Differences in 1HNMR Based Metabolomic Patterns of Malignant and Benign Pleural Effusions.

Abstract

Objective: Malignant pleural effusion (MPE) is a common complication of lung cancer with poor prognosis. Benign pleural effusion (BPE), such as tuberculous and pneumonic pleural effusion, usually has a good prognosis. Differential diagnosis between MPE and BPE remains a clinical challenge.

Methods: 52 MPE, 93 BPE, and their corresponding serum samples were analyzed by hydrogen nuclear magnetic resonance (¹HNMR) based metabolomics.

Results: The ¹HNMR study showed that some amino acids and betaine in MPE are significantly altered in pleural effusion and serum compared to BPE patients. Levels of serum glucose and glutamine have strong positive correlation with those in pleural effusion (r>0.6) for MPE patients. The area under the receiver operating characteristic curve (AUROC) values of metabolites in pleural effusion or serum were less than 0.805 in differentiating MPE from BPE. Improved an AUROC value of 0.901 was observed using pleural effusion-serum ratios of glutamic acid in differentiating MPE from BPE, which was further validated by 15 double-blind samples.

Conclusions: Compared with BPE patients, amino acids and betaine in MPE are significantly altered in pleural effusion and serum. Pleural effusion-serum ratio of glutamic acid may contribute to the rapid diagnosis of MPE from BPE by ¹HNMR analysis.