Adaptive Neuromodulation of Working Memory Networks in Aging and Dementia

点击翻译按钮获取中文摘要

PROJECT SUMMARY Dementia due to Alzheimer’s disease (AD) is a leading public health concern in the US with enormous care costs and no effective pharmacotherapy despite multiple clinical trials. Multiple studies have shown mild cognitive impairment (MCI) to be a precursor risk for AD and to be more amenable to intervention. While preclinical studies have shown that directly modulating activity in the prefrontal cortex (PFC) using non-invasive brain stimulation techniques, such as transcranial magnetic stimulation (TMS), can modulate cognitive function in healthy older adults, there is little evidence of reliable efficacy in MCI. We posit three reasons for this lack of efficacy. First, there is no established means of estimating a reliable biomarker as well as a unique dose-response relationship between TMS intensity and brain activity, which remains a fundamental means of titrating individualized response to neuromodulation. Second, standard TMS protocols fail to capture the dynamic nature of cognitive states and the reaction of endogenous brain states to exogenous neuromodulation. By understanding the dynamic changes associated with a successful brain state, it should be possible to manipulate PFC dynamically in a manner that enhances cognition. Third, no studies using TMS in AD-related populations have accounted for the influence of cerebrovascular disease in the response to TMS. We propose to address these shortcomings by using closed-loop TMS, based on individualized brain networks to establish parameters that can reliably control brain states during normal memory functioning in healthy aging and MCI. To achieve this goal, we will study network activation and neural oscillatory mechanisms underlying the network that regulates working memory (WM), a cognition function with a reliable PFC-based network characterization. We will then target this network using closed-loop TMS to the PFC and measure the impact on WM performance and task-based neural activity. This approach, which builds on our existing K01, U01, and RF1 awards, uses concurrent TMS-fMRI to identify dose-response relationships in the working memory network, which can be used to identify neuroplasticity and optimize targeting for TMS (Aim 1). Next, we apply novel closed-loop TMS to perturb this network using temporally-precise TMS-EEG (Aim 2), optimizing the encoding of memory by minimizing endogenous alpha oscillations. Lastly, we will integrate information collected via fMRI and EEG into a single computational framework in order to model spatiotemporal dynamics of the global brain network, accounting for the influence of both connectivity and cerebrovascular pathology in predicting the success of the TMS-related response in our MCI cohort (Aim 3). In sum, the project will use cutting-edge brain stimulation and network modeling techniques to enhance WM in healthy older adults and MCI and will provide a demonstration of the value of closed-loop, network-guided TMS for future clinical applications.