Functional & Structural Connectivity of Alcohol Use Disorders

National Institute On Alcohol Abuse And Alcoholism

Investigators

Linked publications, trials & patents

Abstract

1. fMRI Network Analysis and Resting State Studies a. Resting state Connectivity and alcohol use disorder domains. We have been in the process of analyzing major networks such as Salience, Default Mode, and Executive Control networks and their relationship with the circuits involved in addiction cycle as defined by Koob et al (2010). Currently we have collected and are processing the resting state data of healthy control volunteers and alcohol dependent patients. Preliminary results indicate that alcohol dependent individuals have less within network connectivity in the Default Mode Network, when compared to healthy controls. Furthermore, we have identified differences in spectra between groups for these same networks. One of the domains of cycle of addiction as described by Koob and Volkow (2016) is withdrawal/negative affect. This domain has similar characteristics to the personality trait of neuroticism (pessimism, nervousness, and anxiousness), which has been shown to be positively correlated to alcohol use. In addition, it has also been demonstrated that neuroticism is correlated to rs-fMRI connectivity with the amygdala. Because both the withdrawal/negative effect domain and neuroticism consist of negative emotional states, with association to amygdala activity, it is of interest of how these factors relate to each other. We are currently working on a project to explores the relationship between AUD, neuroticism, and rs-fMRI connectivity with the amygdala. We hypothesize that the relationship between AUD and rs-fMRI connectivity with the amygdala will be explained by neuroticism. In addition, because of the prominent association between the amygdala and ventral striatum in Koob's withdrawal/negative affect domain, we hypothesize that the relationship between AUD and resting state connectivity of the nucleus accumbens and the amygdala will be significantly accounted for by neuroticism. So far, we have found a significant relationship between neuroticism and amygdala rs-fMRI connectivity with the following regions: insula, middle frontal gyrus, temporal poles, and the nucleus accumbens. These results support our hypothesis. b. Resting State Connectivity under acute alcohol administration. Studies conducted by Dr. Lovinger's Laboratory for Integrative Neuroscience have demonstrated a selective acute ethanol effect on external globus palidus (GPe) neurons that have a specific connectivity with the dorsal striatum (Abrahao et al._2016). Based on this finding and the known role of the basal ganglia in habitual and compulsive behaviors seen in addiction, we investigated the functional resting-state connectivity changes of the GPe and basal ganglia of 25 healthy social drinkers before and after intravenous ethanol administration. Using a network based statistical analysis, our preliminary results show that after acute ethanol administration resting-state connectivity of the GPe with the caudate and putamen is significantly altered. We also found changes in the connectivity of the substantia nigra, subthalamic nucleus and GPi and other cortical regions. Further analysis of this data and its translational implications is underway. 2. Utilize innovative approaches such as machine learning to assist in individualized patient treatment, treatment efficacy, and relapse prediction. a. Classification of AUD using Machine Learning. Currently, classification of AUD is made on clinical grounds; Robust evidence shows that chronic alcohol use leads to neurochemical and neurocircuitry adaptations. Identifications of the neuronal networks that are affected by alcohol would provide a more quantitative and consistent way of diagnosis and provide novel insights into the pathophysiology of AUD. We are attempting to identify network-level brain features of AUD, and further quantified resting-state within- and between-network connectivity features in a multivariate fashion that are classification of AUD, thus providing additional information about how each network contributes to alcoholism. Our results showed that within-networks features were able to identify AUD and control high degree of precision. As expected Executive Control Networks (ECN) and Reward Network (RN) provided the most significant within-network information. The between-network connectivity between RN - Default Mode Network (DMN) and RN - ECN contribute the most to the prediction. Our findings suggest that machine-learning algorithms provide an alternative technique to quantify large-scale network differences and offer new insights into the identification of potential biomarkers for the clinical diagnosis of AUD (Zhu et al., 2018). b. Quantification of AUD Using Multimodal Data and Machine Learning. Most studies examining the neurobiology of AUD treat individuals with this disorder as a homogeneous group; however, the theories of the neurocircuitry of AUD necessitates a quantitative, multidimensional approach to understand the association between severity of alcohol use and the brain. In this study, participants underwent a structural MRI as well as resting-state, monetary incentive delay, and face matching fMRI scans. Machine learning was applied using the neural data from MRI scanning as the features. The model was trained using a subset of the data and tested for generalizability in a naive sample. A multimodal model best predicted AUD severity in the naive sample, compared to task fMRI, structural MRI, or resting state connectivity alone. Neural features corresponding to the basal ganglia, default mode, salience, and executive control networks explained 32% of the variance associated with AUDIT in this model. These findings indicate that the neural effects of AUD vary according to severity. Our results emphasize the utility of multimodal imaging and suggest a future for neuroimaging derived biomarkers for clinical use in the AUD field. 3. Structural Connectivity Sex Differences in Structural Connectivity of AUD. The purpose of this study was to investigate the prevalence of sex differences in white matter microstructure of the brains of individuals with alcohol use disorder. The damage associated with excessive alcohol use on the brain is accepted and previous diffusion tensor imaging (DTI) research has identified white matter microstructural tract damage within alcohol dependent populations. However, it is unclear if there is a sex difference related to these damages. We conducted voxelwise statistical analysis to measure fractional anisotropy (FA), axial diffusivity (AD), and radial diffusivity (RD) from the DTI images of male and female AUD patients. We found widespread differences between alcohol dependent participants and healthy controls, where alcohol dependent participants had damage to their white matter microstructure compared to controls. We also found significant differences between alcohol dependent men and women in RD, with alcohol dependent women having greater RD in the corticospinal tract, anterior thalamic radiation, uncinate fasciculus, superior longitudinal fasciculus, and inferior fronto-occipital fasciculus. Increases in RD has been associated with demyelination, therefore, these results suggest an interaction between alcohol and sex, whereby alcohol has a more deleterious effect on the myelination of womens brains than men. 4. Development of fMRI Neurofeedback as a Potential AUD Intervention In order to study the potential utility of fMRI neurofeedback as an AUD intervention we have developed a new protocol. This study is a two-stage procedure to both provide evidence of a response modulation deficit associated with socioemotioal processing in individuals with alcohol use disorder and investigate how moderating that deficit affects socioemotional processing and negative drinking consequences.