Filling the longitudinal gap: connectivity of the default mode and salience networks in severe depression across three time points

Introduction:

It is widely accepted that disrupted network connectivity within the Default Mode Network (DMN) and the Salience Network (SN) plays a crucial role in the pathophysiology of Major Depressive Disorder (MDD). Numerous cross-sectional studies have investigated this issue. However, due to high methodological heterogeneity in the field of fMRI analyses, as well as substantial inter-individual variability in functional connectivity, there is currently no consensus on whether increased or decreased functional connectivity within the DMN or SN is associated with the severity or the pathophysiology of the depressed state. Moreover, it remains unclear whether functional connectivity in these networks can serve as a predictive marker for positive treatment outcomes. Longitudinal studies, capable of shedding light on how connectivity evolves across the course of MDD, are rare. Even fewer studies examine functional connectivity in severely depressed patients over extended periods and with more than two time points.

Aim:

Therefore, our aim is to longitudinally track the course of major depressive disorder (MDD) in a real-world, severely depressed clinical sample and to examine how functional connectivity within the DMN and the SN evolves over time. This approach may help elucidate the neurobiological underpinnings of individual depressive states and their dynamic changes.

Methods:

Within the IDeA-L study, we collected MRI data and assessed clinical measures in severely depressed patients at three time points: baseline, two weeks, and six months. Depressive symptom severity was evaluated using the Montgomery-Åsberg Depression Rating Scale (MADRS). All patients received pharmacological treatment according to established clinical guidelines (75% received SSRIs). Functional connectivity was assessed during a 20-minute multi-echo fMRI sequence. Functional connectivity within the Default Mode Network (DMN) and Salience Network (SN) at each of the three time points will be quantified using seed-based correlation analysis and incorporated as independent variables in mixed-effects regression models to examine their association with changes in depressive symptom severity over time. All preprocessing and connectivity analyses were performed using Python scripts based on Nilearn and Nipype; subsequent regression analyses were conducted in R.

Results:

Currently, baseline data have been collected from N = 76 patients (46% female; mean MADRS at Baseline = 31.49, SD = 9.98; mean age = 38.39 years). At baseline, higher functional connectivity within the DMN and the SN was associated with higher MADRS scores (DMN: r = .14, SN: r = .12). At the time of abstract submission, n = 40 patients have completed all three assessment time points. While data collection for the final (6-month) assessment is still ongoing and recruitment remains open, we expect to have the complete dataset, including all analyses, finalized and available by the time of the congress.

Conclusion:

By characterising how DMN and SN connectivity evolves across three assessment points, this longitudinal analysis may reveal dynamic neurobiological markers of disease trajectory that can be used for prognostic stratification and the understanding of the development of personalized, network‑targeted treatment strategies.

References

No references