Tumor Biol. DOI 10.1007/s13277-015-3446-7
RESEARCH ARTICLE
Plasma microRNA might as a potential biomarker for hepatocellular carcinoma and chronic liver disease screening Li Jiang 1 & Xue Li 2 & Qi Cheng 3 & Bin-Hao Zhang 3
Received: 30 December 2014 / Accepted: 8 April 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015
Abstract Our study aims to investigate the expression signature of plasma microRNA-106b (miRNA-106b, miR-106b) in hepatocellular carcinoma (HCC) patients and chronic liver disease (CLD) patients compared with healthy controls and further evaluate the potential clinical value of miR-106b as biomarker in HCC detection. In addition, a meta-analysis was conducted to assess the diagnostic performance of miR106a/b as a biochemical marker for cancer screening. This study was divided into two phases. In the first phase, the expression levels of plasma miR-106b obtained from 108 subjects (47 HCC patients, 25 CLD patients, and 36 healthy controls) were measured by using qRT-PCR. Areas under receiver operating characteristic (ROC) curves (AUCs) were used to evaluate the diagnostic accuracy of plasma miR-106. In the second phase, a meta-analysis based on 11 previous researches as well as our current study was conducted to assess the potential clinical value of miR-106 in cancer detection. Plasma levels of miR-106b in HCC patients were significantly higher compared with CLD patients and healthy individuals. ROC curves suggested that plasma miR-106b yielded relative high sensitivities and specificities in differentiating HCC patients from CLD patients or healthy controls with
* Li Jiang [email protected] 1
Department of Biliary and Pancreatic Surgery, Affiliated Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No. 1095 Jiefang Street, Wuhan 430030, China
2
Department of Clinical Immunology, College of Laboratory Medicine, Tianjin Medical University, Tianjin 300203, China
3
Hepatic Surgery Center, Affiliated Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China
corresponding AUC values of 0.726 and 0.879, respectively. In addition, miR-106b showed a relatively high accuracy in distinguishing CLD patients from healthy controls with its AUC value of 0.703. Furthermore, the meta-analysis for diagnostic performance of miR-106a/b showed a pooled sensitivity of 0.74, specificity of 0.75, and an AUC of 0.81. Subgroup analysis based on samples types revealed a higher diagnostic performance of miR-106 for cancer detection by using nonblood samples. Similarly, miR-106 as biomarker showed a higher diagnostic accuracy for gastric cancer detection. We found that plasma miR-106b has clinical value in the detection of HCC from healthy people and CLD patients. Further largescale study may be needed to validate our findings. Keywords miR-106 . Plasma . Hepatocellular carcinoma . Chronic liver disease . Diagnosis
Introduction Hepatocellular carcinoma (HCC) is the major subtype of primary liver cancers (85–90 %), where it represents the fifth most common cancer and the third leading cause of cancerrelated death [1]. More than 500,000 cases and 600,000 HCCrelated deaths occur per year. Chronic hepatitis B and C viral (HBV and HCV) mainly contribute to the development of cirrhosis, nearly 80 to 90 % of HCC cases are reported to have cirrhosis as well [2]. The high mortality rate of HCC may attribute to the lack of efficient diagnostic methods and the delay of diagnosis. In general, patients with cancer at early stage are more likely to receive effective treatments and recuperate. For example, the overall 5-year survival rate for HCC is estimated to be 69 % if the patients are diagnosed at early stage, especially when the tumor is smaller than 2 cm [3, 4]. However, the overall 5-year survival rate drops to only 5–9 %
Tumor Biol.
