PARINT-01415; No of Pages 7 Parasitology International xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Parasitology International journal homepage: www.elsevier.com/locate/parint
Urinary microRNA-192 and microRNA-21 as potential indicators for liver fluke-associated cholangiocarcinoma risk group Runglawan Silakit a,g, Watcharin Loilome a,g, Puangrat Yongvanit a,g, Suyanee Thongchot a,g, Paiboon Sithithaworn b,g, Thidarut Boonmars b,g, Supinda Koonmee c,g, Attapol Titapun d,g, Narong Khuntikeo d,g, Nittaya Chamadol e,g, Anchalee Techasen f,g, Nisana Namwat a,g,⁎ a
Department of Biochemistry, Faculty of Medicine, Khon Kaen University, Khon Kaen, 40002, Thailand Department of Parasitology, Faculty of Medicine, Khon Kaen University, Khon Kaen, 40002, Thailand c Department of Pathology, Faculty of Medicine, Khon Kaen University, Khon Kaen, 40002, Thailand d Department of Surgery, Faculty of Medicine, Khon Kaen University, Khon Kaen, 40002, Thailand e Department of Radiology, Faculty of Medicine, Khon Kaen University, Khon Kaen, 40002, Thailand f Center for Research and Development of Medical Diagnostic Laboratories, Faculty of Associated Medical Science, Khon Kaen University, Khon Kaen, 40002, Thailand g Liver Fluke and Cholangiocarcinoma Research Center, Khon Kaen University, Khon Kaen, 40002, Thailand b
a r t i c l e
i n f o
Article history: Received 31 August 2015 Received in revised form 23 September 2015 Accepted 1 October 2015 Available online xxxx Keywords: Cholangiocarcinoma Opisthorchis viverrini Periductal fibrosis Urinary microRNAs Risk indicator
a b s t r a c t Opisthorchis viverrini infection induces chronic inflammation in the bile ducts, leading to periductal fibrosis (PDF), which possibly associates to cholangiocarcinoma (CCA). Patients with CCA have a poor prognosis, which is linked to asymptomatic disease and late diagnosis. Hence, detecting early stage CCA is essential. Secretory miRNAs have been promoted as biomarkers for pathological changes associated with parasitic infections, fibrosis and/or cancer. We aimed to determine levels of miR-192 and miR-21 in the urine of O. viverrini infected, periductal fibrosis (PDF) and CCA groups using qRT-PCR. We found that miR-192 was significantly higher in O. viverrini infected, PDF and also CCA groups (p b 0.05) than in healthy controls. By utilizing the Receiver Operation Characteristics (ROC) analysis, miR-192 differentiated patients with opisthorchiasis (the area under the curve; AUC = 0.766), PDF subjects (AUC = 0.781) and CCA patients (AUC = 0.682) from healthy controls. MiR-21 was significantly higher in PDF and CCA groups (p b 0.05) than in healthy controls. MiR-21 discriminated PDF subjects (AUC = 0.735) and CCA patients (AUC = 0.682) from healthy controls. Combined levels of these two miRNAs revealed an increased AUC of 0.812 for separating opisthorchiasis, AUC of 0.815 in discriminating PDF subjects, and AUC of 0.849 in differentiating CCA from healthy controls. Odds ratios (OR) indicated high levels of miR-192/miR-21 as risk predictors for opisthorchiasis, PDF and CCA. Levels of these miRNAs declined significantly for patients following praziquantel treatment. In conclusion, urinary miR-192/miR-21 have potential as risk indicators for opisthorchiasis and PDF-associated CCA in the endemic region. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Infection with the food-borne trematode Opisthorchis viverrini leads to chronic inflammation in the host's biliary tract. Importantly, this parasite is categorized as a relative risk factor for cholangiocarcinoma (CCA), which is a malignant cancer of the bile duct epithelium, prevalent in northeastern Thailand [1,2]. Response to the inflammation induced by O. viverrini infection results in a progressive remodeling and destruction of the tissue architecture of the bile duct epithelium, leading to the development of periductal fibrosis (PDF) [1,3]. The fibrotic deposition along the biliary epithelium is a precursor event to the development of CCA observed for animal models [4]. The persistence of PDF ⁎ Corresponding author at: Department of Biochemistry, Faculty of Medicine, Khon Kaen University, Khon Kaen, 40002, Thailand. E-mail address:
[email protected] (N. Namwat).
may ultimately contribute to CCA carcinogenesis even after praziquantel treatment [5]. Therefore, PDF is a risk marker when screening for early CCA using ultrasonography [6,7]. Nevertheless, the successful use of ultrasonography for screening depends on the skill of the investigator who performs the study and assesses the findings [8]. Consequently, identifying a non-invasive biomarker for early detection of CCA is desirable. Recent findings indicate that a class of small, non-coding RNA, termed microRNA (miRNA), acts as a negative regulator of gene expression at the post-transcriptional level [9,10]. Aberrant expression of miRNA in humans appears to be associated with infections with various parasites [11–13]. Several studies review miRNAs as biomarkers for the diagnosis of parasitic diseases [14], also indicating the response to chemotherapy [15]. Recently, miRNA genes in O. viverrini were identified [16]. MiR-21, which is linked to inflammation-related cancers, was overexpressed in O. viverrini-induced CCA in the hamster model
http://dx.doi.org/10.1016/j.parint.2015.10.001 1383-5769/© 2015 Elsevier Ireland Ltd. All rights reserved.
