- MDD shows greater global cortical thickness heterogeneity than controls (Cohen's d = -0.26), indicating increased neuroanatomical variability.
- Regional variability concentrates in cingulate, insular, frontal and temporal cortices, with 5.7-6.8% higher thickness variation in these regions.
- Heterogeneity is greater among patients on antidepressants (Cohen's d = -0.39); individuals with extreme variability show higher symptom severity, supporting stratification.
Transl Psychiatry. 2026 Jun 23. doi: 10.1038/s41398-026-04189-x. Online ahead of print.
ABSTRACT
The clinical and biological heterogeneity of major depressive disorder (MDD) may reflect the aggregation of different conditions with distinct pathologies under a single diagnostic label. Neuroanatomical heterogeneity in MDD was examined using a harmonized, age- and sex-matched sample from the ENIGMA MDD consortium (N = 5146; age range: 9-82 years; 64% female). Analyses of global neurostrucutral variability revealed greater cortical thickness heterogeneity in MDD compared with healthy controls (Cohen’s d = -0.26). Regionally, increased variability in cortical thickness was most prominent in the cingulate (+6.1 to +6.6% more variation in MDD) and insular (+5.8%) cortices, as well as in the frontal (+5.7 to +6.8%) and temporal (+6.1 to +6.8%) lobes. Heterogeneity in cortical thickness was more pronounced among patients using antidepressant medication (Cohen’s d = -0.39). Patient-specific analyses further showed that individuals with markedly increased cortical thickness variability (<5th percentile relative to the normative range) exhibited greater depressive symptom severity than those within the normative range (5th-95th percentile; Cohen’s d = 0.19-0.36). Overall, the results indicate that neuroanatomical heterogeneity in MDD is primarily expressed in cortical thickness, offering refined insights into the neurobiological complexity of structural alterations associated with depression. These findings could guide future stratification efforts examining whether regionally confined changes in cortical thickness within areas of pronounced variability reflect clinically meaningful patient subgroups.
PMID:42337246 | DOI:10.1038/s41398-026-04189-x
Share Evidence Blueprint

Search Google Scholar
Save as PDF

