Rebeaud Jessica, Phillips Nicholas Edward, Thévoz Guillaume, Vigne Solenne, Nassirnia Sedreh, Gauthier-Jaques Aude, Lim-Dubois-Ferriere Pansy, Panda Satchidananda, Théaudin Marie, Du Pasquier Renaud, Greub Gilbert, Bertelli Claire, Kuhle Jens, Collet Tinh-Hai, Pot Caroline
Laboratories of Neuroimmunology, Department of Clinical Neurosciences, Center for Research in Neuroscience and Service of Neurology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland.
Service of Endocrinology and Diabetology, Geneva University Hospitals, Geneva, Switzerland.
Metabolomics. 2025 Aug 12;21(5):114. doi: 10.1007/s11306-025-02315-2.
Multiple sclerosis (MS) is an autoimmune disorder with an unpredictable outcome at the time of diagnosis. The measurement of serum neurofilament light chain (sNfL) and glial fibrillary acidic protein (sGFAP) has introduced new biomarkers for assessing MS disease activity and progression. However, there is a need for additional diagnostic and prognostic tools. In this study, we investigated the predictive abilities of metabolomics, gut microbiota, as well as clinical and lifestyle factors for MS outcome parameters.
The aim of this study was to assess the predictive capacity of plasma metabolites, gut microbiota, and clinical/lifestyle factors on MS outcome measures including MS-related fatigue, MS disability, and sNfL and sGFAP concentrations.
A prospective cohort study was conducted with 54 individuals with MS. Anthropometric, biological, and lifestyle parameters were collected. The least absolute shrinkage and selection operator (LASSO) algorithm with ten-fold cross-validation was used to identify predictors of MS disease outcome parameters based on plasma metabolomics, microbiota sequencing, and clinical and lifestyle measurements obtained from questionnaires and anthropometric measurements.
Circulating metabolites were found to be superior predictors for sNfL and sGFAP concentrations, while clinical and lifestyle data were associated with EDSS scores. Both plasma metabolites and clinical data significantly predicted MS-related fatigue. Combining multiple multi-omics data did not consistently improve predictive performance.
This study highlights the value of plasma metabolites as predictors of sNfL, sGFAP, and fatigue in MS. Our findings suggest that prioritizing metabolomics over other methods can lead to more accurate predictions of MS disease outcomes.
多发性硬化症（MS）是一种自身免疫性疾病，在诊断时其预后难以预测。血清神经丝轻链（sNfL）和胶质纤维酸性蛋白（sGFAP）的测量引入了新的生物标志物，用于评估MS疾病的活动和进展。然而，仍需要额外的诊断和预后工具。在本研究中，我们调查了代谢组学、肠道微生物群以及临床和生活方式因素对MS预后参数的预测能力。
本研究的目的是评估血浆代谢物、肠道微生物群以及临床/生活方式因素对MS预后指标的预测能力，这些指标包括与MS相关的疲劳、MS残疾以及sNfL和sGFAP浓度。
对54例MS患者进行了一项前瞻性队列研究。收集了人体测量学、生物学和生活方式参数。使用具有十折交叉验证的最小绝对收缩和选择算子（LASSO）算法，基于血浆代谢组学、微生物群测序以及从问卷和人体测量中获得的临床和生活方式测量结果，确定MS疾病预后参数的预测因子。
发现循环代谢物是sNfL和sGFAP浓度的更好预测因子，而临床和生活方式数据与扩展残疾状态量表（EDSS）评分相关。血浆代谢物和临床数据均显著预测了与MS相关的疲劳。合并多个多组学数据并没有持续提高预测性能。
本研究强调了血浆代谢物作为MS中sNfL、sGFAP和疲劳预测因子的价值。我们的研究结果表明，将代谢组学置于其他方法之上可以更准确地预测MS疾病的预后。
