Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
The recurrence of hepatocellular carcinoma (HCC) after liver transplantation (LT) for HCC is a well-known problem associated with high mortality and has resulted in the development of restrictive selection criteria, such as the Milan criteria.1–3 Meanwhile, de novo HCC in a graft after LT has been very limited. In previous cases, the diagnosis of de novo HCC was straightforward because the cases did not have HCC at the time of LT.4–10 Discrimination between recurrent (metastatic) and de novo HCC is a difficult task in a nontransplant setting and in a transplant setting with advanced HCCs at the time of LT. Indeed, confirmation of de novo HCC requires genetic analysis of the tumor cells.
In terms of hepatitis C, it has been recently reported that HCC could develop even after achievement of sustained viral response (SVR) after interferon therapy for hepatitis C.11–14 However, there has been no report of de novo HCC occurring in a liver graft with SVR after LT.
In this article, we present the first case of de novo HCC occurring in a liver graft with SVR 6 years after living donor liver transplantation (LDLT) for advanced HCCs and hepatitis C cirrhosis. We used microsatellite analysis for genotyping and confirmed that the HCC originated from the transplanted liver with fibrosis.
The recipient was a 58-year-old Japanese female who was referred to Kyuhsu University Hospital for possible LDLT for multiple HCCs and end-stage liver disease due to hepatitis C. Pretransplant imaging studies showed that she had 4 HCCs with a maximum tumor size of 2.5 cm in her cirrhotic liver. Her Child-Pugh score was 10 (grade C), and her Model for End-Stage Liver Disease score was 13. Her serum alpha-fetoprotein concentration was 416.6 ng/mL, and her serum des-gamma-carboxy prothrombin was 126 mAU/mL. The donor was the recipient's healthy 30-year-old daughter, whose blood type was identical to that of the recipient. A preoperative evaluation of the donor confirmed that she was negative for hepatitis B surface antigen and hepatitis C virus (HCV) antibody and had no evidence of malignancy. A left lobe living donor graft was procured and transplanted with methods described elsewhere.15, 16 The graft liver weight was 400 g, corresponding to 36.2% of the standard liver volume of the recipient. A histopathological examination of the explanted cirrhotic liver revealed 4 viable HCC nodules. The largest HCC, 2.3 cm in histopathological size and located in segment 6, was a moderately differentiated HCC with a trabecular growth pattern. Approximately 80% of the tissue showed coagulation necrosis due to previous transarterial chemoembolization (TACE; Fig. 1A,B). The other HCCs were as follows: (1) a 2.2-cm, moderately differentiated HCC in segment 8; (2) a 1.6-cm, well to moderately differentiated HCC in segment 8; and (3) a 0.8-cm, well-differentiated HCC in segment 4. All 4 HCCs were negative for any macroscopic or microscopic vascular invasion. Immunosuppression was induced with basiliximab (Simulect, Novartis Pharma, Basel, Switzerland), tacrolimus (Prograf, Astellas, Tokyo, Japan), and mycophenolate mofetil (Cellcept, Chugai Pharmaceutical Co., Ltd., Tokyo, Japan). Tacrolimus monotherapy was used for maintenance of the immunosuppression.
The histological recurrence of hepatitis C after LDLT was rapid. The serial follow-up graft liver biopsies performed at 4 months, 2 years, and 3 years after LDLT revealed A0F0, A1F3, and A1F4, respectively, according to the METAVIR scoring system17 (Fig. 1C,D). At 3 years after LDLT, her liver became decompensated with intractable ascites. Therefore, combination therapy with pegylated interferon alpha-2b (PEG-Intron, Schering-Plough KK, Tokyo, Japan) and ribavirin (Rebetol, Schering-Plough) was performed for 48 weeks. Although early viral response was not achieved, the serum was negative for HCV RNA 20 weeks after the induction of the treatment. After she completed the 48-week treatment, the serum was still negative for HCV RNA, SVR was achieved, and hepatic function recovered.
