2026 ECNP Congress

PS01-1069

Altered immune–autonomic and neurovisceral coupling in late-onset depression: a multimodal dimensional study combining gene expression, HRV, and fMRI

Federica Goffi — P. Enrico, G. Enrico, C. Cabasino, L. Sperti, F. Bottaro, M.P. Marra, L. Di Consoli, G. Delvecchio, A. Finardi, G. Schiena, Y. Torrente, F.M. Triulzi, R. Furlan, A.M. Bianchi, P. Brambilla, E. Maggioni · Politecnico di Milano, Department of Electronics Information and Bioengineering, Milan, Italy

2026 ECNP Congress

Background

Major depressive disorder (MDD) is increasingly conceptualized as a systemic condition involving alterations across immune, autonomic, and neural systems [1-2]. Reduced heart rate variability (HRV), reflecting impaired autonomic flexibility, is observed in MDD and may contribute to increased cardiovascular vulnerability [3]. The vagal inflammatory reflex represents a key bidirectional neuroimmune pathway linking autonomic regulation and immune activity [4], and alterations in this axis may reflect disrupted integration across interacting physiological systems. However, relationships between peripheral immune status and autonomic and brain functioning in late-onset MDD remain unclear within a unified framework.

Aims

To characterize immune-brain-heart interactions in late-onset MDD using a multimodal dimensional approach integrating gene expression, HRV, and functional magnetic resonance imaging (fMRI).

Methods

Eighty-eight participants were included, comprising 45 patients diagnosed with late-onset MDD (DSM-IV-TR; 19M/26F, 64.9±7.5 years) and 43 healthy controls (HC; 11M/32F, 68.7±10.2 years). Peripheral blood gene expression was profiled using whole-blood RNA extraction and cDNA synthesis, followed by custom OpenArray PCR of a targeted immune-related gene panel. A subset of 40 participants (20 MDD, 20 HC) underwent 500-second resting-state 3T fMRI with simultaneous electrocardiographic recording. HRV was derived from the electrocardiogram. Time-varying estimates of sympathetic and parasympathetic-vagal activity were obtained using bivariate autoregressive modeling and included as regressors in a general linear model to identify brain regions showing fMRI responses to autonomic dynamics [5]. Immune-related gene-expression data were summarized using principal component analysis (PCA) to derive data-driven immune dimensions. HRV-fMRI measures were then projected onto the immune PCA space. Regression models assessed the effects of immune component scores on HRV measures and HRV-related fMRI responses, adjusting for age and sex, with PC-by-group interactions testing group-specific associations.

Results

PCA identified distinct immune dimensions. PC1 reflected an adaptive lymphocyte-related immune profile, characterized by higher expression of T-/B-cell and regulatory markers, with the strongest positive loadings including CD19, FOXP3, GATA3, and IL2RA. PC2 captured a pro-inflammatory, immune-recruitment pattern, with the strongest positive loadings including IL18, TNF, CXCL8, and PTX3. PC2 was associated with reduced HRV, with significant effects on low-frequency (LF, p=0.010), high-frequency (HF, p=0.033), and root-mean-square of successive differences (RMSSD, p=0.017), indicating decreased autonomic variability with increasing immune activation. PC2-by-group interactions indicated stronger negative associations in HC than in MDD (LF, p=0.047; HF, p=0.033; RMSSD, p=0.024). Multimodal HRV-driven fMRI analyses revealed reduced HRV-related responses in MDD within regions of the central autonomic network (p<0.001). Immune status was associated with vagal-linked brain activity, with PC1 primarily linked to anterior cingulate responses (p=0.028), and both PC1 (trend, p=0.06) and PC2 linked to anterior insula activity, where associations showed nonlinear profiles (PC1², p=0.018; PC2², p=0.016).

Conclusions

A pro-inflammatory, immune-recruitment gene-expression profile is associated with reduced autonomic variability, and immune-autonomic coupling appears altered in late-onset MDD, reflecting reduced physiological flexibility. Immune-related modulation of brain activity in central autonomic regions further supports a multi-level dysregulation, with brain responses suggesting more complex integration of intrinsic regulatory and peripheral inflammatory signals. These findings move beyond single-organ views, supporting a multi-system model of depression. This perspective may help refine patient stratification and inform future strategies aimed at restoring integrated neurovisceral regulation.

References

  1. Vitali, E., Cattane, N., D’Aprile, I., Petrillo, G., & Cattaneo, A. (2025). Systemic Inflammation at the Crossroad of Major Depressive Disorder and Comorbidities: A Narrative Review. International journal of molecular sciences, 26(19), 9382. https://doi.org/10.3390/ijms26199382
  2. Valenza G. (2023). Depression as a cardiovascular disorder: central-autonomic network, brain-heart axis, and vagal perspectives of low mood. Frontiers in network physiology, 3, 1125495. https://doi.org/10.3389/fnetp.2023.1125495
  3. Goffi, F., Maggioni, E., Bianchi, A. M., Brambilla, P., & Delvecchio, G. (2025). Is cardiac autonomic control affected in major depressive disorder? A systematic review of heart rate variability studies. Translational psychiatry, 15(1), 217. https://doi.org/10.1038/s41398-025-03430-3
  4. Thayer J. F. (2009). Vagal tone and the inflammatory reflex. Cleveland Clinic journal of medicine, 76 Suppl 2, S23–S26. https://doi.org/10.3949/ccjm.76.s2.05
  5. Goffi, F., Enrico, P., Reali, P., Torino, G., Marra, P. M., Di Consoli, L., Ferro, A., Schiena, G., Torrente, Y., Lombardi, L., Triulzi, F. M., Bianchi, A. M., Brambilla, P., & Maggioni, E. (2026). Brain-heart interactions in late-onset major depressive disorder revealed by multimodal HRV-driven fMRI. Communications Biology. https://doi.org/10.1038/s42003-026-10156-5