We found marked increases in GMVs of the VS and

We found marked increases in GMVs of the VS and mOFC. As a key region in the orbitofronto-striato-thalamic pathway (Menzies et al., 2008; Milad and Rauch, 2012; Pauls et al., 2014), VS (mainly the nucleus accumbens in this study) demonstrated significant enlargement as in previous studies (Norman et al., 2016; Pujol et al., 2004), while existing reports show either increased (Szeszko et al., 2008) or decreased volumes in OFC (Norman et al., 2016; Rotge et al., 2009). Functional studies have indicated that hyperactivation in the mOFC and caudate may be related to goal-directed dysfunction in OCD (Gillan et al., 2014). Meanwhile, the vegf inhibitor as a central link in CSTC circuitry demonstrated abnormally increased volume, a finding consistently associated with OCD pathology here and in other studies (Boedhoe et al., 2016; Eng et al., 2015; Rotge et al., 2009). In contrast, decreased GMVs of the left dlPMC/pre-SMA are observed, as in prior meta-analyses of OCD studies (Norman et al., 2016; Rotge et al., 2009). Premotor areas are critical for response inhibition, both in suppressing an unwanted action and facilitating a desired one (Duque et al., 2012). Recently de Wit and colleagues found left dlPMC/pre-SMA hyperactivity in OCD patients and their unaffected siblings during response inhibition (de Wit et al., 2012). Interestingly, OCD patients not only exhibit increased attention to the actual outcomes of such actions, but also show heightened intra-individual correlation between error-related negativity and activity in pre-SMA that is consistent with increased worry about future negative outcomes (Grutzmann et al., 2016). In order to glean clinically meaningful volume alterations caused by pharmacotherapy, we integrated effect size analysis with rigorous cross-validation procedures, revealing strikingly differential effects of pharmacological modulation on OCD-specific abnormalities. Firstly, trivial differences remained between the medicated and control groups in premotor and ventral striatum areas, indicating that the most prominent therapeutic effects of serotonergic-based drugs occur in these regions. Secondly, the modulatory effect on the thalamus is relatively ambiguous here, as significant medication-related differences occurred in over 77% of sampling rounds (albeit with small effect sizes; Table 2) despite earlier longitudinal accounts revealing reversible volumetric changes induced by SSRI treatment (Gilbert et al., 2000; Hoexter et al., 2012; Szeszko et al., 2004). Lastly, we unveiled that the orbitofrontal cortex of OCD patients, particularly the medial part of the left hemisphere, exhibits pronounced resistance to pharmacological modulations. Recent evidence based on animal models has shown that hyperstimulation of glutamatergic OFC-ventromedial striatum projections leads to compulsive-like grooming behavior that is reversible with chronic fluoxetine, further substantiating its potential as a therapeutic target for OCD (Ahmari et al., 2013). Intriguingly, our analysis implies divergent pharmacological modulation of the OFC and ventral striatum, which requires further experimental investigation to dissect their distinct predictive responses in the orbitofronto-striato-thalamic loop. It is worth mentioning that prioritized alterations in brain regions and circuits imposed by various medications are a core characteristic of clinical biomarkers (Abi-Dargham and Horga, 2016). The key medication targets identified here overlap partially with regions preferentially targeted by an emerging treatment candidate – ketamine, which mainly acts on the OFC, subgenual and posterior cingulate cortices, and nucleus accumbens, as demonstrated in our recent preclinical work (Lv et al., 2016). Collectively, these findings raise promising potential for the development of specific loci or circuit-targeted therapeutics that are potentially more effective in particular subpopulations of patients involving particular phenotypic/dimensional deficits.