were associated with greater baseline neutrophil (95% highest density interval (HDI) = [0.06, 1.32]; BF10 = 2.92) and monocyte counts (95% HDI = [0.29, 1.46]; BF10 = 19.08), whereas higher anxious-depression scores were associated with a lower baseline monocyte count (95% HDI = [-0.96, -0.02]; BF10 = 1.81) and a greater inflammatory response to stress (e.g., change in neutrophil scores from pre- to post stress induction: (95% HDI = [0.18, 2.14]); BF10 = 5.66). We also found evidence for an association between the social withdrawal factor and an immune principal component most strongly weighted by monocytes, basophils, and IL-6 (95% HDI = [-1.14, -0.01]; BF10 = 1.88). ConclusionsWe find preliminary evidence that different transdiagnostic psychiatric symptom dimensions map onto specific inflammatory profiles, both at baseline and after a stress induction. This represents a proof-of-principle for the use of data-driven and hypothesis-driven approaches to identify and link transdiagnostic factors with inflammatory changes, which may be of use to future studies with larger, clinical populations, or for testing in the context of stratified interventions.