Project Summary (Abstract) Suicide is the second leading cause of death in young people (10 to 34 years) and there are currently no evidence-based pharmacologic anti-suicidal interventions for adolescents. Potent risk factors for adolescent suicide include Major Depressive Disorder (which increases the risk 30-fold), and recent discharge from a higher level of psychiatric care relating to suicide. These risks may be further enhanced treatment-resistant populations. Ketamine is anti-suicidal in adult treatment resistant populations, even after controlling for its antidepressant effects. Despite having no evidence base in pediatric psychiatry, ketamine is increasingly being utilized off label by Child Psychiatrists, who have no evidence-based pharmacologic options beyond the TORDIA recommendations. We have recently completed a midazolam-controlled randomized clinical trial in adolescents with treatment resistant depression (TRD) showing rapid (1 day) antidepressant efficacy and sound tolerability of a single ketamine dose. We have case report and pilot data suggesting tolerability and anti-suicidal promise of repeat dosing paradigms in adolescents with TRD. Here we propose a two-phase study to test the rapid anti-suicidal efficacy of ketamine in adolescents at high suicide risk (operationally defined as having TRD and a suicide event within the 120 days prior to enrollment) using a conservative repeat dosing paradigm (four intravenous infusions over two weeks). The first phase is 2-week parallel, double-blind phase comparing ketamine to midazolam, and the second is a 4-month open phase in which midazolam- assigned participants who remain suicidal or depressed can receive open ketamine. All participants will receive medication management according to an adaptation of the Texas Children’s Medication Algorithm and 8 weeks of cognitive behavioral therapy (CBT). All will be followed weekly in the open phase for efficacy and safety, with trial design developed in consultation with the FDA. Given the need for predictive biomarkers of treatment response, adolescents will participate in task and rest-based fMRI neuroimaging. Using our novel connectome- based predictive modeling, which uses tasks to “tweak” brain networks across RDoC domains, we will determine pre-treatment connectome phenotypes, or “fingerprints”, that predict treatment response. We proposed 3 specific aims: (1) To evaluate the feasibility and safety of treating adolescents at high suicide risk with a conservative repeat-dosing ketamine paradigm followed by standard of care treatment over 4 months. (2) To evaluate the 48-hour impact of ketamine on suicidal ideation (measured via Columbia Suicide Rating Scale, recent ideation subscale) compared to midazolam, and to identify connectome phenotypes predictive of ideation post-treatment. (3) To describe the trajectory of suicidal thinking, depressive symptoms, and use of mental health resources in both ketamine responders and non-responders over 4 months. The data generated here will advance scientific knowledge and influence clinical practice, in addition to providing the foundational data needed for subsequent trial design in youth at severe suicide risk.

We will enroll 66 adolescents (13-17 years) at high suicide risk in a two-phase clinical trial testing the anti- suicidal effects of intravenous ketamine, based on our preliminary data with ketamine in the adolescent population. Phase 1 is a double-blind, parallel design where adolescents are randomized to receive four doses of ketamine or midazolam over two weeks; key outcomes include suicidal ideation 48-hours after the first dose and functional connectome phenotypes associated with treatment response. Phase 2 is a 4-month open extension in which midazolam-treated participants who remain symptomatic are offered open ketamine and all participants receive medication management and cognitive behavioral therapy; key outcomes are safety metrics, psychiatric resource utilization, symptom trajectory in responders and non-responders, and connectome phenotypes that are altered by treatment.