Circulating miR-200c and miR-141 and outcomes in patients with breast cancer

Abstract

Background: The deregulation of microRNAs in both tumours and blood has led to the search for microRNAs to indicate the presence of cancer and predict prognosis. We hypothesize the deregulation of miR-200c/miR-141 in the whole blood can identify breast cancer (BC), and could be developed into a prognostic signature.

Methods: The expression of miR-200c and miR-141 were examined in bloods (57 stage I-IV BC patients and 20 age-matched controls) by quantitative reverse-transcription PCR. The associations of circulating microRNAs with clinic and pathological characteristics were analysed. Their effects on survival were analysed by the Kaplan-Meier method and Cox regressions.

Results: MiR-200c was down regulated (P < 0.0001) in the blood of BC patients, yielded an area under the ROC curve of 0.79 (90% sensitivity, 70.2% specificity) in discriminating BC from controls. Circulating miR-141 was not discriminating. MiR-200c and miR-141 in the blood of BC patients were inversely correlated (P = 0.019). The miR-200c levels were numerically higher in stage IV and tumours with lower MIB-1. MiR-141 was significantly higher in the blood of patients with stage I-III, lymph node metastasis, and HER2 negative tumours. High blood expression of miR-200c and/or low expression of miR-141 was associated with unfavourable overall survival (hazard ratio, 3.89; [95% CI: 1.28-11.85]) and progression-free survival (3.79 [1.41-10.16]) independent of age, stage and hormonal receptors.

Conclusions: Circulating miR-200c and miR-141 were deregulated in BC comparing with controls. Furthermore, miR-200c and miR-141 were independent prognostic factors and associated with distinct outcomes of BC patients.

Figure 1

Real time PCR of miR-200c and miR-141 in blood samples. Plots depicting the relative expression for blood levels of miR-141 (A) and miR-200c (C) between patients and healthy controls, and between stage I-III patients, stage IV patients and healthy controls (B and D). The horizontal bar denotes the mean value for each group. The corresponding P values are provided in plots.

Figure 2

The role of blood miR-200c in breast cancer diagnosis. Receiver-operating characteristic (ROC) curve analysis (A) and odds ratio plot (B) using blood miR-200c expression levels for discriminating breast cancer patients (n = 57) and healthy controls (n = 20). ROC curves and odds ratio plots for discrimination of stage I-III BC from healthy controls (Cand D) and discrimination of stage I-II patients from healthy controls (E and F) are also shown.

Figure 3

miR-200c and miR-141 expression levels measured in peripheral blood are associated with poor prognosis in breast cancer patients. Kaplan-Meier curves showing (A and B) the progression-free survival (PFS) and (C and D) the overall survival (OS) of 57 breast cancer patients with high or low blood expression levels of microRNA. Continuous miRNA expression levels measured using RT-qPCR were converted to dichotomous variables using the Cutoff software (see text). The P values were computed using the Log-rank test.

Figure 4

Poor prognostic blood miRNA signature. Kaplan-Meier analysis and log-rank test showed that patients with higher levels of blood miR-200c and/or low levels of miR-141 had significantly poorer progression-free survival (A) and overall survival (B).

Figure 5

Blood miR-200c and/or miR-141 as prognostic factors in patients with breast cancer. Multivariate models showing the ability of high miR-200c levels, low miR-141 levels, and the combination of both as prognostic factors for predicting progression free survival (A) and overall survival (B) in breast cancer patients.