A follow-up computed tomography scan performed 6 years and 3 months after LDLT detected a 1-cm hepatic tumor in segment 3 of the left lobe graft with enhancement in the arterial phase and low density in the portal phase; these findings were consistent with the diagnosis of classical HCC (Fig. 2A). The tumor was also detectable as a typical defect for classical HCC in gadolinium-ethoxybenzyl-diethylenetriamine pentaacetic acid–enhanced magnetic resonance imaging. In conjunction with the new HCC, the serum alpha-fetoprotein level had increased from 19.5 to 89.7 ng/mL over the preceding 6 months, whereas the serum des-gamma-carboxy prothrombin level was still normal (16 mAU/mL). Liver function tests were as follows: total bilirubin, 0.5 mg/dL; aspartate aminotransferase, 30 IU/L; alanine aminotransferase, 11 IU/L; and prothrombin time (international normalized ratio), 95% (1.03). Partial hepatic resection in segment 3 was successfully performed. Currently, 4 months after the resection, the patient is doing well without any evidence of recurrent HCC. A histopathological examination of the resected liver specimen revealed that the tumor was a moderately differentiated HCC, 0.9 cm in size, with a trabecular growth pattern with fat deposits and desmoplastic changes. The noncancerous liver parenchyma was cirrhotic with a chronic inflammatory infiltrate in the fibrous stroma with a METAVIR score of A1F417 (Fig. 2B-D). The pathological findings could not reveal whether the HCC originated from the recipient or the donor.
To determine the origin of the HCC, we performed genotyping by microsatellite analysis of DNA samples extracted from the resected HCC in the graft, the noncancerous graft liver, the peripheral blood of the donor, the peripheral blood of the recipient, and the noncancerous liver of the explanted liver of the recipient. The HCC samples of the explanted liver at LDLT were not available. Six loci in 5 chromosomes were examined for each sample. Polymerase chain reaction products for the loci were analyzed with an Applied Biosystems 3130 genetic analyzer using Genemapper software (PE Applied Biosystems, Foster City, CA). All genotypes at all 6 loci of the HCC completely matched those of the noncancerous graft liver and the peripheral blood of the donor. Donor-specific alleles were confirmed at 3 loci (D5S346, D13S171, and D17S855) of the HCC, which were not detectable in the peripheral blood of the recipient or explanted liver of the recipient. In contrast, genotypes at 5 loci of the HCC mismatched those of the peripheral blood of the recipient and the explanted liver of the recipient, except at 1 locus (D5S492), which was originally compatible between the recipient and the donor, excluding the recurrent HCC from the recipient. At least 1 of the alleles matched between the donor and the recipient because the donor was the recipient's daughter. These findings confirmed that the HCC originated from the transplanted liver and was thus a de novo HCC (Table 1 and Fig. 3).
Table 1. Microsatellite Analysis of DNA Extracted from the HCC in the Graft, the Noncancerous Graft Liver, the Peripheral Blood of the Donor, the Peripheral Blood of the Recipient, and the Noncancerous Liver Parenchyma of the Explanted Liver of the Recipient
DNA Allelotype (Comparison with HCC)
Abbreviations: HCC, hepatocellular carcinoma; N, mismatched; Y, matched.
After LT for HCC, recurrent HCC is a well-known problem and has been vigorously discussed.1–3 On the other hand, de novo HCC in a liver graft is rare, and only a few cases have been reported, each of which had no previous evidence of HCC.4–10 The etiologies of the previously reported cases of de novo HCC included hepatitis C cirrhosis (4 cases), hepatitis B cirrhosis (4 cases), and alcoholic cirrhosis (1 case). It is certain that HCC occurring in the cirrhotic graft liver without previous evidence of HCC naturally evokes de novo HCC. In contrast, the present case is, to the best of our knowledge, the first reported case of de novo HCC after LT for HCCs, rather than recurrence.
