Magnetic Resonance Imaging 35 (2015) 1146–1155

Contents lists available at ScienceDirect

Magnetic Resonance Imaging journal homepage: www.mrijournal.com

MRI and 1H MRS findings of hepatobilary changes and cholangiocarcinoma development in hamsters infected with Opisthorchis viverrini and treated with N-nitrosodimethylamine Petcharakorn Hanpanich a, b,⁎, Somchai Pinlaor b, c, Lakhanawan Charoensuk b, c, Puangrat Yongvanit b, d, Yaovalux Chamgramol b, e, Chawalit Pairojkul b, e, Eimorn Mairiang a a b c d e

Department of Radiology, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand Liver Fluke and Cholangiocarcinoma Research Center, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand Department of Parasitology, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand Department of Biochemistry, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand Department of Pathology, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand

a r t i c l e

i n f o

Article history: Received 29 August 2014 Revised 23 April 2015 Accepted 21 June 2015 Keywords: Magnetic resonance imaging Magnetic resonance spectroscopy Opisthorchis viverrini Hepatobiliary change Cholangiocarcinoma

a b s t r a c t 3 T MRI and 1H MRS were useful for quantitative investigation of the serial development of hepatobiliary changes in Opisthorchis viverrini infection in hamsters, and the differential diagnosis of cholangiocacinoma (CCA) development from bile duct changes and normal condition is unclear. In this study, we investigated the serial development of hepatobiliary changes and CCAgenesis in O. viverrini-infected and N-nitrosodimethylamine (NDMA) treated hamsters (ON group) using 3 T MRI and 1H MRS and the results were compared with those either in the O. viverrini-infected group (OV group) and uninfected normal controls. In the ON group, CCAs were first found at 9 weeks post-infection, with sizes of ~2 mm. The typical MR signal characteristics of CCA were hypo- and occasionally isointensity signal on T1-weighted images, and mild–moderate to hyper-intensity signal on T2-weighted images compared to the liver parenchyma. T2-weighted images with fat suppression revealed dilatation of the intra- and extrahepatic bile ducts, and often defined the anatomical level of biliary obstruction, cystic lesions, liver abscesses, and CCA which was starting seen of these noticeable abnormalities at 5 weeks onwards. The results of fibrosis grading using MR images showed a positive correlation (r = 0.90, P b 0.038 by Spearman's rank correlation test) with those of the histopathological grading. In addition, 3.0 T 1H MRS showed elevated choline and decreased lipids levels in the liver tissues of the ON group. In conclusion, MRI and 1H MRS are useful for the quantitative investigation of the serial development of hepatobilary changes and CCA in hamsters, and are potentially useful as early diagnostic tools for CCA. © 2015 Elsevier Inc. All rights reserved.

1. Introduction Opisthorchiasis caused by infection with the liver fluke, Opisthorchis viverrini, is endemic in Southeast Asia and is a major health problem in northeastern Thailand [1]. O. viverrini is one of only three metazoan pathogens classified as Group 1 carcinogens by the World Health Organization's International Agency for Research on Cancer (the other two being Clonorchis sinensis and Schistosoma haematobium) [2]. Early histological findings in the liver of opisthorchiasis are infiltration of inflammatory cells, bile duct epithelial hyperplasia, hepatobiliary changes and DNA damage [3,4]. In cases of chronic infection, periductal fibrosis progresses over time, contributing to the risk of cholangiocarcinoma (CCA) development [5]. ⁎ Corresponding author at: Department of Radiology, Faculty of Medicine, Khon Kaen University, Khon Kaen 40002, Thailand. Tel./fax: +66 4334 8389. E-mail addresses: [email protected], [email protected] (P. Hanpanich). http://dx.doi.org/10.1016/j.mri.2015.06.022 0730-725X/© 2015 Elsevier Inc. All rights reserved.

Nitrosamines—potent human carcinogens—are ubiquitous in the human environment and have been detected in food items including cured meat, bacon, seafood, beer etc. [6–8]. Among them, N-nitrosodimethylamine (NDMA) is a strong DNA alkylation adduct that induces CCA and hepatocellular carcinoma (HCC) [9]. People in Thailand traditionally eat fermented food containing high concentration of nitrosamines including NDMA. A combination of O. viverrini infection and nitrosamine intake is assumed to contribute greatly to CCA and HCC development in endemic areas [5,10]. Malignant liver tumors are classified as HCC, CCA and mesenchymal tumors (sarcomas). The prognosis of CCA is poor, with an overall 5-year survival rate of 1% [11,12], mainly because early detection of CCA is difficult and usually too late for therapeutic benefits [13]. Liver fibrosis is a common response to chronic liver injury [14–16]. Early diagnosis of periductal fibrosis could facilitate early intervention and treatment prior to progression toward CCA [17,18].

P. Hanpanich et al. / Magnetic Resonance Imaging 35 (2015) 1146–1155

Ultrasonography (US) and magnetic resonance imaging (MRI) are the major noninvasive techniques for the diagnosis of liver diseases [19,20]. US is much more accessible and less expensive compared with other imaging methods. The resolution of US, however, is highly dependent on the skill of operators. Also, US is unable to precisely quantify the hepatic fat content, nor can it detect small changes in the liver fat, so that its usage in longitudinal clinical studies is limited [21,22]. MRI technique has been used for early cancer detection, staging, and therapy response monitoring [20,23,24] and usually applied with proton magnetic resonance spectroscopy ( 1H MRS) to investigate metabolic and biochemical information of normal and abnormal tissues non-invasively. In addition to the application of MRI for the investigation of hepatobiliary changes, MRS has been used for the evaluation for diffuse hepatic diseases such as fatty liver, chronic hepatitis and cirrhosis [25,26]. In hamsters infected with O. viverrini, combination of NDMA treatment can induce hepatobiliary changes often resulting in CCA and HCC [27,28]. We previously reported that MRI and 1H MRS were useful in monitoring serial changes of periductal fibrosis in the liver of O. viverrini-infected hamsters [29], the differential diagnosis of CCA appearance from those previously reported is unclear. The purpose of this study is to describe a time profile of quantitative changes of the hepatobiliary structure in relation to CCA development using 3 T MRI and 1H MRS in the liver of O. viverrini-infected and NDMA-treated hamsters. Emphasis was placed on the variable low or high-signal-intensity changes in MRI imaging. Moreover, the proton MRS was used to investigate the choline/lipid changes in relation to CCA development. Additionally, the MR imaging appearance was analyzed and compared with pathologic and histological findings.

