47. Early improvement predicts subsequent antidepressant response to intravenous ketamine

Abstract Background Major depressive disorder (MDD) is a debilitating medical condition that places a significant psychological, social, and financial burden on the individuals and families of those affected. However, a third of individuals with MDD do not respond to standard treatments – diagnosed with treatment-resistant depression (TRD). Repeated dosing of intravenous (IV) ketamine has emerged as an effective and rapid-acting intervention for TRD. While early improvement in clinical symptoms is an important predictive marker for end-of-treatment response across a variety of treatments for MDD, further research is required to understand the predictive value of early symptomatic improvement in achieving an antidepressant response from a course of IV ketamine treatments. Aims & Objectives This exploratory study aimed to investigate whether improvements in depression symptom scores at early timepoints within an eight-dose IV ketamine regimen can reliably predict end-of-treatment response status. Using data from a naturalistic clinic setting, we aimed these findings to guide the clinical delivery of IV ketamine as well as provide an evidence-based approach to treatment continuation and discontinuation dialogues.

Method: Patients with TRD (n=77) who received eight doses of IV ketamine (0.5-1.0mg/kg) over the course of 4-5 weeks were included in a retrospective analysis. Patients were diagnosed with MDD according to the DSM-5, and presented with TRD measured as a non-adequate response to 2 or more antidepressants or at least 1 antidepressant and 1 psychotherapy approach. The Patient Health Questionnaire–9 (PHQ-9) was completed before each dosing session. End-of-treatment response was characterized as achieving either a full (≥50% reduction) or partial (25-49% reduction) response. Non-response was <25% score reduction. A Receiver Operating Characteristic (ROC) curve was created based on the percent reduction in PHQ-9 scores after the second dose and the end-of-treatment outcomes. Diagnostic performance metrics were calculated to assess predictive power. A logistic regression was conducted for the strongest predictive timepoint.

Results: There were large and significant reductions in depression symptom scores after the eight treatment days (p<0.0001). The ROC curve displayed a high predictive power of early improvement after the second treatment for assessing end-of-treatment response. The Youden Index found an optimal early improvement threshold of 4.1%. Specificity, sensitivity, positive predictive value (PPV), negative predictive value (NPV), and Youden Index values were 0.86, 0.86, 0.94, 0.69, and 0.71, respectively. High PPV indicates early improvement strongly predicts response, while low NPV shows lack of early improvement does not reliably predict non-response. The logistic regression also found a significant effect of early improvement after second treatment (β = 0.083, p<0.001), demonstrating that for every 1% increase in early improvement, the odds of achieving treatment response increases by 8.7% (OR=1.087, 95% CI: 1.044, 1.148). Discussion & Conclusions In a naturalistic sample of TRD patients receiving an eight-dose treatment course of IV ketamine, a reduction in depression symptom severity after the second dose was a strong predictor of achieving a partial or full antidepressant response by the end of the treatment course. These data suggest that the routine incorporation of measurement-based care into the delivery of IV ketamine may help predict response status for patients with TRD receiving this treatment.