Major depressive disorder (MDD) is the most common psychiatric disease worldwide, with huge socio-economic impacts. Pharmacotherapy represents the first-line therapeutic choice, but about 30% of patients are classified as resistant to treatment (TRD). TRD is associated with specific clinical and biological features; however, taken individually, these signatures have limited power in response prediction. The project’s aim is the development of an innovative algorithm for the early detection of non-responder patients, more prone to later develop TRD. Phase 1 will involve 300 patients with MDD already recruited, including 150 TRD/150 responders, considered as “extremes” in relation to treatment response. A full clinical assessment will be performed, together with a comprehensive molecular evaluation (genomic, transcriptomic and miRNomic profiling). An algorithm integrating all these data will be developed in order to predict response to therapy. In phase 2, a new cohort of 300 MDD patients will be recruited to assess, in real-world conditions, the ability of the algorithm to correctly predict treatment outcome. Moreover, an active participation of patients will be established to consider their perspectives and needs. Project results will provide a new predictive tool for future use in the clinical practice, enabling a better prevention and management of MDD treatment resistance.