From Resistance to Redesign-The Emerging Logic of Hybrid Care in Treatment-Resistant Depression

Background/Objectives: Treatment-resistant depression (TRD) remains one of the most urgent unmet needs in psychiatry, while its therapeutic pipeline is evolving rapidly. To characterize current development trajectories, we conducted a registry-anchored mapping of interventional trials in adults with major depressive disorder and treatment resistance (MDD-TRD), with the aim of defining the distribution of intervention types, endpoint choices, and key design features across the active trial landscape. Methods: We systematically searched ClinicalTrials.gov, the EU Clinical Trials Information System, and ISRCTN for interventional MDD-TRD trials registered up to 18 September 2025. After data cleaning and cross-registry deduplication, 237 unique trials were retained. Interventions were categorized through a taxonomy distinguishing device-based neuromodulation, pharmacological strategies, biologic/novel agents, multimodal non-digital combinations, digital-hybrid programs, psychotherapy, and lifestyle interventions, with classification anchored on structured registry intervention tags and whole-word matching across title and intervention text. Primary endpoints were flagged as standard when they explicitly referenced the Montgomery-Åsberg Depression Rating Scale or Hamilton Depression Rating Scale. We also examined developmental phase, sample size, and recurrent methodological features. Results: Device-based neuromodulation accounted for the largest share of the active pipeline (114/237, 48.1%), followed by pharmacological strategies (86/237, 36.3%), biologic/novel agents (16/237, 6.8%), and multimodal non-digital combinations (11/237, 4.6%). Digital-hybrid programs represented a small but distinctive stratum (5/237, 2.1%), with the remaining records comprising lifestyle interventions (3/237, 1.3%) and psychotherapy (2/237, 0.8%). Standard clinician-rated primary endpoints were used in 63.3% of studies. Trial development was concentrated in mid-phase designs, whereas sample sizes were generally modest (median 49; interquartile range, 19-87). Across modalities, increasing attention was directed to durability of response, functioning, and patient-reported outcomes, with adaptive and enrichment-based designs appearing with greater frequency. Conclusions: The contemporary TRD trial ecosystem is structured around two co-active developmental tracks-device-based neuromodulation and pharmacology with novel mechanisms-accompanied by a smaller but measurably expanding biologic/novel stratum and a still-marginal digital-hybrid presence. This registry-based mapping provides a near-real-time overview of the field and may support future harmonization of trial endpoints and design standards.