Plasma copeptin is coupled to resting-state frontoinsular-subgenual alpha dynamics in major depression with comorbid post-traumatic stress disorder.

Background: Hypothalamic-pituitary-adrenal axis dysregulation is implicated in depression and post-traumatic stress disorder (PTSD), but stress-related neuroendocrine mechanisms that distinguish their comorbid presentation remain poorly understood. Prior work in the broader parent cohort found lower levels of plasma copeptin, a surrogate marker of arginine vasopressin, in 0individuals experiencing a major depressive episode (MDE) who also had PTSD. The present study examined whether copeptin was associated with frequency-specific resting-state neural activity and circuit dynamics in an overlapping magnetoencephalography (MEG) subset of participants with MDE+PTSD, MDE without PTSD, and healthy volunteers (HVs). Methods: Whole blood samples were collected from 105 participants (MDE+PTSD=18, MDE=59, HV=28) and assayed for plasma copeptin concentrations. Eyes-closed resting-state MEG data were acquired and source localized using synthetic aperture magnetometry in the delta, theta, alpha, beta, gamma, and high gamma frequency bands. Linear mixed models were used to examine copeptin-by-diagnosis interactions in whole-brain source-localized power. Two regions of interest implicated in affective, interoceptive, and stress-related processing-the left anterior insula (AI) and subgenual anterior cingulate cortex (sgACC)-were further examined using virtual electrodes and dynamic causal modeling. Parametric empirical Bayesian analysis modeled group differences with mean-centered copeptin values entered as covariates. Results: In the alpha frequency band, a significant group-by-copeptin interaction (pCLC<0.05) was identified in medial and lateral prefrontal cortex, including the sgACC and AI. Compared with the HV and MDE groups, MDE+PTSD participants demonstrated a negative association between alpha power and copeptin levels. In the high gamma band, a significant main effect of diagnosis was identified in early visual cortex, encompassing the bilateral lingual gyrus, with both clinical groups showing greater power than HVs. Dynamic causal modeling of sgACC-AI circuitry showed strong evidence (Pp>0.99) for reduced intrinsic excitatory drive within the sgACC and AI in the MDE+PTSD group. Higher copeptin levels in the comorbid group were associated with local circuit changes consistent with reduced excitatory throughput and faster inhibitory timing. Conclusion: These findings raise the possibility that MDE+PTSD may represent a distinct neuroendocrine-neurophysiologic phenotype in which the relationship between peripheral vasopressinergic signaling and cortical alpha dynamics/frontoinsular-subgenual circuit function is altered.