Background

The high comorbidity rate between depression and physical diseases has long been a challenge in medicine. Patients with Major Depressive Disorder (MDD) are more likely to suffer from various physical conditions such as diabetes, hypertension, and irritable bowel syndrome compared to the general population. Until now, the medical community has attributed these comorbidities to stress hormone secretion caused by depression or unhealthy lifestyle habits. Conversely, it has also been understood that chronic physical diseases can lead to depression.

However, these explanations only demonstrate correlations and have limitations in identifying the biological root causes. In particular, previous genome-wide studies conducted on individual diseases have struggled to comprehensively elucidate the genetic vulnerability of complex, multi-organ disorders involving multiple organ systems and depression. In clinical practice, MDD patients who report physical comorbidities often receive individual treatments for each symptom, resulting in low treatment efficiency and poor medication adherence. This has led to a growing need for a new approach that can simultaneously uncover multidimensional genetic factors influencing multiple diseases.

Key Findings

A joint research team led by Dr. Damian J. Woodward and Professor Eske M. Derks from the QIMR Berghofer Medical Research Institute in Australia used large-scale genomic analysis data to unravel the complex genetic architecture between depression and physical diseases. The team introduced Genomic Structural Equation Modeling (Genomic SEM), a multivariate genetic analysis method. Using this statistical model, they classified physical diseases into four clusters—cardiovascular, metabolic, gastrointestinal, and immune—and applied a method to precisely estimate the genetic overlap with MDD.

The analysis revealed that three disease clusters—cardiovascular, metabolic, and gastrointestinal—formed independent genetic associations with MDD. These three physical disease clusters explained 47% of the SNP-based heritability of MDD, successfully demonstrating the strong genetic connection between mental and physical health. Notably, the gastrointestinal disease cluster showed the highest genetic correlation with MDD (correlation beta = 0.63, P = 3.04 × 10^-30). This is considered strong evidence supporting the Gut-Brain Axis theory at the genetic level, which posits that the gut and brain communicate closely through neural networks and blood flow.

The research team identified hundreds of independent genetic loci shared between each physical disease system and MDD. The metabolic-MDD genetic factors revealed the most loci, with 537 independent loci identified, 417 of which were specific to the metabolic-MDD connection. For CVD-MDD, 172 loci (66 unique) were observed, and for gastrointestinal disease-MDD, 170 loci (42 unique) were identified. In the case of immune disease-MDD, only 141 shared loci were detected. Comparative analysis with databases revealed that many of these loci contained novel variants not previously associated with depression or specific physical diseases in prior research. These variants were found to interact through specific drug pathways, cell types, and biological mechanisms.

Implications and Outlook

This study provides an opportunity to redefine depression not as a purely mental disorder confined to the brain, but as a systemic, multi-organ syndrome closely linked to overall bodily dysfunction. The academic value of this research is high, as it constructs an integrative biological model that bridges psychiatry and physical medicine through large-scale genomic big data analysis. In particular, the discovery of numerous genetic variants mediated through the gut-brain axis is expected to accelerate the development of personalized treatments for patients with depression accompanied by irritable bowel syndrome or gastroesophageal reflux disease.

However, the study primarily used genetic data from individuals of European descent, which limits its generalizability to patients from diverse ethnic backgrounds. Additionally, variations in sample sizes across genome-wide association studies (GWAS) for different disease groups may have introduced statistical power differences that could affect the results. To apply these findings directly in clinical settings, further functional genomics research is needed to clarify how the identified genetic variants are expressed and interact within long-term and brain cells.