due to the late diagnosis [5, 6]. The accurate diagnostic approaches for HCC may help in reducing its mortality rate. Histopathological biopsies are widely used in cancer diagnosis at present. For instance, liver biopsy has been regarded as gold standard for HCC diagnosis, but the invasive and uncomfortable nature, high expense, and potential sampling errors impede its widespread application in HCC detection. Moreover, imaging techniques such as ultrasonography, magnetic resonance imaging (MRI), and contrast-enhanced computed tomography (CT) have also limitations such as high maintenance cost and low sensitivity if tumor size is less than 1 cm [7]. It was discovered in the previous studies that serum alpha-fetoprotein (AFP) has potential to serve as biomarker for HCC detection, but the diagnostic accuracy of AFP is unsatisfactory with a low sensitivity (41–65 %) and relatively high specificity (80–94 %) of AFP [8]. Similarly, des-γcarboxy prothrombin (DCP) was shown to have 44–47 % accuracy to identify the HCC patients while their tumors are less than 3 cm in size [9, 10]. Thus, it’s necessary to explore the novel diagnostic approaches or biomarkers to increase the diagnosis accuracy. Over the past decades, evidences have indicated that microRNAs (miRNAs), a group of endogenous small noncoding RNAs, are of great clinical value in cancer detection. These molecules have been recognized as posttranscriptional regulators of gene expression [11]. MiRNAs were not only detectable in tissues but also in body fluids, hence, miRNAs could be obtained easily by using minimally invasive procedure [12]. Furthermore, miRNAs exhibit a remarkable stability in body fluids (such as plasma, serum, urine, cerebral spinal fluid, and gastric juice). The findings suggested that miRNAs might be served as potential, noninvasive, and high accurate biomarkers for tumor diagnostics [12, 13]. Among a variety of miRNAs, a number of researches have reported the potential relationship between aberrant miR106a/b expression and the prevalence rate of human cancers [14–16]. MiR-106a/b was identified to be linked to various cancers, and its expression level is associated with the stage of cancers [17, 18]. Previous studies have reported that miR106a/b as a noninvasive biomarker in detection of various cancers, including gastric cancer [14], lymphoma [15], and colorectal cancer [16]. However, few studies have been investigated the diagnostic performance of miR-106b in HCC diagnosis. Only one study reported by Shen et al. showed that the expression of miR-106b is evidently up-regulated in hepatoma cells compared with immortalized normal liver epithelial cells [19]. Therefore, this study aims to investigate the difference in the plasma level of miR-106b expression from HCC patients, CLD patients, and healthy controls. This study also examines the sensitivity and specificity of plasma miR106b for distinguishing HCC patients from healthy people or CLD patients.
Moreover, several studies indicated that miR-106a/b level has a significant association with cancers. However, other studies have found no evidence of association between miR106a/b and cancers. There were conflicting results among these studies. Hence, we also performed a systematic metaanalysis to comprehensively investigate the diagnostic performance of miR-106a/b as biomarkers in the cancer diagnosis.
Materials and methods Study design and patients Our study consisted of two phases. In the first phase, we measured the expression level of the plasma miR-106b in all subjects to assess the diagnostic performance of miR-106 for HCC. Forty-seven patients with HCC and 25 CLD patients without HCC were involved in this study, and they were all enrolled from the Affiliated Tongji Hospital, Tongji Medical College. Furthermore, 36 healthy individuals were randomly selected from 500 volunteers taking the routine physical examinations in the hospital. The clinical characteristics of all participants are presented in Table 1. The second phase of the study was to evaluate the diagnostic value of the miR-106a/b in a variety of cancers by a systematic meta-analysis. We conducted a computerized literature search for all related articles published before July 21, 2014 in PubMed, Excerpta Medica Database, Web of Science and Chinese National Knowledge Infrastructure (CNKI). A total of 83 articles were identified after the initial search. Eleven eligible articles meeting the following selection criteria were finally included in the present meta-analysis: (1) studies focused on the evaluation of the clinical value miR-106a/b in cancer diagnosis; (2) more than 20 subjects including patients with cancer and healthy controls participated in these studies; (3) the histopathological biopsy as gold standard was used to confirm the diagnosis of cancer patients; (4) studies provided valid data to allow for the reconstruction of two-by-two table. All subjects who participated in this study have to provide written informed consents. The present study was approved by the review board of Affiliated Tongji Hospital, Tongji Medical College ethics committee, and carried out in compliance with the international ethical guidelines for biomedical research involving human subjects. Plasma preparation and MiR-106 extraction First, 3–4 ml of peripheral venous blood was collected into BD Vacutainer spray-coated K2EDTA Tubes (BD, Franklin Lakes, USA) from subjects using standardized phlebotomy procedures. The blood samples of HCC patients were obtained before surgical operation or adjuvant therapy. Ethylenediaminetetraacetic acid (EDTA) was added into the samples to