Please cite this article as: R. Silakit, et al., Urinary microRNA-192 and microRNA-21 as potential indicators for liver fluke-associated cholangiocarcinoma risk group, Parasitology International (2015), http://dx.doi.org/10.1016/j.parint.2015.10.001
2
R. Silakit et al. / Parasitology International xxx (2015) xxx–xxx
and in patients [17]. In addition, miR-192, a profibrotic-related miRNA, was found to be increased in the sera from PDF subjects and CCA patients [18]. These data suggest that miR-21 and miR-192 could potentially be used as a non-invasive biomarker for screening the risk group of CCA. Urine is one of the most readily accessible biological fluids, and it harbors numerous biologically significant molecules, including miRNAs. The miRNAs in urine exist in a stable form, even after exposure to seven cycles of freezing and incubation at room temperature for 72 h [19]. Thus, the use of urinary miRNAs as biomarkers for screening the risk group of CCA might have promise. In this study, we studied urinary miR-192 and miR-21 in O. viverrini-infected and PDF subjects and CCA patients as possible risk indicators for CCA screening. The miRNA levels in patients before and after treatment of O. viverrini-infected subjects with praziquantel were also assessed as surrogate markers. 2. Materials and methods 2.1. Subjects and samples A total of 105 subjects were included in this study; 21 were healthy persons who had no history of O. viverrini infection, were ultrasoundnegative and had normal urinalysis results during 2012–2014; 24 subjects were diagnosed as having O. viverrini infection determined by the presence of parasite eggs in the feces; 28 subjects were diagnosed as having PDF by abdominal ultrasonography; 10 subjects were diagnosed as having other diseases associated with inflammation, including cholelithiasis, hepatic steatosis, liver cyst, nephrolithiasis, renal cyst, hydronephrosis and haemangioma. Patients with cancers rather than CCA were not included in this study. CCA was diagnosed in 22 patients, all of who were admitted to Srinagarind Hospital, Faculty of Medicine, Khon Kaen University, Thailand. After urine collection, urinalysis was performed to determine the following parameters: glucose, bilirubin, ketone, specific gravity, hemoglobin, pH, protein, urobilinogen, nitrite and leukocytes. The remaining sample was centrifuged at 1000 × g for 10 min at 4 °C to precipitate cellular debris; then, supernatants were aliquoted and frozen at −80 °C until use. The Ethics Committee of the Faculty of Medicine, Khon Kaen University approved the study protocol, according to Helsinki criteria (HE561039 and HE571283). Informed consent for obtaining and using urine samples was obtained from all subjects and patients. The characteristics of study subjects are listed in Table 1. 2.2. Purification of urinary miRNAs Total RNAs were extracted from individual urine samples using the miRNeasy mini kit (Qiagen, Hilden, Germany) according to the
instructions provided by the manufacturer. Briefly, 1000 μl of QIAzol lysis reagent were added to 200 μl of urine. The sample was mixed and incubated at 15–25 °C for 5 min, then 200 μl of chloroform were added and the tube was vortexed. Next, samples were centrifuged at 12,000 × g for 15 min at 4 °C. After transferring the upper aqueous phase to a new tube, a 1.5 volume of 100% (v/v) ethanol was added. The mixture was then applied to a binding column. After washing twice, the RNA sample was then eluted from the membrane into a collection tube with 35 μl of elution buffer or RNase free water.
2.3. Reverse transcription and real-time quantitative RT-PCR The levels of urinary mature miRNAs originated from human miRNA sequences (hsa-miR-192; assay ID 000491, hsa-miR-21; assay ID 000397) quantified using TaqMan MicroRNA Assay Kit (Applied Biosystems, Foster City, CA, USA). Firstly, the reverse transcription (RT) reaction using 5 μl of extracted RNA was carried out using the TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA), according to the instructions supplied by the manufacturer. After cDNA synthesis, real-time quantitative RT-polymerase chain reaction (qRT-PCR) was conducted employing an Applied Biosystems 7500 Fast Real-time RCR system (Applied Biosystems). The thermal cycling conditions were according to the manufacturer's instructions, and all reactions were performed in duplicate. To compare the relative levels of urinary miRNAs, miR-16 was used as a loading control. The 2−ΔCt method was used to calculate the relative level of urinary miRNAs.
2.4. Statistical analysis IBM SPSS version 19.0 software (Chicago, IL, USA) was used for all statistical analyses. Significant differences in the levels of urinary miRNAs between groups were assessed by the Mann–Whitney U-test. To evaluate the differentiating power of urinary miRNA analysis for its ability to differentiate subject groups from healthy controls, receiver operating characteristic (ROC) analysis and the areas under the ROC curve (AUC) were performed. The cut-off values were selected to measure optimal sensitivity and specificity. To examine urinary miRNAs levels as predictors for risk factors and disease outcome, a multivariate logistic regression analysis was performed using both crude and adjusted odds ratios. The association of urinary miRNAs and the clinicopathological parameters of CCA patients were analyzed using the Fisher's exact test. Wilcoxon's signed-rank test was performed for paired samples. A P-value of less than 0.05 was designated as a statistically significant difference.
