MiR-106B Represents an Innovative Bio-marker in Cholangiocarcinoma Detection

Serum levels of miR-106b in CCA patients and healthy control were evaluated by quantitative real-time polymerase chain reaction (qRT-PCR). Chi-square test was used to analyze the association of miR-106b with the clinicopathological features. To evaluate the diagnostic value of miR-106b in CCA, the ROC curve was constructed.


Results
The expression of miR-106b was signi cantly increased in CCA samples compared with the healthy controls (P < 0.001). The overexpression of miR-106b was remarkable correlated with the lymphatic node metastasis (P = 0.038), clinical stage (P = 0.017) and differentiation (P = 0.009). ROC curve suggested that miR-106b was an effective diagnostic biomarker in CCA with the AUC of 0.913. The optimal cutoff value was 2.525, with the sensitivity of 89.7% and the speci city of 79.3%.

Conclusions
MiR-106b functions as an oncogene in CCA, which may be an potential diagnostic biomarker for CCA. Background Cholangiocarcinoma (CCA) is the second common primary hepatic malignancy, with the increased incidence around the world over recent decades [1,2]. Based on the tumor location, CCA can be classi ed into two types, including intrahehepatic cholangiocarcinoma (ICCA) and extrahepatic cholangiocarcinoma (ECCA) [3]. Surgical resection is the optimal strategy for this deadly disease [4].
However, most of patients are diagnosed at an advanced stage, missing the optimal time for operation and leading to poor survival [5,6] Early detection is di cult in CCA, due to the silent nature of the cancer at early stage and the lack of effective diagnostic tools [7,8]. The conventional blood-based biomarkers, such as carcinoembryonic antigen, carbohydrate antigen (CA) 19 − 9, CA242, exhibit low sensitivity and speci city in CCA diagnosis [9]. Therefore, nding novel diagnostic biomarkers for CCA is crucial to survival for patients with this disease. MicroRNAs (miRNAs) are a group of small, single stranded non-coding RNAs with approximately 22 nucleotides in length. MiRNAs play important roles in various cellular and biological processes by regulating gene expression at transcriptional level [10,11]. The aberrant expression of miRNAs is accepted a biomarker for several diseases like cancers [12]. For example, a related study scheduled by Wang et al. suggested that serum levels of miR-26a were signi cantly different between CCA patients and healthy individuals, which might serve as a indicator for CCA formation [13]. The expression pro le of miRNAs shows a signi cant association with tumor development and progression, suggesting their potential as biomarkers for cancer diagnosis and prognosis [14].
In the present study, we aimed to assess the diagnostic signi cance of miR-106b in the CCA patients. The serum level of miR-106b was detected in CCA patients, as well as its association with clinical characteristics. The present study might explore a novel predictor for CCA diagnosis.

Patients and specimens collection
The permission for this study was obtained from the ethics committee of the PLA Rocket Force Characteristic Medical Center. All the individuals collected in this study provided the written informed consent before sampling.
156 patients who were pathologically diagnosed with CCA in the PLA Rocket Force Characteristic Medical Center were collected in the study. None of the patients had received chemotherapy or radiotherapy before blood collection. In addition, 58 healthy volunteers who visited the hospital for routine checkup and have no history of carcinoma were enrolled in the study as healthy controls. The patients group and healthy controls were age and gender matched.
5 ml blood specimens were collected from the participants, and stored in EDTA tubes. Then the samples were centrifuged at 2000 g for 15 min to separate serum specimens. The serum specimens were stored at -20℃ for until use.
The clinicopathological features of CCA patients were recorded, including age, gender, tumor size, lymphatic node metastasis, clinical stage, histological type and differentiation. All these data were listed in Table 1.

Results
The expression level of miR-106b The expression levels of miR-106b in 156 CCA patients and 58 healthy volunteers were measured by qRT-PCR. The results suggested that the expression of miR-106b was signi cantly increased in the CCA patients compared with the healthy controls, and this differentiation was statistically signi cant (P 0.001) (Fig. 1).

Association of miR-106b with clinicopathological characteristics in CCA patients
The patients diagnosed with CCA were divided into high expression group (n = 89) and low expression group (n = 67), according to their median serum levels of miR-106b. In addition, we examined the relationship between miR-106b levels and the clinicopathological data of CCA patients using chi-square test. The results revealed that miR-106b expression was signi cantly correlated with the lymphatic node metastasis (P = 0.038), clinical stage (P = 0.017) and differentiation (P = 0.009). However, there was no signi cant association between miR-106b expression and age, gender, tumor size or histological type (all P > 0.05) ( Table 1).