Tumor Biol. Table 1 Clinicopathological characteristics of patients with hepatocellular carcinoma (HCC), patients with chronic liver disease, and healthy controls
Age (years) Gender Male Female HBV status Positive Negative ALT ≤40 U/L >40 U/L AFP Positive Negative Tumor size ≤3 cm >3 cm Tumor number 1 ≥2 Vein invasion Yes No TNM grade I–II III–IV
P value
HCC (n=47)
Chronic liver disease (n=25)
Healthy control (n=36)
56.1±10.3
52.6±8.9
54.6±11.2
0.391
36 (76.6 %) 11 (23.4 %)
19 (76.0 %) 6 (24.0 %)
27 (75.0 %) 9 (25.0 %)
0.986
37 (78.7 %) 10 (21.3 %)
22 (88.0 %) 3 (12.0 %)
6 (17.0 %) 30 (83.0 %)
RESEARCH ARTICLE
Plasma microRNA might as a potential biomarker for hepatocellular carcinoma and chronic liver disease screening Li Jiang 1 & Xue Li 2 & Qi Cheng 3 & Bin-Hao Zhang 3
Received: 30 December 2014 / Accepted: 8 April 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015
Abstract Our study aims to investigate the expression signature of plasma microRNA-106b (miRNA-106b, miR-106b) in hepatocellular carcinoma (HCC) patients and chronic liver disease (CLD) patients compared with healthy controls and further evaluate the potential clinical value of miR-106b as biomarker in HCC detection. In addition, a meta-analysis was conducted to assess the diagnostic performance of miR106a/b as a biochemical marker for cancer screening. This study was divided into two phases. In the first phase, the expression levels of plasma miR-106b obtained from 108 subjects (47 HCC patients, 25 CLD patients, and 36 healthy controls) were measured by using qRT-PCR. Areas under receiver operating characteristic (ROC) curves (AUCs) were used to evaluate the diagnostic accuracy of plasma miR-106. In the second phase, a meta-analysis based on 11 previous researches as well as our current study was conducted to assess the potential clinical value of miR-106 in cancer detection. Plasma levels of miR-106b in HCC patients were significantly higher compared with CLD patients and healthy individuals. ROC curves suggested that plasma miR-106b yielded relative high sensitivities and specificities in differentiating HCC patients from CLD patients or healthy controls with
* Li Jiang [email protected] 1
Department of Biliary and Pancreatic Surgery, Affiliated Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No. 1095 Jiefang Street, Wuhan 430030, China
2
Department of Clinical Immunology, College of Laboratory Medicine, Tianjin Medical University, Tianjin 300203, China
3
Hepatic Surgery Center, Affiliated Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China
corresponding AUC values of 0.726 and 0.879, respectively. In addition, miR-106b showed a relatively high accuracy in distinguishing CLD patients from healthy controls with its AUC value of 0.703. Furthermore, the meta-analysis for diagnostic performance of miR-106a/b showed a pooled sensitivity of 0.74, specificity of 0.75, and an AUC of 0.81. Subgroup analysis based on samples types revealed a higher diagnostic performance of miR-106 for cancer detection by using nonblood samples. Similarly, miR-106 as biomarker showed a higher diagnostic accuracy for gastric cancer detection. We found that plasma miR-106b has clinical value in the detection of HCC from healthy people and CLD patients. Further largescale study may be needed to validate our findings. Keywords miR-106 . Plasma . Hepatocellular carcinoma . Chronic liver disease . Diagnosis
Introduction Hepatocellular carcinoma (HCC) is the major subtype of primary liver cancers (85–90 %), where it represents the fifth most common cancer and the third leading cause of cancerrelated death [1]. More than 500,000 cases and 600,000 HCCrelated deaths occur per year. Chronic hepatitis B and C viral (HBV and HCV) mainly contribute to the development of cirrhosis, nearly 80 to 90 % of HCC cases are reported to have cirrhosis as well [2]. The high mortality rate of HCC may attribute to the lack of efficient diagnostic methods and the delay of diagnosis. In general, patients with cancer at early stage are more likely to receive effective treatments and recuperate. For example, the overall 5-year survival rate for HCC is estimated to be 69 % if the patients are diagnosed at early stage, especially when the tumor is smaller than 2 cm [3, 4]. However, the overall 5-year survival rate drops to only 5–9 %
Tumor Biol.