Table 1 Characteristics of the study subjects. Characteristics
Healthy
O. viverrini infected
PDF
CCA
Other diseases
Total (n) Sex Male Female Age 30–39 40–49 50–59 ≥60 Histology type Non-papillary type Papillary type Overall metastasis Negative Positive
21
24
28
22
10
6 (28.6%) 15 (71.4%)
12 (50.0%) 12 (50.0%)
10 (35.7%) 18 (64.3%)
11 (50.0%) 11 (50.0%)
2 (20.0%) 8 (80.0%)
1 (4.8%) 6 (28.6%) 7 (33.3%) 7 (33.3%)
0 (0.0%) 7 (29.2%) 7 (29.2%) 10 (41.6%)
0 (0.0%) 10 (35.7%) 7 (25.0%) 11 (39.3%)
0 (0.0%) 0 (0.0%) 11 (50.0%) 11 (50.0%)
0 (0.0%) 2 (20.0%) 4 (40.0%) 4 (40.0%)
13 (59.1%) 9 (40.9%) 16 (72.7%) 6 (27.3%)
PDF, periductal fibrosis; CCA, cholangiocarcinoma.
Please cite this article as: R. Silakit, et al., Urinary microRNA-192 and microRNA-21 as potential indicators for liver fluke-associated cholangiocarcinoma risk group, Parasitology International (2015), http://dx.doi.org/10.1016/j.parint.2015.10.001
R. Silakit et al. / Parasitology International xxx (2015) xxx–xxx
3
Fig. 1. Urinary miR-192 level in healthy subjects, O. viverrini-infected subjects, subjects with PDF, CCA patients and patients with other inflammatory diseases (A). Data are presented as relative expression (normalized with miR-16). The receiver operating characteristics (ROC) curves in opisthorchiasis relative to healthy subjects (B), subjects with PDF relative to healthy subjects (C), and CCA patients relative to healthy subjects (D). Data are expressed as the mean ± standard deviation (SD). Mann–Whitney U-tests were used to compare the different levels of urinary miRNAs between individual subject groups (O. viverrini infected, PDF or CCA) and the healthy control group. *P b 0.05. **P b 0.01. ***P b 0.001.
3. Results 3.1. Urinary miR-192 and miR-21 levels are potential indicators for risk groups and CCA detection The urinary miR-192 and miR-21 levels were examined by qRT-PCR in healthy subjects, O. viverrini-infected subjects, subjects with PDF, CCA
patients and subjects with other inflammatory diseases. To calculate the relative level of these miRNAs, miR-16 was used as a loading control. Intra-assay deviation of urinary miR-16 levels was performed, and the average Ct values were 30.0 ± 1.7 for the same samples analyzed on 4 consecutive days. The results indicated that the urinary miR-192 level was significantly higher in O. viverrini-infected subjects (P = 0.002), subjects with PDF (P = 0.001) and CCA patients (P = 0.041, Fig. 1A)
Table 2 Results of area under the ROC curve (AUC), sensitivity and specificity for the optimal cut-off for the subjects. Comparative groups
MiR-192 Healthy: Ov-infected Healthy: PDF Healthy: CCA Healthy: other diseases Ov-infected: PDF Ov-infected: CCA PDF: CCA Other diseases: CCA MiR-21 Healthy: Ov-infected Healthy: PDF Healthy: CCA Healthy: other diseases Ov-infected: PDF Ov-infected: CCA PDF: CCA Other diseases: CCA Combined miRNAs Healthy: Ov-infected Healthy: PDF Healthy: CCA Healthy: other diseases Ov-infected: PDF Ov-infected: CCA PDF: CCA Other diseases: CCA
ROC curve parameters
P value
Cut-off value
AUC (95% CI)
Sensitivity %
Specificity %
1.038 1.055 0.936 1.055 2.531 1.796 1.658 0.645
0.766 (0.615–0.917) 0.781 (0.644–0.917) 0.682 (0.516–0.848) 0.591 (0.363–0.818) 0.505 (0.345–0.664) 0.525 (0.352–0.697) 0.534 (0.366–0.702) 0.632 (0.440–0.823)
75.0 75.0 63.6 60.0 57.1 54.6 54.6 30.0
71.4 71.4 66.7 71.4 54.2 66.7 67.9 72.7
0.002 0.001 0.041 0.422 0.956 0.775 0.682 0.238
0.943 1.277 1.355 1.718 1.207 1.324 1.730 1.750
0.560 (0.389–0.730) 0.735 (0.592–0.877) 0.682 (0.521–0.842) 0.738 (0.548–0.928) 0.612 (0.452–0.771) 0.593 (0.427–0.758) 0.515 (0.346–0.685) 0.505 (0.293–0.716)
58.3 71.4 63.6 60.0 75.0 63.6 63.6 60.0
47.6 71.4 71.4 85.7 54.2 54.2 57.1 63.6
0.495 0.005 0.041 0.035 0.168 0.281 0.853 0.968
2.534 2.760 2.402 2.317 3.933 3.431 4.440 2.883
0.812 (0.681–0.943) 0.815 (0.690–0.940) 0.849 (0.725–0.972) 0.720 (0.520–0.918) 0.539 (0.380–0.697) 0.540 (0.370–0.710) 0.502 (0.340–0.664) 0.705 (0.522–0.887)
79.2 78.6 81.8 70.0 60.7 68.2 54.6 60.0
76.2 81.0 71.4 66.7 54.2 50.0 53.6 77.3
b 0.001 b 0.001 b 0.001 0.052 0.633 0.644 0.984 0.067
Ov, O. viverrini; PDF, periductal fibrosis; CCA, cholangiocarcinoma; AUC, area under the ROC curve; CI, confidence interval. P value less than 0.05 is considered statistically significant.
