Jason G. Craggs - US grants

? DESCRIPTION (provided by applicant): Chronic pain is now considered a disease of the central nervous system (CNS), affecting approximately 103- million people, the costs associated pain are a staggering $653 billion annually, and increasing. Yet, hampering our progress is the lack of a standardized, objective tool for assessing and diagnosing chronic pain (e.g., a neural signature of pain). Recent evidence suggests that the advances in neuroimaging data, along with biopsychosocial, psychophysiological, behavioral, and self-report measures can identify such an index. However, the studies were narrowly focused and the measurement properties (e.g., reliability, sensitivity, etc.) of these biomarker/signatures were not examined. This information is critical to the establishment of a neural pain signature. To address this need we developed an innovative and novel methodology to assess the measurement properties of three neuroimaging modalities that have been suggested for establishing a signature of pain, including: resting state (i.e., the default mode network [DMN]), task-based fMRI, and high- resolution structural MRI data. These data will be acquired twice from three groups (normal controls [NC], and chronic pain (CP) patients with chronic low back pain [CLBP], and fibromyalgia [FM]). In Aim-1, we will use a mood induction protocol to challenge the reliability, sensitivity/specificity, and stability of resting state data (RS) and the connectedness of regions n the default mode network (DMN), between groups and over time. For Aim-2, we will study the measurement properties fMRI associated with thermal pain and the memory of a painful event. In Aim-3, we will assess the measurement properties of structural MRI measurements for use as an objective biomarker of pain. As with our previous work, in Aims 1 and 2 we will: 1) develop models of effective connectivity among pain related networks; 2) use multi-group analyses to estimate the validity of a single model of effective connectivity being reliably produced across al groups; and 3) use longitudinal growth curve modeling (LGCM) to assess the influence of time on the reliability and stability of model identification across all three groups. Consistent with te development of any other clinical tool, the results provided by this study are necessary in the quest to identify an objective pain signature, and/or use it clinically with any confidence. Findin robust measurement properties across the NC and CP groups, in any or all of the imaging modalities, will provide significant support for a common pain signature. However, instability, poor reliability, and a significant amount of unexplained variance in the measurements would suggest that neuroimaging data might not be valid for the identification of an objective pain biomarker. Rather, if the measurement properties for the entire sample are poor, a case might be made for the presence of condition specific signatures. Regardless of the outcome, the results from this study will advance our understanding of endogenous pain mechanisms, and contribute to the development of new treatment options.