OBJECTIVE

Previous studies revealed that circulating (either from plasma or serum) long non-coding RNA may predict the occurrence or prognosis of multiple human malignant tumors. In this study, we mainly explored whether circulating lncRNAs can be utilized as biomarkers predicting the development of human esophageal squamous cell carcinoma (ESCC).

METHODS

LncRNA microarray was applied to screen the potential biomarkers for ESCC. Each group contained three individual plasma samples. A multi-stage validation and risk score formula detection were used for validation.

RESULTS

Eleven dysregulated lncRNAs were obtained after Venny analysis. Further validation in a larger cohort including 205 ESCC patients, 82 patients suffering from esophagus dysplasia and 210 healthy controls confirmed that increased Linc00152, CFLAR-AS1 and POU3F3 might be potential biomarkers for predicting the early progress with an area under curve (AUC) of 0.698, 0.651 and 0.584, respectively. The merged AUC of the three factors and merged with CEA was 0.765 and 0.955, respectively. We also revealed that circulating levels of three lncRNAs were associated with poor post-surgery prognosis of ESCC patients.

CONCLUSIONS

The three circulating lncRNAs might serve as potential biomarkers for predicting the early occurrence of ESCC.