UHPLC-MS/MS-based untargeted metabolite profiling of Lyme neuroborreliosis

The diagnosis of Lyme neuroborreliosis (LNB) requires the demonstration of intrathecal synthesis of Borrelia antibodies in a patient's cerebrospinal fluid (CSF), which involves the invasive procedure of a lumbar puncture. This study serves as a feasibility study aimed at exploring the potential of using serum samples, which are easily obtainable routine clinical samples, for LNB diagnostics via advanced metabolomics techniques. Serum samples were collected from confirmed LNB patients before and after treatment, with post-treatment samples serving as controls. The objective of the study was to find stable biomarkers for acute LNB through untargeted metabolomics analysis using ultrahigh-performance liquid chromatography coupled with tandem mass spectrometry (UHPLC-MS/MS). The study focused on biomarkers that could be reliably detected in serum samples stored under typical clinical conditions, without the need for special handling, ensuring consistent detection over time. The analysis revealed 26,978 molecular features (MFs), of which 1,746 were statistically significant (p < 0.001). Further manual investigation into 91 of the most prominent MFs revealed 53 potential biomarkers for LNB, individually or in combination. The workflow developed provides a comprehensive platform for biomarker detection, with potential applications in both research and clinical settings for LNB and other infections. This minimally invasive diagnostic approach is promising, and additional validation and future studies are needed for it to be considered as a practical alternative or a complement to CSF-based diagnostics of LNB in everyday clinical practice.