The inability to measure putative pathogenic processes in the central nervous system (CNS) of living subjects precludes the determination of their temporal distribution, intra-individual heterogeneity, and their ability to predict disease course. Using multiple sclerosis (MS) as an example of a complex neurological disorder, we sought to determine if cerebrospinal fluid (CSF) biomarkers can be aggregated to predict future rates of MS progression and provide molecular insight into mechanisms of CNS destruction. 1,305 CSF biomarkers were analyzed blindly in the longitudinal training dataset (N=129) of untreated MS patients, using DNA-aptamer assay. Random forest models, validated in an independent longitudinal cohort (N=64), uncovered signatures of MS severity, measured by clinical scales and volumetric brain imaging. Cluster analysis revealed intra-individual molecular heterogeneity of disease mechanisms that include both CNS- and immune-related pathways and may represent novel targets for inhibiting MS progression.

中文翻译：

---

多发性硬化严重程度的分子模型识别致病机制的异质性

无法测量活体受试者中枢神经系统（CNS）中的假定致病过程，因此无法确定其时间分布，个体内异质性以及预测疾病进程的能力。使用多发性硬化症（MS）作为复杂神经系统疾病的一个例子，我们试图确定脑脊髓液（CSF）生物标记物是否可以聚集起来，以预测MS的未来发病率，并提供有关CNS破坏机制的分子见解。使用DNA适体测定法在未经治疗的MS患者的纵向训练数据集中（N = 129）盲目分析了1,305个CSF生物标志物。在独立的纵向队列（N = 64）中验证的随机森林模型，未发现MS严重性的特征，可通过临床规模和大脑体积成像进行测量。