Purpose: Long noncoding RNAs (lncRNAs) are present in body fluids, but their potential as tumor biomarkers has never been investigated in malignant pleural effusion (MPE) caused by lung cancer. The aim of this study was to assess the clinical significance of lncRNAs in pleural effusion, which could potentially serve as diagnostic and predictive markers for lung cancer-associated MPE (LC-MPE). 
Patients and methods: RNAs from pleural effusion were extracted in 217 cases of LC-MPE and 132 cases of benign pleural effusion (BPE). Thirty-one lung cancer-associated lncRNAs were measured using quantitative real-time polymerase chain reaction (qRT-PCR). The level of carcinoembryonic antigen (CEA) was also determined. The receiver operating characteristic (ROC) curves and the area under the ROC curve (AUC) were established to evaluate the sensitivity and specificity of the identified lncRNAs and other biomarkers. The correlations between baseline pleural effusion lncRNAs expression and response to chemotherapy were also analyzed.
Results: Three lncRNAs (MALAT1 , H19 , and CUDR ) were found to have potential as diagnostic markers in LC-MPE. The AUCs for MALAT1 , H19 , CUDR , and CEA were 0.891, 0.783, 0.824, and 0.826, respectively. Using a logistic model, the combination of MALAT1  and CEA (AUC, 0.924) provided higher sensitivity and accuracy in predicting LC-MPE than CEA (AUC, 0.826) alone. Moreover, baseline MALAT1  expression in pleural fluid was inversely correlated with chemotherapy response in patients with LC-MPE.
Conclusion: Pleural effusion lncRNAs were effective in differentiating LC-MPE from BPE. The combination of MALAT1  and CEA was more effective for LC-MPE diagnosis.
Keywords: malignant pleural effusion, lncRNA, MALAT1 , lung cancer, diagnosis