Please cite this article as: R. Silakit, et al., Urinary microRNA-192 and microRNA-21 as potential indicators for liver fluke-associated cholangiocarcinoma risk group, Parasitology International (2015), http://dx.doi.org/10.1016/j.parint.2015.10.001
4
R. Silakit et al. / Parasitology International xxx (2015) xxx–xxx
compared with healthy subjects. The receiver-operating characteristic (ROC) curve analysis revealed that urinary miR-192 might be used as a risk indicator for distinguishing healthy subjects from O. viverriniinfected subjects, with an AUC of 0.766 (95% CI, 0.615–0.917; P = 0.002, Fig. 1B). Urinary miR-192 also discriminated between healthy and PDF subjects with an AUC of 0.781 (95% CI, 0.644–0.917; P = 0.001, Fig. 1C). Moreover, urinary miR-192 may have the potential to be used as a differentiating marker to discriminate healthy subjects from CCA patients (AUC of 0.682, 95% CI, 0.516–0.848; P = 0.041, Fig. 1D). At the optimal cut-off value of urinary miR-192 for the detection of O. viverrini infection, sensitivity was 75.0% and specificity 71.4%. For identifying subjects with PDF, sensitivity was 75.0% and specificity 71.4%; for detecting CCA, sensitivity was 63.6% and specificity 66.7% (Table 2). Urinary miR-21 (Fig. 2A) levels were significantly higher in subjects with PDF (P = 0.005), CCA patients (P = 0.041) and patients with other inflammatory diseases (P = 0.035) than in healthy subjects. Notably, urinary miR-21 tended to be higher in O. viverrini-infected subjects than that in healthy controls, albeit not significantly. The analysis of data from a larger number of O. viverrini-infected subjects should be undertaken in future. The AUC between subjects with PDF and healthy subjects was 0.735 (95% CI, 0.592–0.877; P = 0.005, Fig. 2B), whereas the AUC between CCA patients and healthy subjects was 0.682 (95% CI, 0.521–0.842; P = 0.041, Fig. 2C). At the optimal cut-off value of urinary miR-21 for distinguishing subjects with PDF from healthy subjects, sensitivity was 71.4% and specificity was 71.4%. For the differentiation of CCA from the healthy group, sensitivity was 63.6% and specificity was 71.4% (Table 2).
3.2. Differentiating power of a combination of urinary miR-192 and miR-21 We estimated the power of differentiation between different disease states using a combination of urinary miR-192 and miR-21 (Fig. 3A). The combined levels of urinary miR-192 and miR-21 yielded an AUC of 0.812 (95% CI, 0.681–0.943; P b 0.001, Fig. 3B), with 79.2% sensitivity and 76.2% specificity for separating O. viverrini-infected subjects from healthy subjects, and an AUC of 0.815 (95% CI, 0.690–0.940; P b 0.001, Fig. 3C) with 78.6% sensitivity and 81.0% specificity for discriminating between subjects with PDF and healthy subjects. The urinary miR-21 was significantly higher in patients with other inflammatory diseases when compared with healthy subjects. The combined level of urinary miR-192 and miR-21 also improved the differentiating power between CCA patients and healthy subjects, resulting in an AUC of 0.849 (95% CI, 0.725–0.972; P b 0.001, Fig. 3D), with a sensitivity of 81.8% and a specificity of 71.4%, showing an enhanced effect in the diagnostic value of these two miRNAs for the diagnosis of CCA. There were no significant correlation between combined levels of urinary miR-192 or miR-21 and the clinico-pathological findings for CCA patients (data not shown). 3.3. Predictive values (odds ratios, ORs) of urinary miR-192 and miR-21 for risk groups and CCA To predict the increasing risk for CCA, a multivariate logistic regression analysis was carried out for both crude and adjusted ORs (Table 3). A high level of urinary miR-192 was a significant predictor to differentiate healthy from O. viverrini-infected subjects (crude OR = 7.5; P = 0.003, adjusted OR = 8.4; P = 0.003), healthy from PDF subjects (crude OR = 7.5; P = 0.002, adjusted OR = 7.6; P = 0.002) and healthy from CCA patients (crude OR = 3.5; P = 0.051, adjusted OR = 5.6; P = 0.025). The ORs for urinary miR-21 revealed a crude OR = 6.3 (P = 0.004) and adjusted OR = 6.3 (P = 0.004) in differentiating healthy from PDF subjects and crude OR = 4.4 (P = 0.024) and adjusted OR = 4.2 (P = 0.043) for the differentiation of healthy subjects from CCA patients. The combined levels also showed an improvement in the differentiation between the risk groups and healthy subjects, with a crude OR of 12.2 (P = 0.001) and an adjusted OR of 19.8 (P = 0.001) for discriminating O. viverrini- infected from healthy subjects, and a crude OR of 15.6 (P b 0.001) and an adjusted OR of 17.3 (P b 0.001) for the discrimination of subjects with PDF from healthy subjects. In addition, the combined level of these two miRNAs could differentiate CCA patients from healthy subjects, with a crude OR = 11.3 (P = 0.001) and an adjusted OR = 11.6 (P = 0.002). 3.4. Reduced levels of urinary miR-192 and miR-21 after praziquantel treatment Urine was collected before and after praziquantel administration of 20 patients of the O. viverrini-infected group. The levels of urinary miR-192 and miR-21 were analyzed in paired pre- and post-treatment samples. The levels of both miRNAs were significantly reduced in posttreatment samples compared with pre-treatment samples (P b 0.001 and P = 0.004 respectively, Fig. 4). 4. Discussion
