Subcortical volumetric alterations in four major psychiatric disorders: a mega-analysis study of 5604 subjects and a volumetric data-driven approach for classification
Naohiro Okada,Masaki Fukunaga,Kenichiro Miura,Kiyotaka Nemoto,Junya Matsumoto,Naoki Hashimoto,Masahiro Kiyota,Kentaro Morita,Daisuke Koshiyama,Kazutaka Ohi,Tsutomu Takahashi,Michihiko Koeda,Hidenaga Yamamori,Michiko Fujimoto,Yuka Yasuda,Naomi Hasegawa,Hisashi Narita,Satoshi Yokoyama,Ryo Mishima,Takahiko Kawashima,Yuko Kobayashi,Daiki Sasabayashi,Kenichiro Harada,Maeri Yamamoto,Yoji Hirano,Takashi Itahashi,Masahito Nakataki,Ryu-Ichiro Hashimoto,Khin K Tha,Shinsuke Koike,Toshio Matsubara,Go Okada,Theo G M van Erp,Neda Jahanshad,Reiji Yoshimura,Osamu Abe,Toshiaki Onitsuka,Yoshiyuki Watanabe,Koji Matsuo,Hidenori Yamasue,Yasumasa Okamoto,Michio Suzuki,Jessica A Turner,Paul M Thompson,Norio Ozaki,Kiyoto Kasai,Ryota Hashimoto
DOI: https://doi.org/10.1038/s41380-023-02141-9
Abstract:Differential diagnosis is sometimes difficult in practical psychiatric settings, in terms of using the current diagnostic system based on presenting symptoms and signs. The creation of a novel diagnostic system using objective biomarkers is expected to take place. Neuroimaging studies and others reported that subcortical brain structures are the hubs for various psycho-behavioral functions, while there are so far no neuroimaging data-driven clinical criteria overcoming limitations of the current diagnostic system, which would reflect cognitive/social functioning. Prior to the main analysis, we conducted a large-scale multisite study of subcortical volumetric and lateralization alterations in schizophrenia, bipolar disorder, major depressive disorder, and autism spectrum disorder using T1-weighted images of 5604 subjects (3078 controls and 2526 patients). We demonstrated larger lateral ventricles volume in schizophrenia, bipolar disorder, and major depressive disorder, smaller hippocampus volume in schizophrenia and bipolar disorder, and schizophrenia-specific smaller amygdala, thalamus, and accumbens volumes and larger caudate, putamen, and pallidum volumes. In addition, we observed a leftward alteration of lateralization for pallidum volume specifically in schizophrenia. Moreover, as our main objective, we clustered the 5,604 subjects based on subcortical volumes, and explored whether data-driven clustering results can explain cognitive/social functioning in the subcohorts. We showed a four-biotype classification, namely extremely (Brain Biotype [BB] 1) and moderately smaller limbic regions (BB2), larger basal ganglia (BB3), and normal volumes (BB4), being associated with cognitive/social functioning. Specifically, BB1 and BB2-3 were associated with severe and mild cognitive/social impairment, respectively, while BB4 was characterized by normal cognitive/social functioning. Our results may lead to the future creation of novel biological data-driven psychiatric diagnostic criteria, which may be expected to be useful for prediction or therapeutic selection.