A Study of Individualized Diagnosis and Treatment for Major Depressive Disorder With Atypical Features

The lifetime prevalence of major depressive disorder (MDD) is 10%~20%. Worldwide, nearly 340 million individuals have suffered the torture of depression. World Health Organization has reported that MDD would become the most serious global burden of disease and eventually turn into a public health problem in 2030. Varied clinical symptoms, inappropriate treatment, unclear pathogenesis, and lack of recurrent risk early-warning predictors cause a series of clinical problems, such as low diagnostic rate, low effective treatment rate, and high recurrent rate. Hence, this study aims to search multidimensional markers for early diagnosis of MDD, to establish optimized personalized therapy, and to explore sensitive recurrence predictors.

Based on the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), MDD is subdivided into eight different clinical specifiers, one of which the incident rate of MDD with atypical features reaches 30%~38%. However, there is still a lack of meta-evidence for the clinical treatment strategy in MDD with atypical features. And 45.4 percentage of MDD with atypical features convert to bipolar disorder. Therefore, this study will focus on three issues about what's the objective endophenotype in MDD with atypical features, how to select appropriate personalized treatment for MDD with atypical features, what's the predictive biomarker of conversion to bipolar disorder.

Based on the investigators' previous findings, this study will investigate adult depression at a cross-sectional study and a prospective cohort study. Multivariate informatics analysis was performed from three research dimensions (cognitive neuropsychology, metabonomics, and multimodal neuroimaging), including atypical features, "cold/hot" cognition assessment, KP (kynurenine pathway) metabolomics and inflammatory factors, multimodal MRI robust property. Referring guidelines for the diagnosis and treatment of depression and evidence-based medicine evidence, MDD with atypical features are divided into f groups (antidepressants, antidepressants+mood stabilizers, mood stabilizers, treat as usual). Then, the investigators perform follow-up to verify optimized treatment strategies and to explore risk factors of conversion from MDD with atypical features to bipolar disorder. Furthermore, this study performs correlation analysis to analyze cross-omics data, weight coefficient analysis to analyze multidimensional indexes, clustering analysis to analyze multivariate bio-information data, and artificial intelligence technologies (such as pattern recognition, and machine learning) to realize the transformation from medical data to practical transformation. Eventually, this study builds three specific models (the multidimensional early diagnosis models for MDD with atypical features, the optimized personalized therapy model, and the recurrence and conversion risk early-warning model), which form the integrated intelligent platform for multidimensional diagnosis, personalized treatment, recovery management of MDD with atypical features.