2026 ECNP Congress

PS02-2100

Cognition meets connectivity: clinical and fMRI-based prediction of functional recovery in major depressive disorder

Silas Katzenbach — K.M. Kachel, L. Hepke, O. Tüscher, K. Lieb, A. Sebastian, J. Engelmann · Johannes Gutenberg University Medical Center Mainz, Department of Psychiatry and Psychotherapy, Mainz, Germany

2026 ECNP Congress

Introduction

Cognitive deficits occur in up to 50 % of patients with Major Depressive Disorder (MDD) and often persist despite remission of depressive symptoms [1, 2]. Such persistent impairments are associated with reduced occupational and everyday functioning and with less favourable treatment trajectories [3]. Recent findings suggest a cognitively characterizable MDD subtype distinguished by alterations in the Central Executive Network (CEN) [4]; executive function (EF) deficits in particular have been linked to altered CEN connectivity [3]. The present analysis examined whether baseline EF performance and CEN connectivity provide prognostic information for the longitudinal change in global functioning (ΔGAF) and depressive symptomatology (ΔIDS) in MDD.

Methods

Data were obtained from the IDeA-L study (Individual Determinants of Antidepressant Response – a Longitudinal study); recruitment is ongoing. All patients (N = 76; 47 female; mean age 38.39 ± 13.96 years) received antidepressant pharmacotherapy. Depressive symptoms (Inventory of Depressive Symptomatology, IDS-30) and global functioning (Global Assessment of Functioning, GAF) were assessed at baseline and after 12 weeks. EF was measured using the Trail Making Test (TMT-A/B) and the Regensburg Word Fluency Test (RWT; phonemic and semantic switching). Resting-state functional connectivity was acquired prior to treatment using two 10-minute multi-echo functional magnetic resonance imaging sequences (3 T Siemens Magnetom; TR = 1435 ms). CEN connectivity was operationalized as Fisher-z-transformed partial seed-to-seed connectivity between the dorsolateral prefrontal cortex and the dorsal anterior cingulate cortex. From six EF and connectivity variables, predictors were selected via elastic-net regression (α = 0.5, 10-fold cross-validation) separately for ΔGAF and ΔIDS. Selected predictors were entered into multiple linear regression models adjusted for the corresponding baseline value; nested models were compared using ANOVA.

Results

Over 12 weeks, patients improved substantially in both depressive symptoms (ΔIDS = –22.78 ± 13.31) and global functioning (ΔGAF = 23.68 ± 21.36). For ΔGAF, elastic-net regression identified semantic category switching (RWT-SW) and CEN connectivity (CEN_C) as relevant predictors. Adjusted for baseline GAF, RWT-SW significantly predicted functional improvement (β = 1.00, p = .015), while CEN_C showed a consistent positive but non-significant trend (β = 55.42, p = .177). The combined model explained the largest share of variance (R² = .379), though the incremental contribution of CEN_C was not significant (ΔR² = .030, p = .100). For ΔIDS, none of the six predictors was retained; only baseline IDS-30 was significant (β = –0.60, p < .001), indicating a dissociation between predictors of functional and symptomatic recovery.

Conclusions

Baseline semantic category switching was significantly associated with the change in global functioning, while CEN connectivity showed a consistent but non-significant trend in the same direction. Neither EF nor CEN connectivity predicted symptom reduction, suggesting a domain-specific association with later functional outcome rather than with depressive symptom course. These findings support the notion that functional recovery represents a distinct dimension in MDD prognosis that may benefit from cognitive markers in future predictive models. Replication in larger samples will be needed to validate these preliminary findings and to evaluate their potential for cognitively-oriented treatment strategies.

References

  1. Semkovska M, Quinlivan L, O’Grady T, Johnson R, Collins A, O’Connor J, Knittle H, Ahern E, Gload T. Cognitive function following a major depressive episode: a systematic review and meta-analysis. Lancet Psychiatry. 2019 Oct;6(10):851-861.
  2. Liu J, Chen Y, Xie X, Liu B, Ju Y, Wang M, Dong Q, Lu X, Sun J, Zhang L, Guo H, Zhao F, Li W, Zhang L, Li Z, Liao M, Li L, Zhang Y. The percentage of cognitive impairment in patients with major depressive disorder over the course of the depression: A longitudinal study. J Affect Disord. 2023 May 15;329:511-518.
  3. Kaiser RH, Andrews-Hanna JR, Wager TD, Pizzagalli DA. Large-Scale Network Dysfunction in Major Depressive Disorder: A Meta-analysis of Resting-State Functional Connectivity. JAMA Psychiatry. 2015 Jun;72(6):603-11.
  4. Hack LM, Tozzi L, Zenteno S, Olmsted AM, Hilton R, Jubeir J, Korgaonkar MS, Schatzberg AF, Yesavage JA, O’Hara R, Williams LM. A Cognitive Biotype of Depression and Symptoms, Behavior Measures, Neural Circuits, and Differential Treatment Outcomes: A Prespecified Secondary Analysis of a Randomized Clinical Trial. JAMA Netw Open. 2023 Jun 1;6(6):e2318411. doi: 10.1001/jamanetworkopen.2023.18411. Erratum in: JAMA Netw Open. 2023 Jul 3;6(7):e2326241.