Abstract: Substance use disorders are prevalent, cause significant morbidity, and are a common cause of preventable death. In contrast to alcohol and opioids, there are currently no FDA approved pharmacotherapies for methamphetamine use disorder (MUD), despite extensive efforts at prior medication trials. Psychosocial interventions are used to treat MUD, but many individuals remain refractory, highlighting the continued need for novel treatment development. Our understanding of the neural basis of addiction has grown in recent decades, underscoring the potential to selectively target addiction-related brain areas such as the nucleus accumbens (NAc). Deep brain stimulation is commonly used to treat movement disorders, as well as obsessive compulsive disorder, and allows chronic stimulation of subcortical brain structures, such as NAc. A growing body of animal model and human clinical data suggests NAc stimulation may be beneficial in the treatment of addiction. Here we propose a two-phase study, to test DBS of bilateral NAc for the treatment of MUD. Under the UG3 phase, we will enroll a small number of subjects (n=5) with treatment-refractory MUD and will employ a subject- and rater-blinded cross-over design using subjects as their own control. Subjects will receive inpatient detoxification, followed by DBS surgery, then 1 month of residential substance use disorder treatment, and 12 weeks of psychosocial interventions, along with 18 months of monitored DBS programming (subjects randomized to 6- months-sham then 12-months-active stimulation vs. active-then-sham treatment). Our clinical outcomes are methamphetamine use as measured by timeline followback and confirmed by urine drug screens, and self- reported methamphetamine craving. We will also carefully assess safety and feasibility of the study procedures. To investigate circuit-based target engagement, we will test for changes in activity during cue-craving and at rest with longitudinal functional MRI (fMRI) and local field potential recording (pre-DBS programming, 6-, 12-, and 18-months post-surgery). The Medtronic Percept implantable pulse generator allows recording from the implanted DBS leads, providing local field potential recording from the NAc itself during craving events experienced in daily life and with cue-craving presentation in the clinic. If our pre-hoc safety, feasibility and clinical and target engagement endpoints are achieved, we will proceed to the UH3 phase. In this phase, we will conduct a randomized, subject- and rater-blinded, sham-controlled trial of DBS for methamphetamine use disorder (n=20). Our design will be informed by UG3 findings (e.g., to determine expected time course for DBS response) and will seek to test whether NAc DBS added to standard psychosocial interventions provides clinical improvement in reducing methamphetamine use and craving above and beyond sham intervention. We will further explore the underlying mechanisms associated with DBS treatment respon...