The neural correlates of cognitive behavioral therapy: Recent progress in the investigation of patients with panic disorder Yunbo Yang * , Tilo Kircher, Benjamin Straube Department of Psychiatry and Psychotherapy, Philipps-University Marburg, Rudolf-Bultmann-Str. 8, 35039 Marburg, Germany article info Article history: Received 14 April 2014 Received in revised form 10 July 2014 Accepted 16 July 2014 Available online 29 July 2014 Keywords: Exposure-based cognitive behavior therapy Panic disorder with agoraphobia fMRI Neural correlates Mechanism of action Outcome prediction Moderator abstract Cognitive behavioral therapy (CBT) is an evidence-based treatment for mental disorders. Several meta- analytical reviews supported its efficacy and effectiveness in the treatment of panic disorder with agoraphobia (PD/AG). Recently, it has been shown that neurobiological changes are associated with the process and outcome of CBT. However, the general and specific neurobiological effects of CBT are still widely unknown. Therefore, the potential of applying neuroscience to clinical practice and optimizing CBT is still limited. The current review summarizes recent findings about the neural correlates of CBT in PD/AG measured with fMRI. Furthermore, the current review will focus on neural activation patterns predicting and moderating therapeutic success of CBT, due to its potential application in personalized treatment in the future. Finally, we will discuss some future perspectives of the neurosciences in CBT research. © 2014 Elsevier Ltd. All rights reserved. Introduction After more than 50 years of development and research on cognitive behavior therapy (CBT), this treatment has become the most widely used and effective evidence-based treatment for many mental disorders (Beck & Dozois, 2011). Meta-analyses of efficacy and effectiveness support its therapeutic effect across a wide range of symptoms and treatment settings (Butler, Chapman, Forman, & Beck, 2006; Hofmann, Asnaani, Vonk, Sawyer, & Fang, 2012; Stewart & Chambless, 2009). With the introduction of neuro- imaging techniques, such as positron emission tomography (PET), structural and functional magnetic resonance imaging (sMRI/fMRI), and single photon emission tomography (SPECT) in psychotherapy research (Carrig, Kolden, & Strauman, 2009; Schwartz, Stoessel, Baxter, Martin, & Phelps, 1996; Weingarten & Strauman, 2014), the neurobiological correlates of therapeutic change in CBT have been increasingly investigated. A number of more recently pub- lished reviews on the neuroscience of psychotherapy suggest that neurobiological changes are associated with the progress and outcome of psychotherapy. The majority of the reviewed research is on the neural correlates of CBT (e.g., Barsaglini, Sartori, Benetti, Pettersson-Yeo, & Mechelli, 2014; Messina, Sambin, Palmieri, & Viviani, 2013; Thomaes et al., 2014; Weingarten & Strauman, 2014). However, previous neuroimaging studies about CBT effects have only proved the concept which shows that changes of the mind through CBT and changes in the brain are somehow intercorrelated (e.g., Prasko et al., 2004; Sakai et al., 2006). The neurobiological mediator and moderator of CBT effects are widely unknown, which makes it very difficult to apply these findings to clinical practice. An in-depth neuroimaging study of CBT needs to reveal the mecha- nism of action in CBT (Gloster et al., 2009). Neuroscientific infor- mation could provide a new foundation for the optimization and individualization of CBT treatments. However, sophisticated and well-controlled neuroscientific experimental designs embedded in randomized controlled trials (RCTs) are needed for the advance- ment of this endeavor (Kraemer, Wilson, Fairburn, & Agras, 2002). Until now, the modulation of brain physiology with CBT in panic disorder (PD) has been investigated only in two PET studies (Prasko et al., 2004; Sakai et al., 2006), one SPECT study (Seo, Choi, Chung, Rho, & Chae, 2014) and two fMRI studies (Kircher et al., 2013; Straube et al., 2014). The two PET studies and the one SPECT study used a resting state paradigm, in which the patients solely had to rest in the scanner. Resting state activity does not provide specific brain states in subjects. Therefore, the direct association of the change in neural activation during resting state and the change in behavior is hard to draw. Although the two PET and one SPECT studies provide first support for CBT modulating brain activation in * Corresponding author. Tel.: þ49 6421 5865837; fax: þ49 6421 5865406. E-mail address: yangy@med.uni-marburg.de (Y. Yang). Contents lists available at ScienceDirect Behaviour Research and Therapy journal homepage: www.elsevier.com/locate/brat http://dx.doi.org/10.1016/j.brat.2014.07.011 0005-7967/© 2014 Elsevier Ltd. All rights reserved. Behaviour Research and Therapy 62 (2014) 88e96