Utility of Adding Primovist Magnetic Resonance Imaging to Analysis of Hepatocellular Carcinoma by Liver Dynamic Computed Tomography

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2013;11:187 192 Utility of Adding Primovist Magnetic Resonance Imaging to Analysis of Hepatocellular Carcinoma by Liver Dynamic Computed Tomography YOUNG JOO JIN,* SO YUN NAH,* JIN WOO LEE,* JUNG IL LEE,* SEOK JEONG,* DON HAENG LEE,*, YOUNG SOO KIM,* SOON GU CHO, and YONG SUN JEON *Department of Internal Medicine, and Department of Radiology, Inha University Hospital, Inha University School of Medicine; and Department of Internal Medicine and Center for Advanced Medical Education by Brain Korea 21 Project, Inha University School of Medicine, and Utah-Inha DDS and Advanced Therapeutics Research Center, Incheon, Republic of Korea BACKGROUND & AIMS: More information is needed on use of Primovist-enhanced magnetic resonance imaging (MRI) during initial staging analysis of patients with hepatocellular carcinoma (HCC) who are candidates for treatment with liver dynamic computed tomography (CT). METHODS: We studied 104 patients who were initially diagnosed as Barcelona Clinic Liver Cancer (BCLC) stage 0 or A, without any other suspicious intrahepatic lesions, by liver dynamic CT from December 2009 to May 2012 at Inha University Hospital in Korea. We evaluated whether an addition of Primovist-enhanced MRI examination affected determination of BCLC stage, compared with liver dynamic CT. RESULTS: On the basis of CT analysis, the median tumor size was 2.3 cm, and 98 patients had HCCs that met the Milan criteria (94.2%). All 104 patients (100%) had nodular type HCC. Abnormal intrahepatic findings were detected by Primovist-enhanced MRI in 41 patients (39.4%). Eighteen patients (17.3%) had new HCCs, but 6 patients who were initially of BCLC stage A remained at this stage, despite increases in tumor numbers. Of the 104 patients, 31 (29.8%) and 73 (70.2%) had BCLC stage 0 and A HCC before Primovist-enhanced MRI, respectively, and 26 (25.0%), 71 (68.3%), and 7 (6.7%) patients had BCLC stages 0, A, and B HCC after Primovist-enhanced MRI, respectively. In 12 of the 104 patients (11.5%), BCLC stage changed from 0 to A (5/31, 16.1%) or from A to B (7/73, 9.6%). CONCLUSIONS: Primovistenhanced MRI can provide additional information that can lead to the detection of new intrahepatic HCC lesions during initial staging analyses of patients with BCLC stage 0 or A HCC by liver dynamic CT, despite the absence of other suspicious liver lesions. Keywords: Diagnosis; Patient Management; Early Detection; Liver Cancer. Hepatocellular carcinoma (HCC) is a leading cause of cancer-related death worldwide, and its incidence continues to rise, even in Europe and the United States. 1 3 Recent advancements in early diagnosis and treatment modalities have significantly improved patient survival, 4,5 but only a small proportion of patients are candidates for curative treatments. 6 Treatment decision-making in HCC is guided by tumor stage, which can be affected by tumor status at presentation. 6,7 Thus, intrahepatic tumor status can be pivotal for treatment decision-making and for determining prognosis, especially in candidates for curative treatment. The American Association for the Study of Liver Diseases guidelines recommend that HCC patients with typical HCC 1 cm by liver dynamic computed tomography (CT) do not require further investigation to confirm the presence of HCC, and that appropriate treatment can be initiated on the basis of tumor stage. 8 However, other intrahepatic HCC lesions not detected by liver dynamic CT can be detected only by Primovist-enhanced magnetic resonance imaging (MRI). 