Prediction of outcomes for deep brain stimulation using multimodal MRI

PROJECT SUMMARY/ABSTRACT Parkinson disease (PD) is a common and growing source of disability worldwide, and adjunctive surgical treatments are often necessary to optimize treatment in more advanced stages of the disease once medical therapy becomes insufficient. Deep brain stimulation (DBS) has become the most common and effective surgical

neuromodulatory technique for motor symptoms of PD, however selection of patients and surgical targets is limited due to an incomplete understanding of predictive factors for DBS outcomes. Historically, levodopa responsiveness has been the primary factor used to establish DBS candidacy, but this correlates only modestly

with motor benefit, does not predict cognitive and psychiatric effects, and offers little information on optimal DBS target. Alternative preoperative predictors based on demographics, volumetric analysis, and functional connectivity have been described, but these predictors vary across studies and have not been validated for

clinical use. As such, a comprehensive understanding of clinical and neuroimaging predictors of DBS effects is necessary to improve outcomes for those with advanced PD. The purpose of this Mentored Patient-Oriented Career Development Award (K23) is to enable the candidate to develop a funded research program focused on the creation of a model of DBS effects, integrating

clinical data and preoperative MRI data to improve patient and target selection in DBS for PD. The candidate's long term goal is to become an independent clinician-scientist investigator capable of developing and implementing clinical neuroimaging tools to investigate DBS mechanisms, improve DBS outcomes, and develop

novel DBS applications for individuals with PD and other neurological disorders. Training and mentorship are proposed in three key areas: (1) conducting clinical research, including study design and ethical conduct, (2) performing advanced statistical analyses including multivariate modeling , and (3) obtaining additional skills in

neuroimaging tools and analysis, including functional and structural MRI. The research plan for this award will be supported by the training plan as well as an expert team of mentors and an outstanding institutional research environment. The objective of this plan is to develop and validate a model of DBS responsiveness, integrating

clinical data and preoperative MRI data, which will improve patient and target selection in DBS for PD. Specifically, this project aims to (1) build a model of motor response to DBS of the subthalamic nucleus (STN) using preoperative behavioral, volumetric, structural and functional connectivity measures, (2) validate this model

using prospective data of individuals undergoing STN DBS for PD, and (3) construct similar predictive models for cognitive and psychiatric outcomes of STN DBS. This will set the stage for future interventional studies using this model as a basis for patient and target selection, and lead to use of more comprehensive objective behavioral

and neuroimaging data in the preoperative evaluation and planning for DBS in PD.