DECIPHER-OCD: Comprehensive Characterization of Brain Structure in OCD

About This Grant

SUMMARY Obsessive-compulsive disorder (OCD) is prevalent, frequently chronic, and linked to elevated mortality and substantial functional impairment. Available pharmacological treatments have only modest response rates. New treatment strategies are urgently needed. New biological insights are likely to facilitate progress. Analyses of brain structure in OCD have begun to shed some light on associated circuit abnormalities. However, studies to date have focused on single structural measures, such as cortical thickness and regional volume. There is no a priori reason to think that these convenient measures are the most informative; a wealth of other measures such as curvature metrics and morphological similarity indices remain unexplored in OCD. Furthermore, few studies have examined biological correlates of the patterns of spatial variation associated with OCD diagnosis. We address these substantial knowledge gaps in the current proposal. We have developed a pipeline to systematically assess all available metrics of brain structure in parallel: cortical surface area (and Jacobian determinant, an analogous measure for subcortical structures), cortical thickness (and radial distance, an analogous measure for subcortical structures), regional volume, five measures of curvature (sulcal depth, curvature index, Gaussian curvature, folding index, and mean curvature), and two brain network-derived measures (MIND and csMIND, a novel measure developed here). Pilot analyses validate the analytic pipeline and suggest that curvature indices and network measures, which have not previously been assessed in OCD, may provide particular insight. We will apply this pipeline to structural data from 2,274 OCD cases and 2,280 control, collected at 26 sites worldwide as part of the ENIGMA-OCD consortium, of which the PI is a longtime member. We will then apply a novel statistical approach based on canonical correlation to assess the relationship of the spatial maps derived from this structural analysis with maps of a range of biological functions: gene expression (i.e. imaging transcriptomics), neurotransmitter receptor and transporter density (derived from PET imaging), excitation/inhibition ratio, myelination, metabolism (derived from PET metabolic imaging), and correlates of various cognitive functions. This will provide new insight into candidate biological contributors to the observed structural differences in OCD. Finally, in an exploratory analysis, we will systematically compare structural findings in OCD, and their functional correlates, with those from commonly comorbid conditions, establishing which findings are specific to OCD and which are more general markers of transdiagnostic pathology. In aggregate, this work will provide by far the most comprehensive study of the structure of the OCD brain to date. Furthermore, the methods developed and validated here are of broad applicability; future applications of this approach to other diagnoses within the ENIGMA consortium will provide additional insight into the structural and functional underpinnings of illness.