9 11 Several studies have evaluated the usefulness of Primovist-enhanced MRI in HCC patients, 12 14 but most of these studies have been limited by lack of an evaluation of a change in HCC stage caused by Primovistenhanced MRI and by the inclusion of patients with suspicious intrahepatic HCC lesions already detected by liver dynamic CT. Furthermore, because of the poor prognosis of HCC patients with early recurrence after curative treatments such as liver transplantation, surgical resection, or radiofrequency ablation (RFA), 15,16 accurate staging work-up is essential for intrahepatic HCC, especially in patients with Barcelona Clinic Liver Cancer (BCLC) stage 0 (very early) or A (early) HCC because they are potential candidates for curative treatments. However, no evidence-based recommendations have been issued regarding the benefit of add-on Primovist-enhanced MRI during initial staging work-ups in HCC patients deemed candidates for curative treatment after diagnosis by liver dynamic CT. Therefore, in the present study, we assessed the usefulness of Primovist-enhanced MRI during the initial staging work-ups of HCC patients with BCLC stage 0 or A, but without any other suspicious intrahepatic HCC lesion by liver dynamic CT. Furthermore, we investigated whether Primovist-enhanced MRI can provide additional information on BCLC stage by directly comparing findings of Primovist-enhanced MRI with those of liver dynamic CT, and we sought to establish whether add-on Primovist-enhanced MRI after a diagnosis of HCC by liver dynamic CT could aid treatment decision-making in these patients. Methods Study Subjects Between December 2009 and March 2012, 283 consecutive patients were initially diagnosed as having HCC by liver dynamic CT at Inha University Hospital. No patient had history Abbreviations used in this paper: BCLC, Barcelona Clinic Liver Cancer; CT, computed tomography; CTAP, CT during arterial portography; CTHA, CT during hepatic arteriography; HCC, hepatocellular carcinoma; MRI, magnetic resonance imaging; RFA, radiofrequency ablation. 2013 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2012.11.002

188 JIN ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 2 of HCC treatment. HCC was diagnosed according to the diagnostic guidelines issued by the American Association for the Study of Liver Diseases. 8 A retrospective database was obtained from these 283 consecutive patients. Of these 283 patients, 101 were excluded because of intermediate, advanced, or terminal stage disease at diagnosis by liver dynamic CT, and 60 were excluded because Primovist-enhanced MRI was performed to confirm a suspicious lesion detected by liver dynamic CT. In addition, 18 patients were excluded because of refusal to undergo an MRI examination (N 12) or another concurrent malignancy (N 6). Therefore, our study cohort consisted of 104 patients who underwent Primovist-enhanced MRI after receiving a diagnosis of BCLC stage 0 or A HCC without another suspicious lesion by liver dynamic CT. The detection protocol used for intrahepatic HCC at our institution was based on the previous comparative reports on the advantages of Primovist-enhanced MRI as compared with liver dynamic CT. 12 14 Acquisition of Radiologic Images Patients were evaluated by Primovist-enhanced MRI within 2 days of the acquisition of liver dynamic CT images. Liver dynamic CT images were obtained craniocaudally from the level of the inferior pulmonary vein to the iliac bone, and it was performed as previously described. 17 Primovist-enhanced MRI was performed by using the liverspecific hepatocyte-directed MRI contrast agent, gadoxetic acid (SH L 569 B, Gd-EOB-DTPA, Primovist; Bayer Schering Pharma AG, Berlin, Germany), as previously described. 