Predicting Transitions From Depression to Wellbeing Through a Network-based Operationalization of Plasticity (PlastNet)

Major depressive disorder (MDD) is one of the leading causes of disability worldwide. Despite available treatments, a substantial proportion of patients do not respond adequately, which highlights the need for approaches that personalize therapeutic strategies.

Plasticity is defined as the ability to modify brain functioning, behaviors and mental states, and is increasingly recognized as a crucial process underlying the transition from psychopathology to wellbeing. Plasticity is neither inherently beneficial nor detrimental: enhanced plasticity increases the likelihood of a mental state transition without determining the direction of the transition, which is modulated by contextual factors such as quality of life. Treatments that enhance plasticity may therefore produce highly context-dependent effects. Lifestyle, environmental and psychotherapeutic interventions are also essential for promoting mental health, but not all individuals benefit equally, partly because of differences in individual plasticity levels. While psychometric tools to evaluate contextual factors have long been available, methods to operationalize and assess plasticity remain limited.

The Network Theory of Plasticity operationalizes plasticity as the inverse of connectivity strength among a system's elements: the weaker the connectivity, the higher the plasticity. Analyses of the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) and Combining Medications to Enhance Depression Outcomes (CO-MED) datasets provided preliminary evidence supporting this operationalization at the population level and showed its high sensitivity. PlastNet aims to validate this approach at the individual level and to use it to develop personalized therapeutic strategies.

Study design: PlastNet is an observational study. No medicinal product or medical device is under investigation, and no intervention is assigned. The procedures (blood sampling, resting-state functional magnetic resonance imaging (fMRI), and clinical and neuropsychological assessments) are standard clinical and research tools, and no additional risk beyond routine clinical practice is anticipated. Two hundred adults with nonpsychotic MDD will be recruited at IRCCS Ospedale San Raffaele over 24 months and followed for 4 weeks each. At baseline, participants undergo resting-state fMRI, blood sampling and clinical assessments; clinical assessments are repeated at the four-week follow-up. Contextual factors are assessed with the World Health Organization Quality of Life-BREF (WHOQOL-BREF) questionnaire (quality of life), the Schedule of Recent Experiences (SRE) and the Perceived Stress Scale (PSS) (psychosocial stress), and socioeconomic status.

Primary objective and analysis: to generalize the operationalization of plasticity to the level of the central nervous system. The investigators hypothesize that lower baseline brain functional connectivity strength, derived from resting-state fMRI, is associated with a greater reduction in depressive symptom severity, measured as change in the total score of the 17-item Hamilton Depression Rating Scale (HDRS-17) from baseline to four-week follow-up. Whole-brain connectivity strength is the primary predictor; connectivity within the default mode network and the salience network is a secondary predictor. Change in the total score of the 13-item Beck Depression Inventory (BDI-13) is analyzed as a supportive endpoint. Because there is no co-primary endpoint, no multiplicity adjustment is required.

Secondary objective: to identify biological signatures of plasticity through integrated multi-omics analyses of peripheral blood collected at baseline. The investigators hypothesize that baseline multi-omics profiles predict individual plasticity and change in the HDRS-17 and BDI-13 total scores. Analyses include transcriptomic profiling of peripheral blood mononuclear cells by RNA sequencing (Illumina NovaSeq 6000), genomic analysis with the Illumina Infinium Global Screening Array-24 (>680,000 markers) and untargeted metabolomics by ultra-performance liquid chromatography coupled with tandem mass spectrometry (UPLC-MS/MS). Data will be integrated using elastic net, random forests and Data Integration Analysis for Biomarker discovery using Latent cOmponents (DIABLO) to generate individualized plasticity scores. Predictive accuracy will be assessed by cross-validation (alpha = 0.05), with Benjamini-Hochberg false discovery rate (FDR) correction (FDR < 0.05).

Exploratory objective: to integrate symptom-based, neuroimaging-based and multi-omics-based plasticity measures into a unified individualized plasticity score and to investigate its interplay with contextual factors in determining treatment outcomes. No formal hypothesis is pre-specified; analyses are hypothesis-generating, and results will be considered preliminary pending replication.

Clinical significance: if validated, this approach could allow clinicians to tailor interventions to individual plasticity profiles, improving the precision and outcomes of psychiatric care.