2. Materials and methods 2.1. Parasites Cyprinid fish obtained from northeastern Thailand, were digested in 0.25% pepsin–HCl solution. Encysted O. viverrini metacercariae were isolated, identified, collected, and counted, and viable cysts were used for the infection of hamsters [30].

2.2. Animals and experimental design This study was approved by the Animal Ethics Committee of Khon Kaen University (AEKKU 52/2553). Six-week-old male golden hamsters (n = 70) weighing 120–150 g were obtained from the Animal Unit, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand. They were used in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Research Council of Thailand. They were housed under conventional conditions, fed a stock diet and given water ad libitum. Animals were divided into two groups: uninfected normal controls (n = 10) and those treated by O. viverrini infection combined with N-nitrosodimethylamine (NDMA) (Wako Pure Chemical Industries Ltd., Osaka, Japan) administration (ON group, n = 60). Fifty metacercariae of O. viverrini were fed orally to each hamster, together with 12.5 ppm of NDMA in water ad libitum for 2 months and withdrawn thereafter. Ten control animals and 60 ON-treated animals were scanned regularly by MRI. For histopathological evaluation of hepatobiliary changes, ON-treated hamsters having lesions seen in MR images were selected and sacrificed at 5, 9, 13, 17 and 21 weeks p.i.. In addition, normal control animals (5 of 10) were sacrificed at week 21.

1147

2.3. MRI and MRS studies In every alternate weeks from 1 to 21 weeks, 60 O. viverriniinfected and NMDA-treated animals and 5 of 10 normal animals were scanned with MRI and 1H MRS. Before imaging, hamsters were anesthetized by an intraperitoneal injection of Nembutal® (pentobarbital sodium, 70–90 mg/kg). The hamsters were examined with a 3 T whole-body MR system (Achieva 3.0 T TX; Philips, Amsterdam, Netherlands) with a maximum gradient capability of 80 mT/m and a maximum slew rate of 200 mT/m/s. An eight-channel human wrist coil was used as an animal holder. A hamster was placed in the supine position inside the coil with the liver region located in the center of the coil; the abdomen was fixed with adhesive tape wrapped around the cradle to reduce respiratory movement artifacts. The animal was covered with a piece of cloth to keep body temperature constant. Scout images were first acquired in three orthogonal planes with a fast, low-angle shot sequence for locating the liver. Conventional liver MRI protocols were as follows: (1) TSE (turbo spin echo) T2-weighted with fat suppression SPAIR (spectral selection attenuated inversion recovery) in the axial, sagittal and coronal orientations (repetition time [TR] 2000 ms, echo time [TE] 80 ms, inversion delay 120 ms, field of view 100–120 mm, slice thickness 1.5–2.0 mm, axial matrix 284 × 210, sagittal matrix 240 × 160, coronal matrix 268 × 240); (2) axial orientation SE (spin echo) T1-weighted images (TR 250 ms, TE 3.53 ms, field of view 100– 120 mm, slice thickness 1.5–2.0 mm, matrices the same as for T2). For the 1H MRS study, the axial, sagittal and coronal T2-weighted images were used to locate the sensitive voxel. First-order shim correction (automatic shimming) in which localized shimming was performed. A single-voxel 1H MRS using a water-suppressed point-resolved spectroscopy sequence (PRESS) was performed with TR = 3000 ms, TE = 30 ms, number of averages = 128, voxel size 7 × 7 × 7 or 8 × 8 × 8 mm 3, and 1024 data points. The acquisition time was 4.52 min. 1H MRS was undertaken in an area within the middle portion of the right or left hepatic lobe, avoiding the large intrahepatic vessels, or the intrahepatic lesions such as cysts, abscesses, and tumor. 2.4. Evaluation of MR images The high signal image intensities of the bile ducts in the MR images were used for the evaluation of inflammation, bile duct dilatation and fibrosis. Bile duct changes were graded 0 to 3, as follows: grade 0 for no high signal image intensity of bile ducts; grade 1 for high signal in one segment of the liver; grade 2 for high signal in two or three segments of the liver; and grade 3 for high signal in more than three segments of the liver [29]. MR images were read by two researchers, a scientist and a radiologist, who have more than 10 years of experience on the MRI diagnosis for human CCA with sufficient knowledge. Each observer made a blind reading without any clinical information on the animals and the time of the study. To evaluate the association between MR findings and clinical outcomes, the joint reading of both observers was performed. 2.5. Data analysis for MRS MRS data were analyzed using a commercial software (Spectroscopy; Phillips) by a single experienced MR technologist for 1H MRS analysis. The 1H MRS data with serious baseline distortion were excluded from data analysis when it was very difficult to recognize the lipid resonance and the choline-containing compounds resonance. The relative choline-to-lipid (Cho/lipid) ratios were obtained by dividing the peak area of the choline-containing compounds at

1148

P. Hanpanich et al. / Magnetic Resonance Imaging 35 (2015) 1146–1155

2.8. Statistical analysis Spearman's rank correlation test was employed for correlation of the histological results of periductal fibrosis and MRI grading, using a statistical software (SPSS for Windows, version 12; SPSS Inc., Chicago, IL, USA). A two-tailed P-value of less than 0.05 was required for statistical significance. 3. Results 3.1. Clinical signs of animal observation

Fig. 1. Graphic present data of the total animals in the ON group which were scanned, sacrificed, and died between 1–21 weeks p.i.