9,18 Briefly, gadoxetic acid solution (0.025 mmol/kg body weight dose of 0.25 mol/l) was injected at a speed of 2 ml/s through an intravenous line placed in an antecubital vein, followed by a flush with 30 ml 0.9% normal saline. Before intravenous injection of contrast agent, T2- and T1-weighted MR images were obtained by using the following conditions: a T2-weighted fast spinecho/turbo spin-echo sequence ( 3000/90120 [repetition time in milliseconds/echo time in milliseconds], 192 256 256 matrix, 5- to 8-mm slice thickness, 0- to 2-mm gap), and a T1-weighted gradient recalled echo sequence with chemically selective fat suppression and without fat suppression by using a 160 192 256 matrix, 5- to 8-mm slice thickness, and a 0- to 2-mm gap. Immediately after injecting intravenous contrast agent, dynamic imaging was performed in the arterial, portal, venous, and equilibrium phases by using the T1-weighted gradient recalled echo sequence without fat suppression. Twenty minutes after injecting the contrast agent, T1-weighted sequence with fat suppression was repeated in the axial and coronal planes. Primovist-enhanced MR images were reviewed by 2 radiologists. Both had at least 5 years of professional experience at interpreting MR images, and both were blinded to the design of the study. To find extrahepatic metastasis, chest CT or 18 F-fluorodeoxyglucose positron emission tomography was performed. However, only patients without extrahepatic metastasis were enrolled in the present study. Detection and Confirmation of Intrahepatic Tumors Definite radiological features of intrahepatic HCC on Primovist-enhanced MR images were defined as arterial enhancement with delayed lesion washout in dynamic phase and hypointense in hepatobiliary phase 19 and/or hypointensity on T1-weighted images with hyperintense on T2-weighted images. 20 The sizes of new intrahepatic lesions, detected by Primovist-enhanced MRI, were recorded as the longest diameter of the largest lesion in at least one dimension. As a diagnostic gold standard of new HCC lesions that were detected by Primovist-enhanced MRI, histologic confirmation by liver biopsy or postoperative specimen was performed. If histologic confirmation was not possible, the lesion concerned was alternatively diagnosed as HCC on the basis of its typical perfusion defects and on enhancement patterns observed by CT during arterial portography (CTAP) and by CT during hepatic arteriography (CTHA) during pretreatment evaluations according to the Japanese guidelines for a diagnosis of HCC. 21,22 Lesions with definite findings of benignity were followed up by liver dynamic CT or Primovist-enhanced MRI at 1- to 3-month intervals for the first 6 months and subsequently at 3- to 6-month intervals. Their median follow-up period was 19 months (range, 5 31 months). Clinical Impact of Primovist-Enhanced Magnetic Resonance Imaging on Hepatocellular Carcinoma Stage BCLC stages were determined before and after Primovist-enhanced MRI. Detection rates of additional lesions by Primovist-enhanced MRI, which had not been previously detected by liver dynamic CT, were analyzed according to BCLC stage. To determine the impact of Primovist-enhanced MRI on treatment decision-making, we analyzed changes in BCLC stage that affected treatment decision-making. The study protocol was approved by the Institutional Review Board of Inha University Hospital, Incheon, South Korea. Statistical Analysis The basic clinical characteristics of patients are expressed as medians (ranges) and frequencies. Differences between categorical or continuous variables were analyzed by using the 2 test, Fisher exact test, or Student t test. Two-tailed P values of less than.05 were considered statistically significant in all analyses. Diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and the diagnostic accuracy of Primovist-enhanced MRI were calculated. Statistical analyses were performed by using SPSS v18.0 (SPSS Inc, Chicago, IL). Results Baseline Characteristics of Patients The baseline characteristics of the 104 patients are summarized in Table 1. Median patient age was 55 years (range, 39 81 years), and 90 (84.1%) were male. The most common etiology of HCC was hepatitis B virus infection, which was observed in 76 patients (73.1%). Most patients had good hepatic reserve function and performance status. Eighty-four patients (80.8%) were of Child Turcotte Pugh class A, and 104 (100%) had an Eastern Cooperative Oncology Group performance status of 0. Median alpha-fetoprotein concentration in the cohort was 19.6 ng/ml (range, 1.6 3.5 10 4 ng/ml) (Table 1). At the time of CT-based diagnosis, 92 of the 104 patients (88.5%) had a single HCC, and 10 (9.6%) and 2 (1.9%) patients had 2 and 3 HCCs, respectively. Median tumor size was 2.3 cm