Fig. 2. Urinary miR-21 levels in healthy subjects, O. viverrini-infected subjects, subjects with PDF, CCA and patients with other inflammatory diseases (A). Data were presented as relative expression (normalized with miR-16). Mann–Whitney U-tests were used to compare the levels of urinary miRNAs between each subject group (O. viverrini infected, PDF or CCA) and the healthy control group. The ROC curves in subjects with PDF relative to healthy subjects (B) and CCA patients to healthy subjects (C). Data are expressed as the mean ± standard deviation (SD). *P b 0.05. **P b 0.01. ***P b 0.001.
Chronic inflammation induced by O. viverrini infection is the major risk for CCA development in northeastern Thailand [2]. The features of opisthorchiasis are epithelial desquamation and hyperplasia, inflammation, granuloma formation and PDF [20]. The study by Sripa and coworkers demonstrated that infection by O. viverrini leads to a chronic inflammatory disease which is mediated by pro-inflammatory cytokines, particularly IL-6 [21]. Pathological changes are significantly associated with the development advanced PDF, which, in turn, might lead to CCA [21]. Due to the poor prognosis for advanced CCA patients,
Please cite this article as: R. Silakit, et al., Urinary microRNA-192 and microRNA-21 as potential indicators for liver fluke-associated cholangiocarcinoma risk group, Parasitology International (2015), http://dx.doi.org/10.1016/j.parint.2015.10.001
R. Silakit et al. / Parasitology International xxx (2015) xxx–xxx
5
Fig. 3. Combined levels of miR-192 and miR-21 in healthy subjects, O. viverrini-infected subjects and subjects with PDF, CCA patients and those with other inflammatory diseases (A). Mann–Whitney U-tests were used to compare the levels of urinary miRNAs between each subject group (O. viverrini infected, PDF or CCA) and the healthy control group. The ROC curve analysis in opisthorchiasis relative to healthy subjects (B), subjects with PDF relative to healthy controls (C) and CCA patients relative to healthy subjects (D). Data are expressed as the mean ± standard deviation (SD). *P b 0.05. **P b 0.01. ***P b 0.001.
screening of the risk population in endemic areas might be a possible approach for detecting CCA at an early stage. To detect PDF caused by the thickening of the bile duct wall, ultrasound is a widely accessible diagnostic imaging modality that can detect an increased echo along
the portal veins and biliary ducts [6,22]. Nonetheless, a physician is required for using ultrasonography. Thus, finding potential non-invasive biomarkers for early CCA detection, O. viverrini-infection and/or PDF are still in needed. Host miRNA expression levels are known to change
Table 3 Odds ratios (OR) for the association of urinary miRNAs with risk factors for CCA. Group comparisons MiR-192 Healthy: Ov-infected Low High Healthy: PDF Low High Healthy: CCA Low High MiR-21 Healthy: Ov-infected Low High Healthy: PDF Low High Healthy: CCA Low High Combine miRNAs Healthy: Ov-infected Low High Healthy: PDF Low High Healthy: CCA Low High
Controls
Cases
Crude OR [95% CI]
P value
Adjusted OR [95% CI]
P value
15 6
6 18
7.5 [1.998–28.15]
0.003
8.4 [2.044–34.33]
0.003
15 6
7 21
7.5 [2.094–26.86]
0.002
7.6 [2.109–27.66]
0.002
14 7
8 15
3.5 [0.997–12.29]
0.051
5.6 [1.244–25.60]
0.025
10 11
10 14
1.3 [0.391–4.141]
0.689
1.4 [0.413–4.763]
0.588
15 6
8 20
6.3 [1.786–21.86]
0.004
6.3 [1.774–22.31]
0.004
15 6
8 14
4.4 [1.210–15.81]
0.024
4.2 [1.048–16.45]
0.043
16 5
5 19
12.2 [2.979–49.64]
0.001
19.8 [3.556–109.9]
0.001
17 4
6 22
15.6 [3.787–64.12]
b 0.001
17.3 [3.955–75.66]
b 0.001
15 6
4 18
11.3 [2.669–47.42]
0.001
11.6 [2.475–54.42]
0.002
Ov, O. viverrini; PDF, periductal fibrosis; CCA, cholangiocarcinoma; OR, odds ratio; CI, confidence interval; Adjusted OR, odds ratio adjusted for age and sex statistical analysis; CI, confidence interval. P value less than 0.05 is considered statistically significant.
