Identification of miRNA-103 in the cellular fraction of human peripheral blood as a potential biomarker for malignant mesothelioma--a pilot study

Abstract

Background: To date, no biomarkers with reasonable sensitivity and specificity for the early detection of malignant mesothelioma have been described. The use of microRNAs (miRNAs) as minimally-invasive biomarkers has opened new opportunities for the diagnosis of cancer, primarily because they exhibit tumor-specific expression profiles and have been commonly observed in blood of both cancer patients and healthy controls. The aim of this pilot study was to identify miRNAs in the cellular fraction of human peripheral blood as potential novel biomarkers for the detection of malignant mesothelioma.

Methodology/principal findings: Using oligonucleotide microarrays for biomarker identification the miRNA levels in the cellular fraction of human peripheral blood of mesothelioma patients and asbestos-exposed controls were analyzed. Using a threefold expression change in combination with a significance level of p<0.05, miR-103 was identified as a potential biomarker for malignant mesothelioma. Quantitative real-time PCR (qRT-PCR) was used for validation of miR-103 in 23 malignant mesothelioma patients, 17 asbestos-exposed controls, and 25 controls from the general population. For discrimination of mesothelioma patients from asbestos-exposed controls a sensitivity of 83% and a specificity of 71% were calculated, and for discrimination of mesothelioma patients from the general population a sensitivity of 78% and a specificity of 76%.

Conclusions/significance: The results of this pilot study show that miR-103 is characterized by a promising sensitivity and specificity and might be a potential minimally-invasive biomarker for the diagnosis of mesothelioma. In addition, our results support the concept of using the cellular fraction of human blood for biomarker discovery. However, for early detection of malignant mesothelioma the feasibility of miR-103 alone or in combination with other biomarkers needs to be analyzed in a prospective study.

Figure 1. Deregulated miRNAs in the cellular fraction of human blood of mesothelioma patients and asbestos-exposed controls.

Differences in miRNA expression were at least threefold. In patients with diagnosed malignant mesothelioma 15 miRNAs were up-regulated (green) and 34 miRNAs were down-regulated (red).

Figure 2. Heat map of miRNA expression of malignant mesothelioma patients and asbestos-exposed controls.

Cluster analysis was performed using miRNA expressions with more than a threefold change in the cellular fraction of human peripheral blood samples. Samples of malignant mesothelioma patients were marked blue and samples of asbestos-exposed controls yellow.

Figure 3. Box plots of relative expression of miR-103.

Relative expression of miR-103 in the cellular fraction of human peripheral blood of malignant mesothelioma patients (blue), asbestos-exposed controls (yellow), and controls from the general population (orange). Expression values were normalized to miR-125a and expressed as 2^−Ratio. Mann-Whitney test was performed to examine group differences.

Figure 4. Scatter plot of relative expression of miR-103 in histological subtypes of malignant mesothelioma.

Relative expression of miR-103 in the cellular fraction of human peripheral blood of patients with epithelioid and biphasic mesothelioma. One sarcomatoid mesothelioma case and three cases without available histological subtype were excluded from analysis. Expression values were normalized to miR-125a and expressed as 2^−Ratio.

Figure 5. Receiver operating characteristics (ROC) curves of miR-103.

The area under curve (AUC) was determined for miR-103 in the cellular fraction of human peripheral blood of (A) malignant mesothelioma patients and asbestos-exposed controls and (B) malignant mesothelioma patients and controls from the general population.

Figure 6. Scatter plots of relative expression of miR-103.

Relative expression of miR-103 in the cellular fraction of human peripheral blood of malignant mesothelioma patients (blue), asbestos-exposed controls (yellow), and controls from the general population (orange) regarding (A) gender, (B) smoking status, and (C) age. The smoking status of one MMP case was not available. Expression values were normalized to miR-125a and expressed as 2^−Ratio. Mann-Whitney tests were performed to examine differences between groups. Spearman correlation coefficient was calculated to evaluate association between miR-103 levels and age.