February 2013 MRI DURING HCC STAGING 189 Table 1. Baseline Clinical Characteristics of 104 Patients With HCC Variable Total (N 104) Age (y) a 55 (39 81) Sex (male), n (%) 90 (84.1) Etiology, n (%) HBV/HCV/alcohol/others 76/10/8/10 (73.1/9.6/7.7/9.6) CTP classification, n (%) A/B 84/20 (80.8/19.2) ECOG (PST), n (%) 0/1 104/0 (100/0) Alpha fetoprotein (ng/ml) a 19.6 (1.6 3.5 10 4 ) HCC lesions b Tumor number, n (%) One/two/three 92/10/2 (88.5/9.6/1.9) Tumor size (cm) a 2.3 (1 16) Tumor type, n (%) Nodular/infiltrative 104/0 (100/0) Within Milan criteria, n (%) 98 (94.2) CTP, Child Turcotte Pugh classification; ECOG, Eastern Cooperative Oncology Group; HBV, hepatitis B virus; HCV, hepatitis C virus; PST, performance status. a Median (range). b Lesions on liver dynamic CT. (range, 1 16 cm), and 98 (94.2%) had HCCs within the Milan criteria. All patients (100%) had nodular type HCC (Table 1). However, no patient had an additional possible HCC lesion except definite HCC lesions, and 28 patients (26.9%) had a benign lesion such as shunt (n 4, 3.8%), cyst (n 21, 20.2%), or hemangioma (n 3, 2.9%). Newly Detected Intrahepatic Lesions by Primovist-Enhanced Magnetic Resonance Imaging Initial Primovist-enhanced MRI showed new intrahepatic lesions not detected by liver dynamic CT in 39 of the 104 patients (37.5%). Truly new benign intrahepatic lesions were observed in 21 patients (20.2%), and the most common benign lesion was a dysplastic nodule (n 12, 11.5%). These dysplastic nodules did not show arterial enhancement on dynamic phase, and they showed isointensity on hepatobiliary phase of Primovist-enhanced MRI. A histologic specimen could be obtained from 5 of the 12 patients with a dysplastic nodule, and all exhibited low-grade dysplasia. Other benign lesions included regenerating nodule, shunt, cyst, or hemangioma, and all of them were small ( 1.5 cm) in size (Table 2). Newly detected definite HCC lesions were observed in 14 patients (13.5%), of whom 10 (71.4%) had a single HCC, 3 had 2 HCCs, and 1 had 3 HCCs. The longest diameters of these newly detected HCC lesions fell in the range of 10 20 mm, although very small-sized ( 1 cm) HCC lesions were also observed in 2 of the 14 patients (14.3%) (Table 2). In 4 of these 14 patients, new lesions were confirmed by postoperative histology, and 3 patients were found to have well-differentiated HCC and 1 to have moderately differentiated HCC. Two of these 14 patients underwent RFA with concurrent liver biopsy, and both had well-differentiated HCC. Follow-up CTAP and CTHA were performed in 8 of the 14 patients, and all showed typical arterial enhancement and delayed perfusion defect. Possible new intrahepatic HCC lesions could not be excluded in 6 patients (5.8%), including 2 with 2 tumors (1.7 and 1.3 cm in longest diameter, respectively) and 4 with a single tumor (1.0 1.6 cm in longest diameter). All tumors showed subtle arterial enhancement and delayed washout on dynamic phase with slight hypointensity on hepatobiliary phase of Primovistenhanced MRI. Percutaneous needle biopsy in 2 of 6 patients showed that 1 patient had true HCC (well-differentiated type), and that the other had a benign lesion. In the remaining 4 patients, typical perfusion defect by CTAP and arterial enhancement by CTHA were observed in 3 (Figure 1), but these were not observed in the remaining 1 patient. These findings