due to the late diagnosis [5, 6]. The accurate diagnostic approaches for HCC may help in reducing its mortality rate. Histopathological biopsies are widely used in cancer diagnosis at present. For instance, liver biopsy has been regarded as gold standard for HCC diagnosis, but the invasive and uncomfortable nature, high expense, and potential sampling errors impede its widespread application in HCC detection. Moreover, imaging techniques such as ultrasonography, magnetic resonance imaging (MRI), and contrast-enhanced computed tomography (CT) have also limitations such as high maintenance cost and low sensitivity if tumor size is less than 1 cm [7]. It was discovered in the previous studies that serum alpha-fetoprotein (AFP) has potential to serve as biomarker for HCC detection, but the diagnostic accuracy of AFP is unsatisfactory with a low sensitivity (41–65 %) and relatively high specificity (80–94 %) of AFP [8]. Similarly, des-γcarboxy prothrombin (DCP) was shown to have 44–47 % accuracy to identify the HCC patients while their tumors are less than 3 cm in size [9, 10]. Thus, it’s necessary to explore the novel diagnostic approaches or biomarkers to increase the diagnosis accuracy. Over the past decades, evidences have indicated that microRNAs (miRNAs), a group of endogenous small noncoding RNAs, are of great clinical value in cancer detection. These molecules have been recognized as posttranscriptional regulators of gene expression [11]. MiRNAs were not only detectable in tissues but also in body fluids, hence, miRNAs could be obtained easily by using minimally invasive procedure [12]. Furthermore, miRNAs exhibit a remarkable stability in body fluids (such as plasma, serum, urine, cerebral spinal fluid, and gastric juice). The findings suggested that miRNAs might be served as potential, noninvasive, and high accurate biomarkers for tumor diagnostics [12, 13]. Among a variety of miRNAs, a number of researches have reported the potential relationship between aberrant miR106a/b expression and the prevalence rate of human cancers [14–16]. MiR-106a/b was identified to be linked to various cancers, and its expression level is associated with the stage of cancers [17, 18]. Previous studies have reported that miR106a/b as a noninvasive biomarker in detection of various cancers, including gastric cancer [14], lymphoma [15], and colorectal cancer [16]. However, few studies have been investigated the diagnostic performance of miR-106b in HCC diagnosis. Only one study reported by Shen et al. showed that the expression of miR-106b is evidently up-regulated in hepatoma cells compared with immortalized normal liver epithelial cells [19]. Therefore, this study aims to investigate the difference in the plasma level of miR-106b expression from HCC patients, CLD patients, and healthy controls. This study also examines the sensitivity and specificity of plasma miR106b for distinguishing HCC patients from healthy people or CLD patients.
Moreover, several studies indicated that miR-106a/b level has a significant association with cancers. However, other studies have found no evidence of association between miR106a/b and cancers. There were conflicting results among these studies. Hence, we also performed a systematic metaanalysis to comprehensively investigate the diagnostic performance of miR-106a/b as biomarkers in the cancer diagnosis.
Materials and methods Study design and patients Our study consisted of two phases. In the first phase, we measured the expression level of the plasma miR-106b in all subjects to assess the diagnostic performance of miR-106 for HCC. Forty-seven patients with HCC and 25 CLD patients without HCC were involved in this study, and they were all enrolled from the Affiliated Tongji Hospital, Tongji Medical College. Furthermore, 36 healthy individuals were randomly selected from 500 volunteers taking the routine physical examinations in the hospital. The clinical characteristics of all participants are presented in Table 1. The second phase of the study was to evaluate the diagnostic value of the miR-106a/b in a variety of cancers by a systematic meta-analysis. We conducted a computerized literature search for all related articles published before July 21, 2014 in PubMed, Excerpta Medica Database, Web of Science and Chinese National Knowledge Infrastructure (CNKI). A total of 83 articles were identified after the initial search. Eleven eligible articles meeting the following selection criteria were finally included in the present meta-analysis: (1) studies focused on the evaluation of the clinical value miR-106a/b in cancer diagnosis; (2) more than 20 subjects including patients with cancer and healthy controls participated in these studies; (3) the histopathological biopsy as gold standard was used to confirm the diagnosis of cancer patients; (4) studies provided valid data to allow for the reconstruction of two-by-two table. All subjects who participated in this study have to provide written informed consents. The present study was approved by the review board of Affiliated Tongji Hospital, Tongji Medical College ethics committee, and carried out in compliance with the international ethical guidelines for biomedical research involving human subjects. Plasma preparation and MiR-106 extraction First, 3–4 ml of peripheral venous blood was collected into BD Vacutainer spray-coated K2EDTA Tubes (BD, Franklin Lakes, USA) from subjects using standardized phlebotomy procedures. The blood samples of HCC patients were obtained before surgical operation or adjuvant therapy. Ethylenediaminetetraacetic acid (EDTA) was added into the samples to
Tumor Biol. Table 1 Clinicopathological characteristics of patients with hepatocellular carcinoma (HCC), patients with chronic liver disease, and healthy controls
Age (years) Gender Male Female HBV status Positive Negative ALT ≤40 U/L >40 U/L AFP Positive Negative Tumor size ≤3 cm >3 cm Tumor number 1 ≥2 Vein invasion Yes No TNM grade I–II III–IV
P value
HCC (n=47)
Chronic liver disease (n=25)
Healthy control (n=36)
56.1±10.3
52.6±8.9
54.6±11.2
0.391
36 (76.6 %) 11 (23.4 %)
19 (76.0 %) 6 (24.0 %)
27 (75.0 %) 9 (25.0 %)
0.986
37 (78.7 %) 10 (21.3 %)
22 (88.0 %) 3 (12.0 %)
6 (17.0 %) 30 (83.0 %)
