Brain-based classification of youth with anxiety disorders: transdiagnostic examinations within the ENIGMA-Anxiety database using machine learning

Aghajani, Moji Wright, Barry Straube, Thomas Straube, Benjamin Strawn, Jeffrey R. Suarez-Jimenez, Benjamin Sylvester, Chad M. Talati, Ardesheer Thomopoulos, Sophia I. Tükel, Raşit van Nieuwenhuizen, Helena Werwath, Kathryn Wittfeld, Katharina Wu, Mon-Ju Stark, Rudolf Yang, Yunbo Zilverstand, Anna Zwanzger, Peter Blackford, Jennifer U. Avery, Suzanne N. Clauss, Jacqueline A. Lueken, Ulrike Thompson, Paul M. Pine, Daniel S. Stein, Dan J. van der Wee, Nic J. A. Veltman, Dick J. Stein, Frederike Bruin, Willem B. Zhutovsky, Paul van Wingen, Guido A. Bas-Hoogendam, Janna Marie Groenewold, Nynke A. Hilbert, Kevin Winkler, Anderson M. Zugman, Andre Agosta, Federica Åhs, Fredrik Andreescu, Carmen Antonacci, Chase Asami, Takeshi Assaf, Michal Barber, Jacques P. Bauer, Jochen Bavdekar, Shreya Y. Beesdo-Baum, Katja Benedetti, Francesco Bernstein, Rachel Björkstrand, Johannes Blair, Robert J. Blair, Karina S. Blanco-Hinojo, Laura Böhnlein, Joscha Brambilla, Paolo Bressan, Rodrigo A. Breuer, Fabian Cano, Marta Canu, Elisa Cardinale, Elise M. Cardoner, Narcís Cividini, Camilla Cremers, Henk Dannlowski, Udo Diefenbach, Gretchen J. Domschke, Katharina Doruyter, Alexander G. G. Dresler, Thomas Erhardt, Angelika Filippi, Massimo Fonzo, Gregory A. Freitag, Gabrielle F. Furmark, Tomas Ge, Tian Gerber, Andrew J. Gosnell, Savannah N. Grabe, Hans J. Grotegerd, Dominik Gur, Ruben C. Gur, Raquel E. Hamm, Alfons O. Han, Laura K. M. Harper, Jennifer C. Harrewijn, Anita Heeren, Alexandre [UCL] Hofmann, David Jackowski, Andrea P. Jahanshad, Neda Jett, Laura Kaczkurkin, Antonia N. Khosravi, Parmis Kingsley, Ellen N. Kircher, Tilo Kostic, Milutin Larsen, Bart Lee, Sang-Hyuk Leehr, Elisabeth J. Leibenluft, Ellen Lochner, Christine Lui, Su Maggioni, Eleonora Manfro, Gisele G. Månsson, Kristoffer N. T. Marino, Claire E. Meeten, Frances Milrod, Barbara Jovanovic, Ana Munjiza Mwangi, Benson Myers, Michael J. Neufang, Susanne Nielsen, Jared A. Ohrmann, Patricia A. Ottaviani, Cristina Paulus, Martin P. Perino, Michael T. Phan, K. Luan Poletti, Sara Porta-Casteràs, Daniel Pujol, Jesus Reinecke, Andrea Ringlein, Grace V. Rjabtsenkov, Pavel Roelofs, Karin Salas, Ramiro Salum, Giovanni A. Satterthwaite, Theodore D. Schrammen, Elisabeth Sindermann, Lisa Smoller, Jordan W. Soares, Jair C. et al.[show all]

Neuroanatomical findings on youth anxiety disorders are notoriously difficult to replicate, small in effect size and have limited clinical relevance. These concerns have prompted a paradigm shift toward highly powered (that is, big data) individual-level inferences, which are data driven, transdiagnostic and neurobiologically informed. Here we built and validated supervised neuroanatomical machine learning models for individual-level inferences, using a case–control design and the largest known neuroimaging database on youth anxiety disorders: the ENIGMA-Anxiety Consortium (N = 3,343; age = 10–25 years; global sites = 32). Modest, yet robust, brain- based classifications were achieved for specific anxiety disorders (panic disorder), but also transdiagnostically for all anxiety disorders when patients were subgrouped according to their sex, medication status and symptom severity (area under the receiver operating characteristic curve, 0.59–0.63). Classifications were driven by neuroanatomical features (cortical thickness, cortical surface area and subcortical volumes) in fronto- striato-limbic and temporoparietal regions. This benchmark study within a large, heterogeneous and multisite sample of youth with anxiety disorders reveals that only modest classification performances can be realistically achieved with machine learning using neuroanatomical data.