The response to Li is highly variable between individuals. No clinical or biological biomarkers have been consistently associated to long-term response to this treatment.

We previously identified 3 DMRs that discriminated good from non-responders to Li in BD patients. Here we validated 4 additional DMRs using MS-HRM.

90% of the BD-I individuals were correctly classified according to their response status using the 7 DMR associated to clinical variables.

Lithium (Li) is the most frequently recommended first-line treatment in bipolar disorder (BD).

Nevertheless, response to Li is highly variable between individuals and no clinical or biological biomarkers have been consistently associated to long-term response. We previously identified 3 differentially methylated regions (DMRs) that discriminated good from non-responders to Li. In this study, we validated 4 additional DMRs using Methylation Specific High-Resolution Melting (MS-HRM), a PCR-based low-cost method with easy transferability from bench to bedside. In 61 individuals with BD-I, we investigated associations between these 7 DMRs and Li response, alone or combined with demographic and clinical variables. Using 7 DMRs and 6 demographic and clinical variables, 90 % of the BD-I individuals were correctly classified according to their response status (AUC = 0.95). In particular, combining DNA methylation markers with clinical predictors improved the identification of non-responders (83 % correctly classified). This epigenetic signature might be useful in clinical practice especially as it is imperative to avoid exposing individuals to Li if they have a very low likelihood of response.

Bipolar disorder

Lithium

Biomarker

Response

MS-HRM

DNA methylation

© 2025 The Authors. Published by Elsevier B.V.