Untargeted metabolomics analysis of cerebrospinal fluid in patients with leptomeningeal metastases from non-small cell lung cancer

作者：Huiying Li, Yongjuan Lin, Shengnan Zheng, Tingting Yu, Yu Xie, Zhenyu Yin · 发表于：Biotechnology and Genetic Engineering Reviews · 年份：2023 · DOI：10.1080/02648725.2023.2191069 · 被引用次数：7 · 研究领域：Brain Metastases and Treatment、Lung Cancer Research Studies、Lung Cancer Treatments and Mutations

Objective: To explore and analyze the diagnostic value of metabolic markers in cerebrospinal fluid (CSF) in leptomeningeal metastases (LM) of non-small cell lung cancer (NSCLC). Methods: Forty-six CSF samples from patients with NSCLC-LM were collected. Another 48 CSF samples from patients with nonmalignant neurological diseases were selected as control group. Metabolomic analysis of CSF was performed by high-performance liquid chromatography-mass spectrometry. Principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) were applied for modeling. A multi-criteria evaluation system (variable importance value >1, multiple of change >2 and P < 0.05 for univariate analysis) was used to find differential metabolites between two groups. The subject working characteristic curves and pathway enrichment analysis were used to screen metabolites and pathways associated with NSCLC-LM. Results: The PCA model and OPLS-DA model showed good overall data quality. Thirty endogenous differential metabolites were screened, and six potential biomarkers were further identified, including tyrosine (t = 3.37, P = 0.024, AUC = 0.967), phenylalanine (t = 3.98, P < 0.001, AUC = 0.992), pyruvate (t = 4.48, P < 0.001, AUC = 0.976), tryptophan (t = −2.5, P = 0.014, AUC = 0.935), adenosine monophosphate (t = −6.13, P < 0.001, AUC = 0.932) and glucose (t = −4.00, P < 0.001, AUC = 0.993). Thirty differential metabolites screened were subjected to metabolic pathway e...