10.4 MAPPING NEURO-BEHAVIORAL RELATIONSHIPS IN DIMENSIONAL GEOMETRIC EMBEDDING (N-BRIDGE) VIA PHARMACOLOGY, COMPUTATION AND CLINICAL NEUROIMAGING: UNIFYING CATEGORIES AND DIMENSIONS ALONG THE PSYCHOSIS SPECTRUM. (9th April 2019)

• Record Type: Journal Article
• Authors: Anticevic, Alan
• Is Part Of: Schizophrenia bulletin. Volume 45(2019)Supplement 2
• Page Start: S103
• Page End: S104
• Publication Date: 2019-04-09
• Subjects: Schizophrenia -- Periodicals; Schizophrenia -- Research -- Periodicals; 616.898005
• DOI: 10.1093/schbul/sbz022.037
• Languages: English
• ISSNs: 0586-7614

Abstract

Abstract: Background: A key challenge in the field of neuropsychiatry lies in matching patients with effective treatments. Most studies in psychiatry operate under the canonical assumption that categorical diagnostic clinical grouping and/or pre-existing clinical assessments are the 'gold standard' for describing behavioral - and therefore neural - variation in patients. Attempts to robustly characterize the neural substrates of these predefined variables have yielded limited success, suggesting an inadequate mapping to neurobiologically meaningful variation. Notably, a great deal of heterogeneity exists even within groups of patients with the same categorical diagnosis. Thus, understanding the mapping between specific behaviors and clinically-meaningful variation in neural properties is critical to develop and ultimately administer effective individualized neurobehavioral treatments. Methods: Here, we describe a multivariate neuro-behavioral framework under which behavioral variation can be mapped to features of specific neural systems in a data-driven way. We leverage neural (fMRI-derived) and behavioral data from 436 psychosis-spectrum patients across genders that were publicly available via the NIMH Data Archive as part of the Bipolar & Schizophrenia Consortium for Parsing Intermediate Phenotypes study (https://ndar.nih.gov/edit_collection.html?id=2274 ). We relate these findings to effects from two pharmacological neuroimaging experiments manipulating the NMDA glutamate receptor via ketamine (N=40, both genders) and the 5-HT receptor via LSD (N=24, both genders). Results: We first identify dimensions of maximal behavioral variation in patients by performing a principal component analysis across all behavioral measures. Importantly, these dimensions are not parallel to traditional clinical symptom scales derived from pre-existing clinical instruments used in psychiatry, and do not reflect conventional categorical diagnostic boundaries. We then demonstrate that variation along our identified behavioral dimensions relates to variation in specific neural systems, using a data-driven measure of functional connectivity (global brain connectivity) (p<.05 whole-brain corrected). Critically, these robust neuro-behavioral relationships were not observed using either traditional diagnostic groups or a priori clinical scales. We also demonstrate the flexibility to embed both categorical and continuous/dimensional features within the same generalized multivariate geometry. We further show that this framework can inform the identification of pharmacological targets for developing drugs for specific symptom profiles and may provide an assisted selection of behavioral measures that precisely pinpoint variation in a specific neural circuit at the individual subject-level in relation to ketamine and LSD effects. Conclusions: Characterizing how and which specific sets of symptoms map to neural circuitry is a key step towards developing targeted and effective treatments for psychiatric disorders. We propose the Neuro-Behavioral Relationships In Dimensional Geometric Embedding (N-BRIDGE) framework as a key step towards unified mapping between the geometry of data-driven behavioral variation and the geometry of data-driven neural variation, thus integrating both categories and continuous dimensions in psychiatry.