Estimation of the diagnostic value of miR-106b in patients with CCA
In order to evaluate the diagnostic accuracy of miR-106b in CCA, ROC curve was constructed in this study. ROC curve showed that the area under the curve (AUC) was 0.913, suggesting the high diagnostic accuracy of miR-106b in CCA. The cutoff value of miR-106b for CCA diagnosis was 2.525, with the sensitivity of 89.7% and the speci city of 79.3% (Fig. 2).

Discussion
CCA is a highly malignant cancer, which originates from ductular epithelium of biliary tree [21]. Evidences show that the incidence and mortality have been on the rise, especially in several Southeast Asian countries [22]. Due to the asymptomatic nature and inability to early diagnosis, CCA patients are usually diagnosed at the advanced stage, leading the high mortality and poor prognosis [7]. What is worse, CCA is insensitivity to chemotherapy, making the treatments more di cult [23]. Thus, novel targets for diagnosis and treatments are urgently needed.
MiRNAs are a class of small and nocoding RNAs, which have been reported to be involved in the progression of various human cancers [24]. The expression levels of miRNAs are signi cantly correlated with tumor formation and development, which can serve as biomarkers for cancer diagnosis and prognosis [14]. In the previous tumor investigation, a variety of miRNAs were reported to be involved in CCA tumorigenesis. Wang et al. demonstrated that miR-26b as an oncogene could promote aggressive tumor progression of CCA, leading to poor survival [13]. Liu et al. reported that overexpression of miR-122 in vitro could inhibit CCA cell proliferation, invasion and promote apoptosis, which might be a potential target for CCA diagnosis and treatments [25]. Based on the related analysis results, we speculated that miRNAs played important roles in tumorigenesis of CCA, which might improve the accuracy of early detection.
MiR-106b, a member of the miR-106b-25 cluster, is proved to serve as an oncogene in several human cancers, including hepatocellular carcinoma [18], renal cell carcinoma [26], bladder cancer [27]. However, in gastric cancer and melanoma, miR-106b could inhibit malignant tumor development [28,29]. All the related studies might reveal that miR-106b played distinct roles in various malignancies. In the present study, we examined the expression of miR-106b in CCA patients and healthy controls. The analysis results demonstrated that the expression of miR-106b was upregulated in CCA patients compared with healthy controls. Moreover, the elevated miR-106b expression was signi cantly associated with positive lymphatic node metastasis, advanced clinical stage and poor differentiation. These experimental data might suggest that miR-106 was an oncogene in CCA. The conclusion was consistent with the previous studies. Amy et al. reported that the miR-106b expression in the CCA samples was signi cantly higher than that in the normal samples [30]. The study carried out by Li et al. demonstrated that miR-106b could regulate biological behaviors of renal cell carcinoma cell, thus promoting cancer progression [26]. However, the mechanisms for miR-106b regulating CCA were not investigated in the present study. Further researches were needed.
MiR-106b has been reported to be a biomarker for several cancers. Staby et al. suggested that upregulated miR-106b predicted early metastasis for renal cell carcinoma patients who were treated with nephrectomy [31]. The study scheduled by Li et al. indicated that miR-106b might serve as a promising prognostic biomarker for hepatocellular carcinoma patients. The elevated levels of miR-106b predicted poor prognosis for the patients [32]. Zheng et al. reported that plasma level of miR-106b showed high accuracy in breast cancer diagnosis [33]. All the related researches might reveal that miR-106b was a useful biomarker for cancers. In the present study, ROC curve was constructed to evaluate the diagnostic value of miR-106b in CCA. Analysis results demonstrated that miR-106b could distinguish the CCA patients from healthy individuals with high sensitivity and speci city.

Conclusions
In summary, the expression level of miR-106b is upregulated in the CCA patients, and associated with aggressive clinical characteristics. MiR-106b may be a potential diagnostic biomarker for CCA.

List Of Abbreviations
Cholangiocarcinoma ( The subjects had been informed the objective. Certainly, written consents were signed by every subject in this study.
Consent for publication The relative expression level of miR-106b evaluated by qRT-PCR in CCA patients and healthy controls. Analysis results suggested that serum levels of miR-106b were signi cantly up-regulated in CCA patients, compared with healthy controls. ***: indicated P<0.001.

Figure 1
The relative expression level of miR-106b evaluated by qRT-PCR in CCA patients and healthy controls.
Analysis results suggested that serum levels of miR-106b were signi cantly up-regulated in CCA patients, compared with healthy controls. ***: indicated P<0.001.

Figure 1
The relative expression level of miR-106b evaluated by qRT-PCR in CCA patients and healthy controls.
Analysis results suggested that serum levels of miR-106b were signi cantly up-regulated in CCA patients, compared with healthy controls. ***: indicated P<0.001.  showed high diagnostic value for CCA patients, with the AUC value of 0.913, combing with the sensitivity of 89.7% and the speci city of 79.3%. The cut-off value of miR-106b for CCA diagnosis was 2.525.