suggest that 4 of the 6 patients with possible HCC lesions (66.7%) finally turned out to have a true intrahepatic HCC lesion (Table 2). Moreover, 1 patient who showed nondiagnostic imaging findings for HCC by CTAP and CTHA was followed up by CT or MRI for 24 months, and no change in size or shape was evident. Therefore, of the 20 patients with definite (n 14) or possible (n 6) HCC by initial Primovist-enhanced MRI, 18 had a true new intrahepatic HCC lesion. On the other hand, 86 patients showed definite new benign lesions (n 21) or no new lesion (n 55) by initial Primovist-enhanced MRI, and all showed no definite lesion or possible HCC lesion by Primovistenhanced MRI. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of Primovistenhanced MRI were 1.0 (18/18), 0.977 (84/86), 0.9 (18/20), 1.0 (84/84), and 0.981 (102/104), respectively. Clinical Impact of Primovist-Enhanced Magnetic Resonance Imaging on Hepatocellular Carcinoma Stage Of the 18 patients (17.3%, 18/104) with a newly detected intrahepatic HCC evident only by Primovist-enhanced Table 2. Newly Detected HCC Lesions Only on Primovist- Enhanced MRI Findings on Primovist-enhanced MRI Number (n 104) % Newly detected intrahepatic 41 39.4 lesions a Definite HCC lesions 14 13.5 Size (mm) 10/10 20/ 20 2/12/0 14.3/85.7 Number One/two/three 10/3/1 71.4/21.5/7.1 Possible HCC lesions 6 5.8 Definite benign lesions 19 18.3 Final findings, total b 41 39.4 Truly new HCC lesions 18 17.3 Size (mm) 10/10 20/ 20 2/16/0 11.1/88.9 Number One/two/three 12/5/1 66.7/27.8/5.5 Truly new benign lesions 21 20.2 DN/RN 12/3 11.5/2.9 Shunt/cyst/hemangioma 2/3/1 1.9/2.9/1.0 DN, dysplastic nodule; RN, regenerative nodule. a New lesions detected only on Primovist-enhanced MRI. b Final findings on Primovist-enhanced MRI after liver dynamic CT.

190 JIN ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 2 Figure 1. Atypical nodule on Primovist-enhanced MR image. Primovist-enhanced MRI depicted an atypical nodule with subtle arterial enhancement (A, long arrow) and delayed washout (B, long arrow) on dynamic phase with slight hypointensity on hepatobiliary phase (C, long arrow). CTHA and CTAP showed arterial enhancement (D, long arrow) and perfusion defect (E, long arrow), respectively. Angiography showed hypervascular tumor staining on atypical nodule (long arrow, F) and on the primary definite HCC (short arrow, F). MRI (Table 2), none exhibited a significant increase in tumor size on Primovist-enhanced MRI. Of the 92 patients with a single HCC by initial liver dynamic CT, 15 (16.3%) were finally diagnosed to have multifocal HCC by Primovist-enhanced MRI. Of the 12 patients with 2 or 3 HCCs by initial liver dynamic CT, 3 patients (25%) were diagnosed to have an additional new HCC lesion by Primovist-enhanced MRI. However, of these 18 patients, 6 patients who were initially classified as BCLC stage A by liver dynamic CT did not exhibit change in BCLC stage even after add-on Primovist MRI because total tumor numbers remained 3 or 3, and the longest tumor dimension was less than 3 cm despite the increase in tumor number. Before Primovist-enhanced MRI, 31 (29.8%) and 73 (70.2%) patients were classified as having BCLC stage 0 or A HCC, respectively. After evaluating Primovist-enhanced MRI, 26 (25.0%), 71 (68.3%), and 7 (6.7%) patients were classified as having BCLC stage 0, A, or B HCC, respectively. Therefore, 12 of the 104 patients (11.5%) showed a change in BCLC stage from 0 to A (5/31, 16.1%) or from A to B (7/73, 9.6%) (Figure 2). Discussion We have shown here that add-on Primovist-enhanced MRI performed during initial HCC staging provides useful additional information regarding the detection of new intrahepatic HCC lesions in patients with very early or early BCLC stage by liver dynamic CT. True new intrahepatic HCC lesions were detected by Primovist-enhanced MRI, but not by liver