Follow-up after LT for HCC with HCV should be systematic to survey the recurrence due to both the donor-derived de novo HCC and the recipient's remnant HCC. An assessment of hepatitis C and possible treatment for hepatitis C is also essential. Of the 4 previous cases with de novo HCC with recurrent hepatitis C, none had achieved SVR. This is the first report of de novo HCC in a liver graft with SVR after LT for hepatitis C. Although SVR reduces HCC development in patients with hepatitis C cirrhosis in a nontransplant setting,11 HCC can occur after SVR is achieved.11–14 Similarly, after LT for hepatitis C, the achievement of SVR is expected to reduce the risk of de novo HCC but cannot entirely exclude it. Makiyama et al.12 reported that male sex, older age (≥50 years old), and more advanced histological fibrosis (≥F3) before interferon therapy were independent risk factors for the development of HCC in SVR patients. In the present case, the liver biopsies were graded as F3 before treatment was started and as F4 during the treatment, indicating the high risk of HCC development. The achievement of SVR before the progression of fibrosis is thought to be important to prevent the development of de novo HCC after LT. The molecular mechanism of carcinogenesis in the liver with SVR is unknown but is presumed to differ from that for non-SVR livers. We previously reported that HCCs with SVR more frequently expressed p53 than those without SVR.13 Hayashi et al.14 reported that the methylation of p16 in HCCs with SVR was more frequent than that in HCCs without SVR. After LT, the transplanted liver shows chimerism of the vascular endothelium, bile duct epithelium, and hepatocytes derived from a recipient's extrahepatic stem cells.18 Carcinogenesis with SVR after LT may be more complex because of intrahepatic chimerism and immunosuppression. To consider de novo HCC, the time that elapsed between LT and the identification of the new HCC is of interest. Most recurrences occur within 2 years after LT,1–3 whereas the earliest case of de novo HCC occurred 4 years and 11 months after LT,5 and the latest occurred 22 years after LT.7 In the present case, the HCC occurred 6 years after the primary LDLT for HCCs; this is too late for usual recurrence after LT, suggesting de novo HCC instead. Meanwhile, HCC in the natural history of chronic hepatitis C usually occurs after about 30 years of chronic infection.19 The present de novo HCC after LT occurred earlier than the usual hepatitis C–induced HCC. The early carcinogenesis in the present case may be associated with the immunosuppression and rapid establishment of cirrhosis.
The therapeutic strategy for de novo HCC after LT has never been established because of the small number of cases. In the previous cases, various therapies were performed, including 5 cases of retransplantation: 1 case involved pretransplant hepatic resection, 1 case involved pretransplant TACE,4, 6, 8–10 1 case underwent hepatic resection,8 1 case underwent radiofrequency ablation, and 1 case underwent TACE.7 De novo HCC could be treated as a form of primary HCC; thus, it seems reasonable to perform hepatic resection with consideration of tumor factors, liver function, and the patient's general condition. Although retransplantation for de novo HCCs with established cirrhosis could be the procedure of choice, it is difficult to perform, particularly in countries such as Japan in which the availability of deceased donors is very restricted. The tapering or withdrawal of immunosuppression may be more effective for de novo HCC than for recurrent HCC because of immunogenic differences. In the present case, the trough level of tacrolimus has been kept below the measurable limit, without graft rejection, since the de novo HCC was detected.
There are various methods available for identifying the origin of a tumor. In the previous cases with de novo HCC, none of which had a previous history of HCC, genetic analysis to confirm the tumor origin was not thought essential and was performed in only 3 of the cases. Fluorescent in situ hybridization for X and Y chromosomes is available only in sex-mismatched cases,4 and this was not applicable to the present case of a female recipient and her daughter's donation. Microsatellite analysis is useful and definitive for confirming the origin of the tumor without the limitation of sex mismatch.8 The technical procedure has already been established in our institute for the analysis of microsatellite instability in human cancers.20 It is natural that the microsatellite analysis of the resected HCC in this case showed chimerism.18 However, the clear donor-dominant allelic pattern of the HCC strongly suggested that the HCC had originated from the donor's hepatocytes. If the HCC had been recurrent HCC due to the presence of remnant HCC cells of the recipient, the allelic pattern would have been recipient-dominant or recipient-only.
In conclusion, this is, to the best of our knowledge, the first report of a genetically confirmed de novo HCC in a liver graft with SVR after LT for advanced HCCs and hepatitis C cirrhosis. The de novo HCC originated from the graft with sustained HCV clearance and continued graft fibrosis. De novo HCC can even occur after the achievement of SVR in a graft liver. Therefore, regardless of SVR achievement and a previous history of HCC, patients undergoing LT for hepatitis C should be strictly monitored for HCC occurrence with imaging tests, measurements of tumor markers, and liver biopsies for evaluating the degree of fibrosis.
The authors thank Tomoko Konishi, Megumi Kiyota, and Takako Shishino for their excellent technical assistance.