Please cite this article as: R. Silakit, et al., Urinary microRNA-192 and microRNA-21 as potential indicators for liver fluke-associated cholangiocarcinoma risk group, Parasitology International (2015), http://dx.doi.org/10.1016/j.parint.2015.10.001
6
R. Silakit et al. / Parasitology International xxx (2015) xxx–xxx
Fig. 4. Urinary miR-192 (A) and miR-21 (B) levels in pre- and post-praziquantel treatment in O. viverrini-infected subjects. Wilcoxon's signed-rank test was performed for paired samples. *P b 0.05. **P b 0.01. ***P b 0.001.
after specific parasitic infections [11,23,24]. Furthermore, several studies showed that miRNAs have roles in the development of fibrosis in many organs such as the heart, lung, liver and kidney [25–27]. The present study is the first to show miRNA levels in urine sample as potential risk indicators of opisthorchiasis, subjects with PDF and CCA. We found that urinary miR-192 was significantly higher in O. viverriniinfected subjects than in healthy subjects. ROC analyses for discriminating healthy subjects from O. viverrini-infected people yielded an AUC of 0.766 with 75.0% sensitivity and 71.4% specificity. We also found that urinary miR-192 was significantly higher in subjects with PDF than in healthy individuals. ROC analyses revealed an AUC of 0.781 with 75.0% sensitivity and 71.4% specificity for the differentiation of healthy subjects from people with PDF. The level of urinary miR-192 was also found to be significantly higher in CCA patients than in healthy subjects with an AUC of 0.682, and 63.6% sensitivity and 66.7% specificity. Thus, urinary miR192 may also have potential to be used as a screening marker to differentiate between healthy and O. viverrini infected, PDF and/or CCA patients. We also found that the level of urinary miR-21 was significantly higher in subjects with PDF and CCA than in healthy controls. ROC analyses for distinguishing healthy subjects from subjects with PDF yielded an AUC of 0.735 with 71.4% sensitivity and 71.4% specificity, and an AUC of 0.682 with 63.6% sensitivity and 71.4% specificity for the differentiation of people with CCA from healthy subjects. However, we found that the urinary miR-21 level was significantly higher in patients with other inflammatory diseases than in healthy controls. Thus, using only urinary miR-21 as a biomarker may not be specific for liver fluke infection-related CCA. For this reason, the differentiating power of the combination of urinary miR-192 and miR-21 was assessed. Examining the ratios of relative levels of miRNAs has been reported to be a useful method to improve diagnostic potential [28]. The study from Tsujiura and coworkers investigated miRNA expression ratios by dividing the upregulated miRNA concentration by that of the downregulated miRNA, and calculating the highest AUC value [29]. Combined miRNAs revealed an elevated AUC and showed strong potential as a novel non-invasive biomarker for the early detection of colorectal cancer [30]. In the present study, the levels of miR-192 and miR-21 showed an upregulation in urine samples. We combined their levels by addition of the normalized relative level of urinary miR-192 and miR-21. Interestingly, the combined levels improved the power for differentiating O. viverrini-infected subjects from healthy subjects, yielding an AUC of 0.812 with 79.2% sensitivity and 76.2% specificity. In addition, an AUC of 0.815 with 78.6% sensitivity and 81.0% specificity was determined for the differentiation of subjects with PDF from healthy subjects, and an AUC of 0.849 with 81.8% sensitivity and 71.4% specificity for discriminating CCA patients from healthy subjects. These findings suggest that the measurement of urinary miR-192 and miR-21 levels could be used as screening markers for the O. viverrini infection and CCA risk group (PDF) in the endemic area.
We also predicted an increasing risk of CCA (Table 3). The high levels of urinary miR-192 and miR-21 is predictable for opisthorchiasis, PDF and CCA. We found higher OR values in the opisthorchiasis and PDF groups than in CCA. This result may be explained if miR-192 and miR21 are induced by O. viverrini infection [17,18] and are also involved in fibrosis development [31–33]. Notably, this study demonstrated that urinary miR-192 and miR-21 were significantly decreased in the subjects after praziquantel treatment. The findings indicated that urinary miR-192 and miR-21 levels were increased by O. viverrini infection and that they were decreased following parasite clearance. Recently, miRNA genes were identified in three opisthorchiids, O. viverrini, O. felineus and Clonorchis sinensis. The miR-192 and miR-21 expressions have not been reported in these three flukes [16], suggesting that these miRNAs might be released by the mammalian host in response to parasite infection. Thus, determination of host urinary miR-192 and/or miR-21 might be used as risk indicators for O. viverrini infection, which is a risk factor for CCA development. In conclusion, this study revealed that the levels of urinary miR-192 and miR-21 were higher in the risk group of subjects than in healthy individuals. Nevertheless, relatively small sample sizes were assessed in the present study. Therefore, larger sample sizes from different subject groups should be studied in the future to validate the differentiating power of these miRNAs. Further studies are also required to explain miR-192 and miR-21 involvement in the human host's response to O. viverrini infection and the development of fibrosis. The present results suggest that the host urinary miRNAs miR-192 and miR-21, which are increased in opisthorchiasis and PDF-related CCA development, might have better predictive values in areas endemic for O. viverrini than non-endemic regions. Acknowledgements This work was supported by the Thailand Research Fund through the Royal Golden Jubilee PhD Program (Grant No. PHD/0185/2552) to RS and NN, the Targeted Research Grant (Grant No. TR54101 to PY), the Invitation Research Grant (I56328 to RS), the Research Assistantship grant (AS58101 to NN) from the Faculty of Medicine, a grant from Khon Kaen University (KKU56-58), the Higher Education Research Promotion and National Research University Project of Thailand, Office of the Higher Education Commission, through the Health Cluster (SHeP-GMS) Khon Kaen University to NN and the grant of CASCAP. References [1] B. Sripa, S. Kaewkes, P. Sithithaworn, E. Mairiang, T. Laha, M. Smout, et al., Liver fluke induces cholangiocarcinoma, PLoS Med. 4 (7) (2007), e201. [2] P. Yongvanit, S. Pinlaor, H. Bartsch, Oxidative and nitrative DNA damage: key events in opisthorchiasis-induced carcinogenesis, Parasitol. Int. 61 (1) (2012) 130–135.
