RNA editing‐based biomarker blood test for the diagnosis of bipolar disorder: results of the EDIT-B consortium study

Abstract Background Differentiating bipolar disorder (BD) from major depressive disorder (MDD) during depressive episodes represents one of the most critical challenges in clinical psychiatry as patients may not recall prior hypomanic or manic episodes. RNA editing biomarkers combined with AI may improve diagnostic accuracy and support personalized therapeutic approaches. Methods The assay used is an in vitro diagnostic (IVD) blood test based on 8 genes that leverages adenosine-to-inosine (A-to-I) RNA editing signatures to aid in the differentiation of BD from MDD. A set of RNA-editing blood biomarkers within these 8 genes was previously identified and published. In this multicenter European study, we assessed the performance of three distinct RNA editing–based algorithms developed prior to this project. Sequencing data were processed via a secure HDS–certified platform by central clinical laboratory and the results for each algorithm were compared against physician’s diagnosis to evaluate diagnostic performance. Results Across 426 patients with a current major depressive episode recruited from four European centers, the three algorithms demonstrated robust diagnostic performance. Algorithm A284, exhibited the strongest results and highest performance, with high sensitivity (82.6%) and specificity (80.3%) across sites and patient subgroups, thus reproducing the findings and performance obtained on a previous independent cohort. Conclusion This study validates A-to-I RNA editing signatures coupled with AI as a reliable approach to distinguish BD from MDD highlighting the clinical utility of the assay and algorithm A284 as an innovative, robust, complementary diagnostic tool.