3.2 ppm by the peak area of lipid at 0.9–1.4 ppm. The mean and standard deviations of the ratio values were calculated at the designated time points.

2.6. Histological study

Unlike previously studied in O. viverrini-infected hamsters alone [29], 20 death animals in the ON group after MRI investigation during 9–19 weeks p.i. were observed and included in the data analyses. The number of hamsters death was 2 animals at 10–13 weeks (total animals for MRI scan in the ON group = 46), 4 animals at 14–17 weeks (total animals for MRI scan in the ON group = 36), and 14 animals at 18–21 weeks (total animals for MRI scan in the ON group = 16). They died with clinical signs such as depression, abdominal enlargement and were autopsied just before the designed time point. Total Hamsters in the ON groups starting for MRI and MRS scan were 60 animals and decreased total scan for histopathological evaluation of hepatobiliary changes selected and

To compare the MR imaging results with the histopathological appearances, the radiologists and pathologists have worked in concert. The serial development of bile duct lesions in the livers of hamsters with O. viverrini infection combined with NDMA administration were determined by histological grading after MRI scanning, the infected and NDMA-treated animals were euthanized; 6 animals in the ON group were selected at 5, 9, 13, 17 and 21 weeks post-infection (p.i.). Control animals (n = 5) were sacrificed at week 21. The livers were removed and fixed in 4% buffered formalin, dehydrated in an ascending series of ethanol, and embedded in paraffin. Sections of 5 μm were cut and stained with hematoxylin– eosin for histological examination. In addition, collagen fibers were stained using a Picrosirius Red Stain Kit (Polysciences Inc., Warrington, PA, USA). Histological features were graded and staged using the guidelines set out by Goodman [31]. Peribiliary fibrosis, fibrosis of the bile duct and hepatic septa, was scored in five stages: 0 for no fibrosis, 1 + for mild fibrosis, 2 + for moderate fibrosis with fibrous portal expansion, 3 + for severe fibrosis with few septa, and 4 + for more severe fibrosis with numerous septa [32].

2.7. Immunohistochemical study To examine the localization of biliary cytokeratin (CK)-19, a marker of bile duct proliferation, 4 μm sections were deparaffinized in xyline and rehydrated in a descending alcohol series. The samples were then autoclaved at 110 ° C for 10 min in citrate buffer (pH 6.0) for antigen retrieval. The sections were blocked for endogenous peroxidase activity with 3% H2O2 for 30 min at room temperature and then were incubated with rabbit anti-CK19 (1:100; Abcam, Cambridge, MA, USA) overnight at room temperature. Next, the sections were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG antibody (1:400; Amersham Biosciences, Amersham, UK) at room temperature for 1 h. DAB solution (3′,3′-diaminobenzidine) (0.02%) in 0.05 M Tris–HCl (pH 7.6) and 0.01% H2O2 (v/v) was used as a chromogenic substrate. Sections were counterstained with Mayer's hematoxylin. The stained sections were examined using a microscope (Carl Zeiss, Jena, Germany).

Fig. 2. Representative MRI coronal plane images of the hamster livers. T1-weighted image: (A) normal and (B) ON group. T2-weighted image: (C) normal and (D) ON group. The presence of tumors is shown by arrow. k = kidney, GB = gallbladder.

P. Hanpanich et al. / Magnetic Resonance Imaging 35 (2015) 1146–1155

1149

Fig. 3. Representative axial plane MRI T2-weighted images. Series time profile of inflammation and bile duct hyperplasia was monitored from 1 week to 21 weeks in the ON group compared with normal control. The degree of periductal fibrosis and bile duct dilatation (thin arrow) increased over time and multi-cystic lesions (bold arrows) were seen during 19 and 21 weeks.

Fig. 4. Representative coronal plane MRI T2-weighted images. Series time profile of inflammation and bile duct fibrosis and dilatation (thin arrow) was monitored at 1 to 21 weeks of the ON group compared with normal control. The images show more severe periductal fibrosis, intrahepatic bile duct fibrosis and dilatation, extra-hepatic duct which increased over time. The liver changed and liver masses were found from 13 weeks onwards (bold arrow).

1150

P. Hanpanich et al. / Magnetic Resonance Imaging 35 (2015) 1146–1155

Fig. 5. Representative gross study and T2-weighted images (coronal and axial plane). (A–C) One of the hamsters in the ON group at 19 weeks p.i. showing tumor (arrow) size ~2 × 4 mm which has higher signal intensity than the liver. (D–F) One of the hamsters in the ON group at 13 weeks p.i. showing one abscess (arrow) with heterogeneous signal intensity, representing debris dependent on layering material. (G–I) One of the hamsters in the ON group at 5 weeks p.i. showing two small cystic (arrow) lesions which have higher signal intensity than the tumor and abscess.