February 2013 MRI DURING HCC STAGING 191 Figure 2. Clinical impact of Primovist-enhanced MRI on BCLC stage. Twelve patients (11.5%) showed a change in BCLC stage, from 0 to A (5/31, 16.1%) and from A to B (7/73, 9.6%), after Primovist-enhanced MRI. dynamic CT, in 17.3% of patients. Furthermore, in 11.5% of the patients, HCC stage as determined by using the BCLC staging system was revised after Primovist-enhanced MRI. This is a large-scale study to investigate the usefulness of add-on Primovist-enhanced MRI as a crucial component of initial intrahepatic staging work-up in HCC patients with BCLC stage 0 or A by liver dynamic CT. The accurate determinations of tumor number, size, and location are essential for deciding on the most appropriate treatment modality. 8 Recently, Primovist-enhanced MRI has been reported to improve significantly the identification and characterization of focal hepatic lesion vs liver dynamic CT. 12 14 However, there is a dearth of clinical evidence on the merits of Primovist-enhanced MRI, vs those of liver dynamic CT, in the context of HCC staging, which is the basis for treatment decision-making. Therefore, we focused on BCLC stage changes caused by add-on Primovist-enhanced MRI after a diagnosis of HCC had been made by liver dynamic CT. In previous studies, Primovist-enhanced MRI was found to aid the detection of intrahepatic HCC lesions with high sensitivity and specificity. 9 11,23 Interestingly, our findings showed that in 12 of our 104 patients (11.5%), HCC stage was changed by Primovist-enhanced MRI. Although this percentage of change in HCC stage seems not to be substantial, its clinical implication for HCC patients with BCLC stage 0 or A should not be underestimated because the new HCC lesions detected by Primovist-enhanced MRI but not detected by liver dynamic CT would not be included for target lesions in the initial treatment if add-on Primovist-enhanced MRI was not performed. Furthermore, 7 patients (9.6%) with BCLC stage A HCC were restaged at BCLC B; thus, they were not finally candidates for surgery, RFA, or transplantation. Because of the critical shortage of donor livers, operative risk, and the poor prognosis associated with unnecessary surgery or transplantation in HCC patients, it appears that add-on Primovist-enhanced MRI provides important information for the selection of the most effective treatment modality in patients with BCLC stage 0 or A HCC. In addition, incomplete surgery, noncurative RFA, and unnecessary transplantation rates could be reduced by add-on Primovist-enhanced MRI in these patients. Therefore, our findings suggest that add-on Primovist-enhanced MRI should be performed for the initial HCC staging in patients with very early or early HCC as determined by liver dynamic CT, despite the absence of any suspicious HCC lesion by liver dynamic CT images. However, patients with intermediate or advanced HCC, as determined by BCLC staging, may not be restaged when new HCC lesions are identified by Primovist-enhanced MRI, because unlike vascular invasion or extrahepatic metastasis, additional detected intrahepatic HCC lesions do not affect BCLC stage in these patients. 8 Furthermore, curative treatments cannot be considered in these patients after the identification of another HCC lesion by Primovist-enhanced MRI. On the other hand, treatment strategies and prognoses could be changed in HCC patients with very early or early HCC if new HCC lesions are detected by add-on Primovist-enhanced MRI in the clinical situation that definite HCCs are already diagnosed by liver dynamic CT. Thus, in the present study, we evaluated HCC patients with very early or early HCC who were candidates for curative treatment. In the present study, 13 patients who were initially BCLC stage A by liver dynamic CT showed new intrahepatic HCCs after evaluation of Primovist-enhanced MRI. However, BCLC stage did not change for 6 of these 13 patients