PBMC lipidomic profiling as diagnostic and prognostic biomarkers in treatment-resistant depression: a pilot esketamine study in women

Treatment-resistant depression (TRD), defined as the failure to respond to at least two adequate antidepressant trials, affects 30–40% of patients with major depressive disorder and represents a major clinical and economic burden worldwide. Esketamine, a non-competitive NMDA receptor antagonist approved by the EMA in 2019, produces rapid antidepressant effects, yet treatment response remains variable and no validated biomarkers exist to identify potential responders. Lipids constitute promising biomarker candidates given their active roles in neurotransmission, synaptic plasticity, and intracellular signalling, with previous evidence suggesting that peripheral lipidome alterations are associated with both depressive states and antidepressant response following ketamine administration in TRD patients, and may be related to the underlying mechanisms of treatment resistance.

To investigate the lipid profile as a potential biomarker for TRD diagnosis and prediction of esketamine treatment response, the lipidomic profile of peripheral blood mononuclear cells (PBMCs) was analyzed in 15 female TRD patients recruited at Basurto University Hospital and 12 female healthy volunteers, at baseline and after four weeks of treatment. Patients were classified as responders (n = 8) or non-responders (n = 7) based on a ≥50% reduction in Montgomery–Åsberg Depression Rating Scale (MADRS) scores. Correlation analyses between lipid species intensities and MADRS scores were conducted to assess their clinical relevance, and Random Forest classification models were implemented to evaluate their discriminative and predictive capacity in both diagnostic and prognostic contexts.

A diagnostic model based on a ten-lipid panel discriminated TRD patients from healthy volunteers, achieving an AUC of 0.756 and an overall accuracy of 73.8%. Certain species of phosphatidylethanolamines and phosphorylated ceramides emerged as the most discriminative features, both of which are linked to neuroinflammatory processes. Within the TRD group, specific species of phosphatidic acids, phosphatidylinositols, and phosphatidylglycerols showed positive correlations with clinical severity measured by MADRS, suggesting that higher lipid intensities reflect greater depressive symptomatology. A prognostic model comprising eight lipids distinguished treatment responders from non-responders with an AUC of 0.797 and an accuracy of 79.2%. Among these, phosphatidylinositol species were identified as the most critical prognostic markers.Moreover, a direct association was observed between phosphatidylinositol baseline intensity and treatment response magnitude, whereby higher lipid levels correlated with poorer therapeutic response, reflected in a smaller reduction in MADRS scores following esketamine administration.

Therefore, these results highlight the clinical potential of the lipidomic profile as a promising tool for the identification of both diagnostic and prognostic biomarkers in TRD, particularly in females. These findings may contribute to the advancement of personalized medicine approaches in the context of TRD because they may improve patient stratification as well as therapeutic decision-making. Nevertheless, further studies including male subjects are required to determine whether the observed differences are sex-specific or may serve as general biomarkers of TRD and predictors of response to esketamine treatment.

References

No references