sacrificed in each time point at 5, 9, 13, 17 and 21 weeks p.i. (n = 30). The data of the animals in the ON group data are summarized in Fig. 1. 3.2. MRI study In the normal control group, the T1-weighted image (Fig. 2A) showed that the fat surrounding the liver had higher (brighter) signal intensity than the liver parenchyma. The T2-weighted image (Fig. 2C) revealed no dilatation of the bile ducts (dark) and a normal homogenous liver appearance. In the ON group, the tumors were mostly recognized as low signal intensity in the T1-weighted image (Fig. 2B) and intermediate/high signal intensity in the T2-weighted image (Fig. 2D). The degree of intra- and extrahepatic duct dilatation increased with time, appearing brighter in the regions of the bile ducts in the livers on T2-weighted axial plane images (Fig. 3) and coronal plane images (Fig. 4). MRI showed inflammation and dilatation of the intrahepatic bile ducts, which were variably seen in different liver lobes of each animal at the same time point. At weeks 1–3 p.i., MRI showed beaded dilatation of a few intrahepatic ducts. At week 5, MRI showed progression of intrahepatic duct dilatation throughout the liver. The changes in the bile ducts were graded 1, 2, and 3 at 1, 3 and 5 weeks p.i, respectively, which were essentially the same as those seen in our previous study on O. viverrini infection alone (OV group) [29].

Notably, from 13 weeks onwards, dilated extrahepatic ducts were prominently seen in the ON group. Comparison between the ON group in this study and the OV group in the previous study [29] revealed that periductal fibrosis and dilatation in intra and extra-hepatic duct were much more severe in the ON group than in the OV group from 1 to 21 weeks p.i. CCA was developed in the ON group between 9 and 21 weeks p.i. Tumors of ~ 2 mm in size (Fig. 5A–C) were first found at 9 weeks p.i. Abscesses (Fig. 5D–F) were also first found at 9 weeks p.i., with sizes ~ 4 mm. Cystic lesions (Fig. 5G–I) were first found at 5 weeks p.i.; the size of the cysts was ~ 2 mm. All those lesions were seen only in the ON group but not in the OV group. In order to identify the development of hepatobiliary changes and CCAgenesis, we measured the high signal intensity of cystic lesions on T2-weighted images and found the value to be between 1500 and 3500. In MR images, cystic lesions were seen as high signal intensity on T2-weighted images. Because of the long relaxation time of the cyst fluid, signal intensity of the cysts was much brighter than that of the liver parenchyma on T2-weighted images (Fig. 5H–I). On T2-weighted images, abscesses were seen as mixed signal intensity values between low and high and generally seen as the lower signal intensity images than cystic lesions, with heterogeneous signal intensity dependent on the layering debris materials. Abscesses were difficult to differentiate from metastatic foci of adenocarcinoma which also gave signal intensity values between low and high (Fig.5E–F).

P. Hanpanich et al. / Magnetic Resonance Imaging 35 (2015) 1146–1155

1151

Fig. 6. Representative T2-weighted coronal plane images. (A) One of the hamsters in the ON group at 9 weeks p.i. showing tumor (arrow) size ~2 mm which has higher signal intensity than the liver. Gross pathology (B) shows a liver mass (arrow) in the liver. Histological study (C) showed a sub-capsular early CCA nodule. (D) Immunohistochemical study presented bile duct proliferation.

A solitary liver mass of intrahepatic CCA ranging from 2 to 6 mm in diameter, with irregular, lobulated or smooth margins was seen in 9–21 weeks p.i. Typically, the tumor was seen as low signal intensity on T1-weighted images and intermediate or high signal intensity on T2-weighted images relative to the liver parenchyma. Occasionally, tumors were isointensity to the liver on T1- and T2-weighted images. Tumor appearance on MRI images was supported by histological and immunohistochemical study (Fig. 6). 3.3. MRS study A typical 1H MR spectrum of choline and lipids of the liver of one of the subjects is shown in Fig. 7. The signals of the corresponding lipid groups were seen at 1.4 ppm for (–CH2) and 0.9 ppm for (–CH3). Another peak was the signals of the choline-containing compounds at 3.2 ppm. In the normal control group, the calculated mean (± SD) choline-to-lipid ratio was 0.12 ± 0.013. The cholineto-lipid ratio in the ON group increased gradually with time from 1 to 21 weeks p.i. (mean ± SD at 1 week, 0.20 ± 0.03; at 21 weeks, 0.58 ± 0.42) in association with the degree of intra- and extrahepatic bile duct dilatation. These changes were really the same as that seen in the OV group in our previous study [29] (Figs. 7 and 8). 3.4. Histopathological study The bile ducts and liver parenchyma of the normal control group had normal architecture, with very narrow portal spaces and no sign

of inflammation, as described previously [32]. Infection with O. viverrini caused inflammatory changes of the intrahepatic bile ducts. In the ON group, flukes were found in dilated fibrotic (thickened) large intrahepatic bile ducts (Fig. 9A). There was bile duct proliferation, with diffuse periductal fibrosis and chronic inflammatory cell infiltration; occasionally bile ducts containing acute inflammatory exudate were noted. Both early and late liver abscesses were found at the subcapsular area; early abscess contained suppurative exudates, and later abscesses showed xanthomatous (Fig. 9B–D). Complex glandular arrangement of proliferative bile ducts with gland formation is suggestive of CCA in situ (Fig. 9E) and early CCA (Fig. 9F). The mucosa in large bile ducts showed adenomatous hyperplasia and periductal fibrosis as described previously [33,34]. The degrees of fibrosis by histological grading were 2.2 ± 1.29, 2.6 ± 0.63, 2.8 ± 0.45, 3.2 ± 0.84 and 3.8 ± 0.45 (mean ± SD) at weeks 5, 9, 13, 17 and 21, respectively, which were higher grading score than in O. viverrini infection as previously study [29]. 3.5. Correlation of MRI and histopathological study In order to confirm MRI scanning, 6 animals in the ON group were selected and sacrificed in each time point at 5, 9, 13, 17 and 21 weeks p.i. Correlation between the hepatobiliary changes based on the MRI findings such as fibrosis, cysts, abscesses or tumor and histopathological findings was analyzed by radiologists and pathologists who worked in concert. The results revealed that there was a positive