Please cite this article as: R. Silakit, et al., Urinary microRNA-192 and microRNA-21 as potential indicators for liver fluke-associated cholangiocarcinoma risk group, Parasitology International (2015), http://dx.doi.org/10.1016/j.parint.2015.10.001
R. Silakit et al. / Parasitology International xxx (2015) xxx–xxx [3] S. Prakobwong, S. Pinlaor, P. Yongvanit, P. Sithithaworn, C. Pairojkul, Y. Hiraku, Time profiles of the expression of metalloproteinases, tissue inhibitors of metalloproteases, cytokines and collagens in hamsters infected with Opisthorchis viverrini with special reference to peribiliary fibrosis and liver injury, Int. J. Parasitol. 39 (7) (2009) 825–835. [4] W. Thamavit, N. Bhamarapravati, S. Sahaphong, S. Vajrasthira, S. Angsubhakorn, Effects of dimethylnitrosamine on induction of cholangiocarcinoma in Opisthorchis viverrini-infected Syrian golden hamsters, Cancer Res. 38 (12) (1978) 4634–4639. [5] S. Pinlaor, S. Prakobwong, T. Boonmars, C. Wongkham, P. Pinlaor, Y. Hiraku, Effect of praziquantel treatment on the expression of matrix metalloproteinases in relation to tissue resorption during fibrosis in hamsters with acute and chronic Opisthorchis viverrini infection, Acta Trop. 111 (2) (2009) 181–191. [6] N. Chamadol, C. Pairojkul, N. Khuntikeo, V. Laopaiboon, W. Loilome, P. Sithithaworn, et al., Histological confirmation of periductal fibrosis from ultrasound diagnosis in cholangiocarcinoma patients, J. Hepatobiliary Pancreat. Sci. 21 (5) (2014) 316–322. [7] N. Khuntikeo, N. Chamadol, P. Yongvanit, W. Loilome, N. Namwat, P. Sithithaworn, et al., Cohort profile: cholangiocarcinoma screening and care program (CASCAP), BMC Cancer 15 (2015) 459. [8] R. Robledo, A. Muro, M.L. Prieto, Extrahepatic bile duct carcinoma: US characteristics and accuracy in demonstration of tumors, Radiology 198 (3) (1996) 869–873. [9] R.W. Carthew, Gene regulation by microRNAs, Curr. Opin. Genet. Dev. 16 (2) (2006) 203–208. [10] J. Lu, A.G. Clark, Impact of microRNA regulation on variation in human gene expression, Genome Res. 22 (7) (2012) 1243–1254. [11] D. Delic, M. Dkhil, S. Al-Quraishy, F. Wunderlich, Hepatic miRNA expression reprogrammed by Plasmodium chabaudi malaria, Parasitol. Res. 108 (5) (2011) 1111–1121. [12] G. Hu, R. Zhou, J. Liu, A.Y. Gong, X.M. Chen, MicroRNA-98 and let-7 regulate expression of suppressor of cytokine signaling 4 in biliary epithelial cells in response to Cryptosporidium parvum infection, J. Infect. Dis. 202 (1) (2010) 125–135. [13] L. Zhu, J. Dao, X. Du, H. Li, K. Lu, J. Liu, et al., Altered levels of circulating miRNAs are associated Schistosoma japonicum infection in mice, Parasitol. Vectors 8 (2015) 196. [14] R. Manzano-Roman, M. Siles-Lucas, MicroRNAs in parasitic diseases: potential for diagnosis and targeting, Mol. Biochem. Parasitol. 186 (2) (2012) 81–86. [15] X. He, X. Sai, C. Chen, Y. Zhang, X. Xu, D. Zhang, et al., Host serum miR-223 is a potential new biomarker for Schistosoma japonicum infection and the response to chemotherapy, Parasitol. Vectors 6 (2013) 272. [16] V.Y. Ovchinnikov, D.A. Afonnikov, G.V. Vasiliev, E.V. Kashina, B. Sripa, V.A. Mordvinov, et al., Identification of microRNA genes in three opisthorchiids, PLoS Negl. Trop. Dis. 9 (4) (2015), e0003680. [17] P. Chusorn, N. Namwat, W. Loilome, A. Techasen, C. Pairojkul, N. Khuntikeo, et al., Overexpression of microRNA-21 regulating PDCD4 during tumorigenesis of liver fluke-associated cholangiocarcinoma contributes to tumor growth and metastasis, Tumour Biol. 34 (3) (2013) 1579–1588. [18] R. Silakit, W. Loilome, P. Yongvanit, P. Chusorn, A. Techasen, T. Boonmars, et al., Circulating miR-192 in liver fluke-associated cholangiocarcinoma patients: a prospective prognostic indicator, J. Hepatobiliary Pancreat. Sci. 21 (12) (2014) 864–872.