after add-on Primovist MRI despite an increase in tumor number. Nevertheless, these new lesions were considered as therapeutic targets during treatment decision-making because if left untreated, especially in patients with BCLC stage A, they could have resulted in early recurrence after surgery or RFA, leading to more serious outcomes. Therefore, the present study shows that the identification of new lesions by additional Primovist-enhanced MRI has clinical implications for patients with BCLC stage A even when newly detected nodules do not change BCLC stage. In terms of benign lesions, small-sized dysplastic nodules were also detected by Primovist-enhanced MRI and subsequently confirmed by CTAP and CTHA 21,24 or histologically. Because of the roles played by these lesions in HCC development, they might have resulted in de novo HCC lesion development after curative therapy, which could also be diagnosed early by intensive monitoring by using Primovist-enhanced MRI. However, well-differentiated HCCs do not necessarily exhibit low signal intensity on T1-weighted MR images during the liver-specific phase. 25 Although the relation between HCC differentiation and MRI features could not be evaluated in the present study because not all patients underwent surgery or liver biopsy, it should be noted that Primovist-enhanced MR images require careful interpretation to distinguish well-differentiated HCCs from benign hepatic nodules in cirrhotic patients. This study has several limitations that warrant consideration. First, it is intrinsically limited by its retrospective nature. However, to avoid potential selection bias, we enrolled all patients diagnosed as having BCLC stage 0 or A HCC by liver dynamic CT who underwent add-on Primovist-enhanced MRI

192 JIN ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 2 during staging work-up. This was possible because MRI is covered by health insurance in HCC patients in our country. Second, positron emission tomography CT was not routinely performed to find extrahepatic metastasis, which is rare in very early or early stage HCC; instead, chest CT was used to detect extrahepatic metastasis in these patients. Third, we were unable to explore the histologic features of all HCCs because not all patients underwent surgery, a liver biopsy was not possible because of tumor inaccessibility, or patients requested benign lesions be followed rather than biopsied. In conclusion, this study shows that Primovist-enhanced MRI provides additional information regarding the detection of new intrahepatic HCC lesions missed by liver dynamic CT during initial staging work-ups in patients with BCLC stage 0 or A HCC without any other suspicious HCC lesion by liver dynamic CT. Because of the usefulness of Primovist-enhanced MRI for detecting intrahepatic HCC lesions of even a very small size ( 1 cm) with high sensitivity and specificity, we believe that routine Primovist-enhanced MRI may be necessary for the staging work-up of HCC patients with very early or early stage HCC, especially when curative treatment is being considered. References 1. Parkin DM. Global cancer statistics in the year 2000. Lancet Oncol 2001;2:533 543. 2. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999;340:745 750. 3. Bosch FX, Ribes J, Díaz M, et al. Primary liver cancer: worldwide incidence and trends. Gastroenterology 2004;127:S5 S16. 4. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996;334:693 699. 5. Llovet JM, Fuster J, Bruix J. Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology 1999;30:1434 1440. 6. Bruix J, Llovet JM. Prognostic prediction and treatment strategy in hepatocellular carcinoma. Hepatology 2002;35:519 524. 7. Forner A, Reig ME, de Lope CR, et al. Current strategy for staging and treatment: the BCLC update and future prospects. Semin Liver Dis 2010;30:61 74. 