1152

P. Hanpanich et al. / Magnetic Resonance Imaging 35 (2015) 1146–1155

Fig. 7. Representative T2-weighted axial, sagittal, and coronal images for localization and 1H MRS spectrum in ON. The 1H MRS of the liver from weeks 5 to 21 showed lipid compound peaks at 0.9–1.4 ppm and the Cho (choline-containing compounds) peak at 3.2 ppm. Comparison between normal controls and the NO group showed that the amount of lipid decreased, but the Cho/lipid ratio increased significantly.

correlation between hepatobiliary changes evaluated by MR images and histopathological grading (r = 0.90, P b 0.038). Abscesses were seen as mixed signal intensity as the lower signal intensity images than cystic lesions on T2-weighted images. CCA was seen as mild/ moderate hyper signal intensity on T2-weighted images relative to the liver parenchyma and confirmed histopathologically in 23.3% (7 out of 30 animals) as was summarized in Fig. 10.

Fig. 8. Comparison of mean of choline/lipid ratios in normal controls and the ON group, and the OV group in our previous study [29] in weeks 1 to 21.

4. Discussion CCA development has been extensively studied in O. viverriniinfected and NDMA-treated hamsters [35,36]. In O. viverrini infection alone [29], intra- and extra-hepatic bile duct fibrosis and dilatation increased over time but there was no detectable CCA. The degree of high signal intensity on T2-weighted images was variable and has been described as mild, moderate or marked for human CCA [37,38]. In this study, MR images demonstrated that hepatobiliary changes were much more severe in the ON group than those of the OV group, which were in agreement with those described by Prakobwong et al. [5]. In addition, the anatomical level of biliary obstruction, cystic lesions, liver abscesses, and CCA were seen in MR imaging of the liver of the ON group. At 9 weeks p.i., early, small-sized tumor lesions of ~ 2 mm in diameter with irregular, lobulated or smooth margins were observed by MRI. Those features are similar to those seen in human CCA [37,39], as evaluated by T2-weighted images. MRI can effectively detect peritoneal metastases as small as 1 cm in diameter [40]. In this study, CCA was first seen ~ 2 mm and found histopathologically in 23.3% in the ON group. Although some hamsters in the ON group did not show histopathological evidence of CCA, all of them showed cholangiofibrosis, goblet cell metaplasia and dysplasia as reported previously [27,29]. In the present study, MRI grading of the hepatobiliary changes in the ON groups correlated positively with the histopathological evaluation of inflammation and fibrosis, implying that MR imaging is a powerful tool for detecting abnormal hepatobiliary changes until CCA happening.

P. Hanpanich et al. / Magnetic Resonance Imaging 35 (2015) 1146–1155

1153

Fig. 9. Representative image of histopathological changes of the bile ducts of combination of O. viverrini infection and NDMA (ON) from weeks 5, 9, 13, 17, and 21. (A) Week 5: A fluke (O. viverrini; arrow) in a dilated fibrotic bile duct. Diffuse periductal fibrosis and chronic inflammatory cell infiltration. (B) Week 9: early abscess, containing suppurative exudates. (C) Week 13: late abscess, containing macrophages phagocytose necrotic debri (xanthomatous change). (D) Week 17: subcapsular abscesses. (E) Week 21: complex glandular arrangement of the proliferative bile duct, with gland formation, suggesting CCA in situ. (F) Week 21: early CCA; duct lining with atypical cells with enlarged nuclei and complex glandular arrangement.

MR images could efficiently discriminate between cystic lesions and CCA. Cystic lesions showed homogeneous hypo-intensity signal on T1-weighted images and homogeneous hyper intensity signal on T2-weighted images, whereas the typical MR signal characteristics of CCA were hypo- and occasionally isointensity signal on T1-weighted images, and mild/moderate hyper-intensity signal on T2-weighted images compared with the liver parenchyma [41,42]. This variation of MR images of CCA is mostly due to the degree of fibrosis and necrosis, and mucin content in the tumor [43]. Nowadays, advances in MRI have brought about a growing interest in optimizing and applying functional MRI methods for the assessment of liver diseases including CCA. One of such examples is diffusion-weighted imaging (DWI) for liver fibrosis and cirrhosis [44]. In humans, MRCP and contrast enhanced MR imaging can provide all information necessary for the pretherapeutic staging of hilar CCA, because of their intrinsic high soft tissue contrast and multiplanar capability [45,46]. However, in the present study, intravenous injections are practically impossible for hamsters because of the lack of easily accessible veins [47]. MRCP and the administration of contrast media require long study time with

Fig. 10. Graphic present data of MR image demonstrated hepatobilary changes such as fibrosis, cystic lesions, liver abscesses, and CCA from 6 animals in the ON group selected in each time point at 5, 9, 13, 17 and 21 weeks post-infection and sacrificed for confirmed gross and histological study.

