Prognostic and Prescriptive Predictors of Treatment Response to Adjunctive VNS Therapy in Major Depressive Disorder: A RECOVER Trial Report.

Vagus nerve stimulation (VNS) therapy is a long-term intervention for treatment-resistant major depression (TRD) adjunctive to treatment as usual (TAU). To enhance clinical decision- making, we identified subgroups that respond especially well or poorly with active VNS vs no stimulation sham VNS (prognostic predictors) and subgroups that specifically benefit from active VNS vs sham VNS (prescriptive predictors). In the RECOVER trial, patients with marked TRD (N=493) were randomized to either active VNS (N=249) or sham VNS (N=244); both groups continued TAU. Baseline demographic, clinical, and treatment history characteristics were evaluated as potential prognostic and/or prescriptive outcome predictors. Outcome assessment was based on a tripartite measure that combined depressive symptoms (Quick Inventory of Depressive Symptomatology-Clinician), psychosocial function (Work Productivity and Activity Impairment Questionnaire item 6), and quality of life (Mini-Quality of Life Enjoyment and Satisfaction Questionnaire). Generalized linear mixed models were employed to identify both prognostic and prescriptive predictors of tripartite outcomes. Several baseline features predicted outcomes across the entire sample and within the sham VNS group (prognostic prediction). History of treatment with electroconvulsive therapy (ECT; lifetime and current episode) or transcranial magnetic stimulation (TMS; current episode) was associated with poorer prognosis. However, these same features were associated with greater benefit from active VNS vs sham VNS. The presence of comorbid anxiety disorders was predictive of a better prognosis overall, but smaller benefit from active VNS vs sham VNS. Marked TRD patients with a history of ECT or TMS had especially poorer outcomes when receiving sham VNS plus TAU for 1 year than those without this history. These same subgroups showed significant differential benefit with active VNS than with sham VNS (positive prescriptive effect). The absence of a comorbid anxiety disorder was linked to superior benefit from active VNS vs sham VNS. These predictors may inform clinical decision-making when considering VNS. ClinicalTrials.gov identifier: NCT03887715.