BACKGROUND Several recent studies demonstrated that microRNAs (miRNAs) are stably detectable in plasma/serum. We hypothesised that plasma miRNAs concentrations contributed to potential biomarkers in patients with oesophageal squamous cell carcinoma (ESCC). METHODS We selected three oncogenic miRNAs (miR-21, miR-184, miR-221) and one tumour suppressive miRNA (miR-375), which are frequently reported in squamous cell carcinoma, as candidate targets for this plasma miRNA assay. This study was divided into three steps: (1) Determination of appropriate plasma miRNAs in preliminary tests. (2) Evaluation of whether the plasma miRNA assays could monitor tumour dynamics. (3) Validation study on the clinical application of plasma miRNA assays in 50 ESCC patients and 20 healthy volunteers. RESULTS (1) In preliminary tests, the plasma level of miR-21 was significantly higher (P=0.0218) and that of miR-375 (P=0.0052) was significantly lower in ESCC patients than controls. (2) The high plasma miR-21 levels reflected tumour levels in all cases (100%). The plasma level of miR-21 was significantly reduced in postoperative samples (P=0.0058). (3) On validation analysis, the plasma level of miR-21 tended to be higher in ESCC patients (P=0.0649), while that of miR-375 was significantly lower (P<0.0001) and the miR-21/miR-375 ratio was significantly higher (P<0.0001) in ESCC patients than in controls. The value of the area under the receiver-operating characteristic curve (AUC) was 0.816 for the miR-21/miR-375 ratio assay. Patients with a high plasma level of miR-21 tended to have greater vascular invasion (P=0.1554) and to show a high correlation with recurrence (P=0.0164). CONCLUSIONS Detection of circulating miRNAs might provide new complementary tumour markers for ESCC.