increased dose of anesthesia, which will be harmful and high risk for the animal death and MR image result. Moreover, animal movement during the scan often produces an image problem. In these cases, those methods are not suitable for animal models. The principal metabolite that has been targeted in focal liver disease is choline. In general, choline is elevated in tumors, because choline is a cell membrane component, which increases in association with the increased cell turnover in malignancy [48]. In the present study, the results of 1H MRS showed that the choline/ lipid ratio increased significantly in the ON group, although the degree of increase was compatible to that of the OV group [29]. Related to this, Laothong et al. [49] reported that O. viverrini infection induces inflammation-mediated oxidative stress and liver injury, which may alter α-tocopherol and lipid metabolism. Apart from the constituent of the cell membrane, choline is a nutrient essential for normal function of all cells, especially as the precursor for acetylcholine. The liver parenchyma physiologically contains large choline metabolite pools by catabolic and anabolic reactions leading to elevated choline-containing compounds (CCC) levels [48]. In the ON group, the increase of the choline/lipid ratio in the ON group started as early as 1 week p.i. An increase in choline levels as detected by 1H MRS was reported in cases of HCC, and a reduction in the choline/lipid ratio was observed after trans-arterial embolization for HCC [50,51]. 1H MRS is a valuable tool for detecting hepatobiliary changes at an early phase in hamster liver fibrosis prior to the development of CCA, and is potentially useful in determining treatment strategies and evaluating therapeutic outcomes. Although ultrasonography is still the gold standard for screening of hepatobiliary abnormalities, detection of small tumors in the biliary system at the early stage is limited [52,53]. A high-resolution imaging technique like MRI is necessary for an early-stage diagnosis of CCA or suspected cases. MRI evaluation of hepatobiliary changes in hamsters reported here may provide valuable information about early-stage findings of CCA genesis caused by liver fluke infection and oral carcinogen. 5. Conclusion

1

The present study demonstrated the feasibility of using MRI and H MRS for the evaluation of the anatomical and metabolic

1154

P. Hanpanich et al. / Magnetic Resonance Imaging 35 (2015) 1146–1155

hepatobiliary changes of O. viverrini-infection-associated CCA. The choline/lipid ratios correlated well with the increase in cellular proliferation and cell density in bile ducts of hamsters infected with O. viverrini and treated with NDMA. The MRI protocol described here is recommended for detecting fibrosis and CCA in small animals. MRI can be used in combination with US in detecting periductal fibrosis at an early stage of CCA and to monitor its progression in human subjects who are infected with O. viverrini. They are also valuable for the evaluation of hepatobiliary changes for possible tumor resection, follow-up of tumor progression, and early CCA monitoring in humans. Acknowledgments This work was supported by the National Research Universities Project and Research Promotion in Higher Education, Office of the Higher Education Commission, Thailand, through the Center of Excellence in Specific Health Problems in the Greater Mekong Subregion cluster (SHeP-GMS), Khon Kaen University. The Department of Radiology, Faculty of Medicine, and the Liver Fluke and Cholangiocarcinoma Research Center, Khon Kaen University, are gratefully acknowledged for instrument support. We also thank Dr. Christopher C. Salisbury and Yukifumi Nawa, Publication Clinic, Research Affairs, Faculty of Medicine, Khon Kaen University, for their suggestions and critical reading of the manuscript. References [1] Rangsin R, Mungthin M, Taamasri P, Mongklon S, Aimpun P, Naaglor T, et al. Incidence and risk factors of Opisthorchis viverrini infections in a rural community in Thailand. Am J Trop Med Hyg 2009;81(1):152–5 [Epub 2009/ 06/27]. [2] Bouvard V, Baan R, Straif K, Grosse Y, Secretan B, El Ghissassi F, et al. A review of human carcinogens—part B: biological agents. Lancet Oncol 2009;10(4):321–2 [Epub 2009/04/08]. [3] Kawanishi S, Hiraku Y. Oxidative and nitrative DNA damage as biomarker for carcinogenesis with special reference to inflammation. Antioxid Redox Signal 2006;8(5–6):1047–58 [Epub 2006/06/15]. [4] Pinlaor S, Sripa B, Sithithaworn P, Yongvanit P. Hepatobiliary changes, antibody response, and alteration of liver enzymes in hamsters re-infected with Opisthorchis viverrini. Exp Parasitol 2004;108(1–2):32–9 [Epub 2004/10/20]. [5] Prakobwong S, Yongvanit P, Hiraku Y, Pairojkul C, Sithithaworn P, Pinlaor P, et al. Involvement of MMP-9 in peribiliary fibrosis and cholangiocarcinogenesis via Rac1-dependent DNA damage in a hamster model. Int J Cancer 2010;127(11): 2576–87 [Epub 2010/02/18]. [6] Lee SJ, Shin JH, Sung NJ, Kim JG, Hotchkiss JH. Effect of cooking on the formation of N-nitrosodimethylamine in Korean dried seafood products. Food Addit Contam 2003;20(1):31–6 [Epub 2003/01/10]. [7] Sen NP, Donaldson B, Charbonneau C, Miles WF. Effect of additives on the formation of nitrosamines in meat curing mixtures containing spices and nitrite. J Agric Food Chem 1974;22(6):1125–30 [Epub 1974/11/01]. [8] Hammes WP. Metabolism of nitrate in fermented meats: the characteristic feature of a specific group of fermented foods. Food Microbiol 2011;29(2):151–6 [Epub 2011/12/29]. [9] Rabes HM, Kerler R, Wilhelm R. DNA alkylation and tumor induction in regenerating rat liver after cell cycle-related continuous N-nitrosodimethylamine infusion. J Natl Cancer Inst 1983;70(1):193–8 [Epub 1983/01/01]. [10] Sripa B, Kaewkes S, Sithithaworn P, Mairiang E, Laha T, Smout M, et al. Liver fluke induces cholangiocarcinoma. PLoS Med 2007;4(7):e201 [Epub 2007/07/12]. [11] Manfredi R, Masselli G, Maresca G, Brizi MG, Vecchioli A, Marano P. MR imaging and MRCP of hilar cholangiocarcinoma. Abdom Imaging 2003;28(3):319–25 [Epub 2003/04/30]. [12] Baer HU, Stain SC, Dennison AR, Eggers B, Blumgart LH. Improvements in survival by aggressive resections of hilar cholangiocarcinoma. Ann Surg 1993;217(1): 20–7 [Epub 1993/01/01]. [13] Lankisch TO, Metzger J, Negm AA, Vosskuhl K, Schiffer E, Siwy J, et al. Bile proteomic profiles differentiate cholangiocarcinoma from primary sclerosing cholangitis and choledocholithiasis. Hepatology 2011;53(3):875–84 [Epub 2011/03/05]. [14] Brenner DA. Molecular pathogenesis of liver fibrosis. Trans Am Clin Climatol Assoc 2009;120:361–8 [Epub 2009/09/22]. [15] Friedman SL. Seminars in medicine of the Beth Israel Hospital, Boston. The cellular basis of hepatic fibrosis. Mechanisms and treatment strategies. N Engl J Med 1993;328(25):1828–35 [Epub 1993/06/24]. [16] Hernandez-Gea V, Friedman SL. Pathogenesis of liver fibrosis. Annu Rev Pathol 2010;6:425–56 [Epub 2010/11/16].