8. Bruix J, Sherman M, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma: an update. Hepatology 2011;53:1020 1022. 9. Kim SH, Lee J, Kim MJ, et al. Gadoxetic acid-enhanced MRI versus triple-phase MDCT for the preoperative detection of hepatocellular carcinoma. AJR Am J Roentgenol 2009;192:1675 1681. 10. Hammerstingl R, Huppertz A, Breuer J, et al. Diagnostic efficacy of gadoxetic acid (Primovist)-enhanced MRI and spiral CT for a therapeutic strategy: comparison with intraoperative and histopathologic findings in focal liver lesions. Eur Radiol 2008;18: 457 467. 11. Haradome H, Grazioli L, Tinti R, et al. Additional value of gadoxetic acid-dtpa-enhanced hepatobiliary phase MR imaging in the diagnosis of early-stage hepatocellular carcinoma: comparison with dynamic triple-phase multidetector CT imaging. J Magn Reson Imaging 2011;34:69 78. 12. Ahn SS, Kim MJ, Lim JS, et al. Added value of gadoxetic acidenhanced hepatobiliary phase MR imaging in the diagnosis of hepatocellular carcinoma. Radiology 2010;255:459 466. 13. Golfieri R, Renzulli M, Lucidi V, et al. Contribution of the hepatobiliary phase of Gd-EOB-DTPA-enhanced MRI to dynamic MRI in the detection of hypovascular small ( / 2 cm) HCC in cirrhosis. Eur Radiol 2011;21:1233 1242. 14. Sano K, Ichikawa T, Motosugi U, et al. Imaging study of early hepatocellular carcinoma: usefulness of gadoxetic acid-enhanced MR imaging. Radiology 2011;261:834 844. 15. Roayaie S, Schwartz JD, Sung MW, et al. Recurrence of hepatocellular carcinoma after liver transplant: patterns and prognosis. Liver Transpl 2004;10:534 540. 16. Minagawa M, Makuuchi M, Takayama T, et al. Selection criteria for repeat hepatectomy in patients with recurrent hepatocellular carcinoma. Ann Surg 2003;238:703 710. 17. Kim SE, Lee HC, Shim JH, et al. Noninvasive diagnostic criteria for hepatocellular carcinoma in hepatic masses 2 cm in a hepatitis B virus-endemic area. Liver Int 2011;31:1468 1476. 18. Bellin MF, Webb JA, Van Der Molen AJ, et al. Safety of MR liver specific contrast media. Eur Radiol 2005;15:1607 1614. 19. Huppertz A, Haraida S, Kraus A, et al. Enhancement of focal liver lesions at gadoxetic acid-enhanced MR imaging: correlation with histopathologic findings and spiral CT initial observations. Radiology 2005;234:468 478. 20. Taouli B, Losada M, Holland A, et al. Magnetic resonance imaging of hepatocellular carcinoma. Gastroenterology 2004;127: S144 S152. 21. Peterson MS, Baron RL, Dodd GD 3rd, et al. Hepatic parenchymal perfusion defects detected with CTAP: imaging-pathologic correlation. Radiology 1992;185:149 155. 22. Kudo M, Izumi N, Kokudo N, et al. Management of hepatocellular carcinoma in Japan: consensus-based clinical practice guidelines proposed by the Japan Society of Hepatology (JSH) 2010 updated version. Dig Dis 2011;29:339 364. 23. Mita K, Kim SR, Kudo M, et al. Diagnostic sensitivity of imaging modalities for hepatocellular carcinoma smaller than 2 cm. World J Gastroenterol 2010;16:4187 4192. 24. Lim JH, Kim EY, Lee WJ, et al. Regenerative nodules in liver cirrhosis: findings at CT during arterial portography and CT hepatic arteriography with histopathologic correlation. Radiology 1999;210:451 458. 25. Lee MH, Kim SH, Park MJ, et al. Gadoxetic acid-enhanced hepatobiliary phase MRI and high-b-value diffusion-weighted imaging to distinguish well-differentiated hepatocellular carcinomas from benign nodules in patients with chronic liver disease. AJR Am J Roentgenol 2011;197:W868 W875. Reprint requests Address requests for reprints to: Jin-Woo Lee, MD, PhD, Department of Internal Medicine, Inha University Hospital and School of Medicine, 7206, 3-ga, Shinheung-dong, Jung-gu, Incheon, 400-711, Republic of Korea. e-mail: jin@inha.ac.kr; fax: 82-32-8902549. Acknowledgements Drs Jin and Nah contributed equally to this work and share first authorship. Conflicts of interest The authors disclose no conflicts. Funding Supported by an Inha University Research Grant.