7
[19] S.J. Yun, P. Jeong, W.T. Kim, T.H. Kim, Y.S. Lee, P.H. Song, et al., Cell-free microRNAs in urine as diagnostic and prognostic biomarkers of bladder cancer, Int. J. Oncol. 41 (5) (2012) 1871–1878. [20] B. Sripa, Pathobiology of opisthorchiasis: an update, Acta Trop. 88 (3) (2003) 209–220. [21] B. Sripa, E. Mairiang, B. Thinkhamrop, T. Laha, S. Kaewkes, P. Sithithaworn, et al., Advanced periductal fibrosis from infection with the carcinogenic human liver fluke Opisthorchis viverrini correlates with elevated levels of interleukin-6, Hepatology 50 (4) (2009) 1273–1281. [22] N.J. Benedetti, T.S. Desser, R.B. Jeffrey, Imaging of hepatic infections, Ultrasound Q. 24 (4) (2008) 267–278. [23] S.P. O'Hara, P.L. Splinter, G.B. Gajdos, C.E. Trussoni, M.E. Fernandez-Zapico, X.M. Chen, et al., NFkappaB p50-CCAAT/enhancer-binding protein beta (C/EBPbeta)mediated transcriptional repression of microRNA let-7i following microbial infection, J. Biol. Chem. 285 (1) (2010) 216–225. [24] X.M. Chen, P.L. Splinter, S.P. O'Hara, N.F. LaRusso, A cellular micro-RNA, let-7i, regulates toll-like receptor 4 expression and contributes to cholangiocyte immune responses against Cryptosporidium parvum infection, J. Biol. Chem. 282 (39) (2007) 28929–28938. [25] X. Jiang, E. Tsitsiou, S.E. Herrick, M.A. Lindsay, MicroRNAs and the regulation of fibrosis, FEBS J. 277 (9) (2010) 2015–2021. [26] M. Kato, J. Zhang, M. Wang, L. Lanting, H. Yuan, J.J. Rossi, et al., MicroRNA-192 in diabetic kidney glomeruli and its function in TGF-beta-induced collagen expression via inhibition of E-box repressors, Proc. Natl. Acad. Sci. U. S. A. 104 (9) (2007) 3432–3437. [27] L.L. Lv, Y.H. Cao, H.F. Ni, M. Xu, D. Liu, H. Liu, et al., MicroRNA-29c in urinary exosome/microvesicle as a biomarker of renal fibrosis, Am. J. Physiol. Ren. Physiol. 305 (8) (2013) F1220–F1227. [28] M. Avissar, B.C. Christensen, K.T. Kelsey, C.J. Marsit, MicroRNA expression ratio is predictive of head and neck squamous cell carcinoma, Clin. Cancer Res. 15 (8) (2009) 2850–2855. [29] M. Tsujiura, D. Ichikawa, S. Komatsu, A. Shiozaki, H. Takeshita, T. Kosuga, et al., Circulating microRNAs in plasma of patients with gastric cancers, Br. J. Cancer 102 (7) (2010) 1174–1179. [30] Z. Huang, D. Huang, S. Ni, Z. Peng, W. Sheng, X. Du, Plasma microRNAs are promising novel biomarkers for early detection of colorectal cancer, Int. J. Cancer 127 (1) (2010) 118–126. [31] A.C. Chung, X.R. Huang, X. Meng, H.Y. Lan, miR-192 mediates TGF-beta/Smad3driven renal fibrosis, J. Am. Soc. Nephrol. 21 (8) (2010) 1317–1325. [32] A. Zarjou, S. Yang, E. Abraham, A. Agarwal, G. Liu, Identification of a microRNA signature in renal fibrosis: role of miR-21, Am. J. Physiol. Ren. Physiol. 301 (4) (2011) F793–F801. [33] F. Glowacki, G. Savary, V. Gnemmi, D. Buob, C. Van der Hauwaert, J.M. Lo-Guidice, et al., Increased circulating miR-21 levels are associated with kidney fibrosis, PLoS One 8 (2) (2013), e58014.
Please cite this article as: R. Silakit, et al., Urinary microRNA-192 and microRNA-21 as potential indicators for liver fluke-associated cholangiocarcinoma risk group, Parasitology International (2015), http://dx.doi.org/10.1016/j.parint.2015.10.001