[17] Arthur MJ. Reversibility of liver fibrosis and cirrhosis following treatment for hepatitis C. Gastroenterology 2002;122(5):1525–8 [Epub 2002/05/02]. [18] Chow AM, Tan M, Gao DS, Fan SJ, Cheung JS, Qiao Z, et al. Molecular MRI of liver fibrosis by a peptide-targeted contrast agent in an experimental mouse model. Invest Radiol 2012;48(1):46–54 [Epub 2012/11/30]. [19] Faria SC, Ganesan K, Mwangi I, Shiehmorteza M, Viamonte B, Mazhar S, et al. MR imaging of liver fibrosis: current state of the art. Radiographics 2009;29(6): 1615–35 [Epub 2009/12/05]. [20] Aube C, Moal F, Oberti F, Roux J, Croquet V, Gallois Y, et al. Diagnosis and measurement of liver fibrosis by MRI in bile duct ligated rats. Dig Dis Sci 2007; 52(10):2601–9 [Epub 2007/04/13]. [21] Saverymuttu SH, Joseph AE, Maxwell JD. Ultrasound scanning in the detection of hepatic fibrosis and steatosis. Br Med J (Clin Res Ed) 1986;292(6512):13–5 [Epub 1986/01/04]. [22] Roldan-Valadez E, Favila R, Martinez-Lopez M, Uribe M, Mendez-Sanchez N. Imaging techniques for assessing hepatic fat content in nonalcoholic fatty liver disease. Ann Hepatol 2008;7(3):212–20 [Epub 2008/08/30]. [23] Fass L. Imaging and cancer: a review. Mol Oncol 2008;2(2):115–52 [Epub 2009/04/23]. [24] Hyodo T, Kumano S, Kushihata F, Okada M, Hirata M, Tsuda T, et al. CT and MR cholangiography: advantages and pitfalls in perioperative evaluation of biliary tree. Br J Radiol 2012;85(1015):887–96 [Epub 2012/03/17]. [25] Longo R, Ricci C, Masutti F, Vidimari R, Croce LS, Bercich L, et al. Fatty infiltration of the liver. Quantification by 1H localized magnetic resonance spectroscopy and comparison with computed tomography. Invest Radiol 1993;28(4):297–302 [Epub 1993/04/01]. [26] Cho SG, Kim MY, Kim HJ, Kim YS, Choi W, Shin SH, et al. Chronic hepatitis: in vivo proton MR spectroscopic evaluation of the liver and correlation with histopathologic findings. Radiology 2001;221(3):740–6 [Epub 2001/11/24]. [27] Thamavit W, Pairojkul C, Tiwawech D, Itoh M, Shirai T, Ito N. Promotion of cholangiocarcinogenesis in the hamster liver by bile duct ligation after dimethylnitrosamine initiation. Carcinogenesis 1993;14(11):2415–7 [Epub 1993/11/01]. [28] Thamavit W, Pairojkul C, Tiwawech D, Shirai T, Ito N. Strong promoting effect of Opisthorchis viverrini infection on dimethylnitrosamine-initiated hamster liver. Cancer Lett 1994;78(1–3):121–5 [Epub 1994/04/01]. [29] Hanpanich P, Pinlaor S, Charoensuk L, Yongvanit P, Thomas C, Kothan S, et al. MRI and (1)H MRS evaluation for the serial bile duct changes in hamsters after infection with Opisthorchis viverrini. Magn Reson Imaging 2013;31(8):1418–25 [Epub 2013/07/31]. [30] Pinlaor S, Ma N, Hiraku Y, Yongvanit P, Semba R, Oikawa S, et al. Repeated infection with Opisthorchis viverrini induces accumulation of 8-nitroguanine and 8-oxo-7,8-dihydro-2′-deoxyguanine in the bile duct of hamsters via inducible nitric oxide synthase. Carcinogenesis 2004;25(8):1535–42 [Epub 2004/04/03]. [31] Goodman ZD. Grading and staging systems for inflammation and fibrosis in chronic liver diseases. J Hepatol 2007;47(4):598–607 [Epub 2007/08/19]. [32] Prakobwong S, Khoontawad J, Yongvanit P, Pairojkul C, Hiraku Y, Sithithaworn P, et al. Curcumin decreases cholangiocarcinogenesis in hamsters by suppressing inflammation-mediated molecular events related to multistep carcinogenesis. Int J Cancer 2010;129(1):88–100 [Epub 2010/09/09]. [33] Riganti M, Pungpak S, Sachakul V, Bunnag D, Harinasuta T. Opisthorchis viverrini eggs and adult flukes as nidus and composition of gallstones. Southeast Asian J Trop Med Public Health 1988;19(4):633–6 [Epub 1988/12/01]. [34] Harinasuta T, Riganti M, Bunnag D. Opisthorchis viverrini infection: pathogenesis and clinical features. Arzneimittelforschung 1984;34(9B):1167–9 [Epub 1984/ 01/01]. [35] Loilome W, Yongvanit P, Wongkham C, Tepsiri N, Sripa B, Sithithaworn P, et al. Altered gene expression in Opisthorchis viverrini-associated cholangiocarcinoma in hamster model. Mol Carcinog 2006;45(5):279–87 [Epub 2006/03/22]. [36] Boonmars T, Wu Z, Boonjaruspinyo S, Puapairoj A, Kaewsamut B, Nagano I, et al. Involvement of c-Ski oncoprotein in carcinogenesis of cholangiocacinoma induced by Opisthorchis viverrini and N-nitrosodimethylamine. Pathol Oncol Res 2011;17(2):219–27 [Epub 2010/09/21]. [37] Zhang Y, Uchida M, Abe T, Nishimura H, Hayabuchi N, Nakashima Y. Intrahepatic peripheral cholangiocarcinoma: comparison of dynamic CT and dynamic MRI. J Comput Assist Tomogr 1999;23(5):670–7 [Epub 1999/10/19]. [38] Low RN. MR imaging of the liver using gadolinium chelates. Magn Reson Imaging Clin N Am 2001;9(4):717–43 [vi. Epub 2001/11/06]. [39] Worawattanakul S, Semelka RC, Noone TC, Calvo BF, Kelekis NL, Woosley JT. Cholangiocarcinoma: spectrum of appearances on MR images using current techniques. Magn Reson Imaging 1998;16(9):993–1003 [Epub 1998/12/05]. [40] Low RN, Barone RM, Lacey C, Sigeti JS, Alzate GD, Sebrechts CP. Peritoneal tumor: MR imaging with dilute oral barium and intravenous gadolinium-containing contrast agents compared with unenhanced MR imaging and CT. Radiology 1997;204(2):513–20 [Epub 1997/08/01]. [41] Soyer P, Bluemke DA, Reichle R, Calhoun PS, Bliss DF, Scherrer A, et al. Imaging of intrahepatic cholangiocarcinoma: 2. Hilar cholangiocarcinoma. AJR Am J Roentgenol 1995;165(6):1433–6 [Epub 1995/12/01]. [42] Jarnagin WR. Cholangiocarcinoma of the extrahepatic bile ducts. Semin Surg Oncol 2000;19(2):156–76 [Epub 2000/12/29]. [43] Vilgrain V, Van Beers BE, Flejou JF, Belghiti J, Delos M, Gautier AL, et al. Intrahepatic cholangiocarcinoma: MRI and pathologic correlation in 14 patients. J Comput Assist Tomogr 1997;21(1):59–65 [Epub 1997/01/01]. [44] Aube C, Racineux PX, Lebigot J, Oberti F, Croquet V, Argaud C, et al. Diagnostic et quantification de la fibrose hepatique par IRM de diffusion: resultats preliminaires (Diagnosis and quantification of hepatic fibrosis with diffusion

