Association between plasma ketamine metabolites and antidepressant response in treatment-resistant depression

Introduction

Ketamine in subanaesthetic doses can produce rapidly acting antidepressant effects in patients with treatment-resistant depression (TRD). A treatment response is observed in approximately 50% of patients with TRD [1]. Reliable predictors of individual treatment response are still lacking. Certain metabolites may play a key role in explaining variability in treatment outcomes [2].

Aims

In this study, we investigated plasma concentrations of (R,S)-ketamine metabolites as potential biomarkers of treatment response within the NeuroMarKet study (“Neuroimaging and Blood Markers as Indicators of Ketamine Efficacy in Treatment Resistant Depression”).

Methods

Thirty-one patients with TRD received six weight-adjusted ketamine infusions over a period of three weeks at the Central Institute of Mental Health in Mannheim. Plasma levels of ketamine (K), norketamine (NK), and dehydronorketamine (DHNK) were measured immediately before, one hour after, and 24 hours after each infusion. D18 was defined as day of the last ketamine infusion after six infusions in total. Depression severity was assessed using the Montgomery–Åsberg Depression Rating Scale (MADRS). To examine whether pharmacokinetic markers were associated with overall treatment response, we analyzed the relationship between mean peak concentrations across the initial five infusions.

Results

Twenty-four hours after the first ketamine infusion, 11 of 31 patients showed a response (≥50% reduction in MADRS), and 5 achieved remission (MADRS ≤10). At D18, 16 patients (51.6%) met criteria for treatment response, while 15 (48.4%) were classified as non-responders. Twelve patients (38.7%) met criteria for remission. Lower NK/K ratios one hour after infusion were significantly associated with greater reductions in MADRS scores (ρ = 0.44, p = 0.014); a borderline association was observed for DHNK/K (ρ = 0.36, p = 0.049). No significant associations were found between metabolite ratios at 24 hours and treatment response. Linear regression confirmed a significant effect of log-transformed NK/K ratios on early changes in MADRS scores (β = 18.45, p = 0.034). A similar association was observed for log-transformed DHNK/K ratios (β = 12.76, p = 0.043). In addition, log(NK/K) was significantly associated with the probability of response (β = −3.02, p = 0.022; OR = 0.05). No significant associations were observed between pharmacokinetic markers and overall symptom change at D18 in correlation analyses (ketamine: r = −0.176, p = 0.34, norketamine: r = 0.143, p = 0.44; DHNK: r = 0.175, p = 0.35; NK/K: r = 0.276, p = 0.13; DHNK/K: r = 0.216, p = 0.24). When additional covariates were included, the NK/K ratio became significantly associated with MADRS scores at D18 in models adjusted for age, sex and BMI, (β = 5.4, p = 0.0499, Adj. R² = 0.293).

Conclusions

Slower ketamine metabolism, as indicated by lower metabolite-to-parent compound ratios one hour after infusion, was associated with greater early antidepressant effects and a higher likelihood of treatment response. Overall, associations between pharmacokinetic markers and treatment response at D18 were less consistent than those observed for early response and appeared to depend on model specification. A limitation of the study is the lack of inclusion of hydroxynorketamine, a potentially active metabolite.

References

1. Medeiros GC, Demo I, Goes FS, Zarate CA Jr, Gould TD. Personalized use of ketamine and esketamine for treatment-resistant depression. Transl Psychiatry. 2024 Nov 29;14(1):481.
2. Zanos P, Moaddel R, Morris PJ, Riggs LM, Highland JN, Georgiou P, Pereira EFR, Albuquerque EX, Thomas CJ, Zarate CA Jr, Gould TD. Ketamine and Ketamine Metabolite Pharmacology: Insights into Therapeutic Mechanisms. Pharmacol Rev. 2018 Jul;70(3):621-660.