FUTURE-D = Forecasting Depression Trajectories: Early Network-Guided Prediction of Severe and Persistent Disease Courses

FUTURE-D redefines early detection and intervention strategies for individuals at risk of severe and persistent depression by adopting a systems-level framework grounded in network science, dynamical systems, and control theory. The MACS and NESDA cohorts provide a unique foundation for modeling the dynamic evolution of neural, symptom, and psychosocial networks underlying severe and persistent depression. To validate and generalize our findings, we will leverage large, harmonized longitudinal datasets—UK Biobank, NAKO, and diverse treatment datasets. Using neuroimaging, digital phenotyping, and novel measures of lived experience, we aim to identify early warning signals of network destabilization predictive of an adverse disease course. A central innovation is to guide TMS interventions tailored to individual network trajectories and informed by robust Patient and Public Involvement. This paradigm shift moves from symptom-based reaction to mechanism-informed prevention, offering a precision mental health approach tailored to individual network trajectories. Early warning signals and individualized control targets will be developed and validated through a co-design process, with strong emphasis on real-world scalability and suitability in LMICs. These efforts will lay the groundwork for future clinical decision-support tools, enabling clinicians to tailor treatments and prevent adverse outcomes.