P. Hanpanich et al. / Magnetic Resonance Imaging 35 (2015) 1146–1155

[45]

[46] [47]

[48]

[49]

weighted MR imaging: preliminary results). J Radiol 2004;85(3):301–6 [Epub 2004/06/12]. Yin LL, Song B, Xu J, Li YC. Hilar cholangiocarcinoma: preoperative evaluation with a three dimensional volumetric interpolated breath-hold examination magnetic resonance imaging sequence. Chin Med J (Engl) 2007;120(8):636–42 [Epub 2007/05/23]. Masselli G, Gualdi G. Hilar cholangiocarcinoma: MRI/MRCP in staging and treatment planning. Abdom Imaging 2008;33(4):444–51 [Epub 2007/07/20]. Turner PV, Pekow C, Vasbinder MA, Brabb T. Administration of substances to laboratory animals: equipment considerations, vehicle selection, and solute preparation. J Am Assoc Lab Anim Sci 2011;50(5):614–27 [Epub 2012/02/15]. Xu L, Liu B, Huang Y, Liu X, Zhang SW, Xin XG, et al. 3.0 T proton magnetic resonance spectroscopy of the liver: quantification of choline. World J Gastroenterol 2013;19(9):1472–7 [Epub 2013/03/30]. Laothong U, Pinlaor P, Boonsiri P, Hiraku Y, Khoontawad J, Hongsrichan N, et al. alpha-Tocopherol and lipid profiles in plasma and the expression of alpha-

[50]

[51]

[52]

[53]

1155

tocopherol-related molecules in the liver of Opisthorchis viverrini-infected hamsters. Parasitol Int 2013;62(2):127–33 [Epub 2012/11/28]. Xu H, Li X, Yang ZH, Xie JX. In vivo 1H MR spectroscopy in the evaluation of the serial development of hepatocarcinogenesis in an experimental rat model. Acad Radiol 2006;13(12):1532–7 [Epub 2006/12/02]. Lin YJ, Koretsky AP. Manganese ion enhances T1-weighted MRI during brain activation: an approach to direct imaging of brain function. Magn Reson Med 1997;38(3):378–88 [Epub 1997/10/27]. Mairiang E, Laha T, Bethony JM, Thinkhamrop B, Kaewkes S, Sithithaworn P, et al. Ultrasonography assessment of hepatobiliary abnormalities in 3359 subjects with Opisthorchis viverrini infection in endemic areas of Thailand. Parasitol Int 2011;61(1):208–11 [Epub 2011/07/21]. Mairiang E, Chaiyakum J, Chamadol N, Laopaiboon V, Srinakarin J, Kunpitaya J, et al. Ultrasound screening for Opisthorchis viverrini-associated cholangiocarcinomas: experience in an endemic area. Asian Pac J Cancer Prev 2006;7(3):431–3 [Epub 2006/10/25].