Back

Dissection of the clinical phenome of major depressive disorder into subdomains and subgroups in relation to activated immune-inflammatory profiles

Niu, M.; Zhang, Y.; Zhang, X.; Yangyang, C.; Zhuang, X.; Li, J.; Maes, M.

2025-10-23 psychiatry and clinical psychology
10.1101/2025.10.21.25338439 medRxiv
Show abstract

BackgroundMajor depressive disorder (MDD) is a prevalent mood disorder characterized by low remission rates and a high risk of suicide. It encompasses various clinical domains, including depression, anxiety, vegetative, physiosomatic and melancholic symptoms, and suicidal behaviors. AimsTo dissect the clinical phenome of MDD, identify potential constructs within the symptom domains, delineate MDD subclasses, investigate the predictive role of recurrence of illness (ROI) on the phenome, and analyze the relationship between those clinical features and immune activation biomarkers. MethodsA total of 125 MDD patients and 40 healthy controls were recruited for this study. Assays included serum immunological features (M1, Th1, Th2, Th17, IRS, CIRS, and TNF signaling), negative acute phase proteins (APPs; transferrin, albumin), sCD40L, EGF, and Flt-3L. ResultsBifactor analysis integrated multiple clinical domains into a general factor, labeled "OSOD" (overall severity of depression), while extracting a single group factor of physiosomatic symptoms. Cluster analysis identified two subgroups of MDD: major dysmood disorder (MDMD) and simple dysmood disorder (SDMD). ROI mediates the impact of adverse childhood experiences on OSOD and correlates with enhanced immune activation (increased Th17). Downregulation of negative APPs and Flt-3L alongside increased EGF and TNF signaling distinguish MDMD versus SDMD or MDD from controls with great accuracy. Up to 60% of MDD patients show immune aberrations with a specificity of 92.1%. ConclusionThis study promotes a novel diagnostic approach that integrates ROI, OSOD, physiosomatic symptoms, MDMD versus SDMD, and immune profiling. It has the potential to establish the groundwork for future personalized therapeutic strategies.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.