| |
Hepatocellular carcinoma in long-term sustained virological responders following antiviral combination therapy for chronic hepatitis C
|
| |
| |
Jnl of Viral Hepatitis Sept 2008
T.-M. Scherzer 1 , K. R. Reddy 2 , F. Wrba 3 , H. Hofer 1 , K. Staufer 1 , P. Steindl-Munda 1 , A. Gangl 1 and P. Ferenci 1
1 Department of Internal Medicine III, Division of Gastroenterology and Hepatology, Medical University of Vienna, Vienna, Austria ; 2 Division of Gastroenterology, Department of Medicine, University of Pennsylvania, Philadelphia, PA, USA ; and 3 Department of Clinical Pathology, Medical University of Vienna, Vienna, Austria
"In the present paper we report five cases of HCC in European and North American long-term sustained virologic responders to antiviral combination treatment with interferon alpha and ribavirin for chronic hepatitis C, including three patients without cirrhosis. The occurrence of HCC in noncirrhotic patients with chronic hepatitis C in general, and in sustained virologic responders in particular, is rare. This observation......calls for long-term follow up of successfully treated patients......The risk factors for HCC in patients with sustained virologic response to interferon therapy were the stage of fibrosis (F3/F4), age (>60 years) and excessive alcohol (>50 g/day) consumption [20]. Similar observations were made in Italian patients followed for up to 10 years after interferon treatment [19]......In our study, three of the five patients were noncirrhotic, both before antiviral therapy and at diagnosis of HCC....Since this is an observational study we do not know the true incidence of HCC in sustained virologic responders long term......In the Austrian cohort at least we can estimate the incidence of HCC to be around 1% within five years after successful antiviral treatment.....One (patient no. 1; then considered as a 'long time sustained virologic responder') of the 187 sustained virologic responders with long-term follow up published by us [11] developed HCC. Since then we have followed a total of about 300 patients including the three cases reported here......most sustained virologic responders in the Austrian cohort were noncirrhotic, thus the HCC incidence was expected to be lower....The reason for HCC development in non-cirrhotic patients with SVR is still unknown, but could be explained by a direct oncogenic effect of HCV before antiviral therapy and the development of microcarcinomas, which may take a long time to cause symptoms. As time goes by and we have had the opportunity to follow large cohorts of sustained virologic responders, particularly those with advanced fibrosis/cirrhosis, we are likely to get a better idea of the incidence of HCC in such populations."
ABSTRACT
Summary. Antiviral treatment results in a sustained virologic response (SVR) in 50-75% of patients with chronic hepatitis C. Long-term follow up studies have observed ongoing SVR in the overwhelming majority of them. Thus chronic hepatitis C is considered 'cured' if an SVR is achieved. Consequently, it is expected that in sustained virologic responders long-term complications of hepatatic C virus (HCV) related chronic liver disease including hepatocellular carcinoma are eliminated or have a decreased incidence. We report on five patients (three from Austria, two from USA) who developed hepatocellular carcinoma during follow up (3-6 years) after achieving SVR. During follow up and at diagnosis all were HCV-RNA neg. None of the patients had other liver diseases. One patient presented with bilateral adrenal metastasis, the remaining four with large hepatic tumours. Three patients were noncirrhotic at the start of treatment at the time of tumour diagnosis. Successful antiviral treatment in HCV patients does not prevent development of hepatocellular carcinoma even in non-cirrhotic livers. Long-term follow up of patients with SVR is mandatory and should include surveillance for hepatocellular carcinoma.
Introduction
Hepatocellular carcinoma (HCC) is the fifth most common carcinoma in men and the ninth in women and affects approximately half a million persons worldwide [1]. Main risk factors for HCC in the USA are cirrhosis due to chronic hepatitis C (47%), chronic hepatitis B (15%), and coinfection with HCV and HBV (5%) [2]. The risk factors for HCC in Austria are similarly dominated by viral hepatitis (HBV: 9.8%, HCV: 36.7% HBV/HCV: 1.6%) followed by alcoholic cirrhosis (35.1%) [3]. In the United States an estimated 3.9 million people are infected with the hepatitis C virus (HCV) [4], whilst in Western Europe the prevalence of chronic HCV infection varies from about 0.1% in the North to 1% of the population in the South and ranges from 0.4% to 4.5% (median 2%) in Eastern Europe. Chronic hepatitis C is a slowly progressive disease leading to liver cirrhosis [5]. Once cirrhosis is present the risk to develop HCC is high [6].
Chronic hepatitis C is considered a curable disease following combination therapy with peginterferon and ribavirin. Sustained virologic response (SVR) rates range between 45% and 60% in patients infected with genotypes 1 and 4 and up to 80-90% in patients with genotypes 2 and 3 [7-10]. Sustained virologic response is maintained in about 98-99% of patients over long observation periods [11,12] and is associated with improvement of liver fibrosis [13,14]. Hepatitis C is considered 'cured' if SVR in non-cirrhotic patients is achieved [15]. Thus, successful antiviral treatment may substantially decrease the rate or even prevent the occurrence of HCC in patients with or without cirrhosis [13,16-19]. However, most of these studies observing a favourable impact on the incidence of HCC are from Asia, are retrospective and have a potential selection bias [17,18,20]. Overall the rate of HCC occurrence in sustained virologic responders was much lower than in nonresponders. In an Italian study the incidence of HCC in patients with HCV cirrhosis treated by interferon monotherapy was 2.59-fold higher in non responders than in sustained responders [19]. There are no large follow up studies in patients successfully treated with pegylated interferons and ribavirin. In a long-term follow up study in Austria, 187 sustained virologic responders were observed for more than 12 months post therapy (median 29 months). No recurrence of HCV or evidence for progressive liver disease was seen and not a single case of HCC was detected [11].
In this report we describe five unique cases in European and North American patients, including three non-cirrhotic patients who developed HCC several years after successful antiviral therapy with pegylated interferon alpha and ribavirin.
Following the Discussion are the Case Reports.
Discussion
In the present paper we report five cases of HCC in European and North American long-term sustained virologic responders to antiviral combination treatment with interferon alpha and ribavirin for chronic hepatitis C, including three patients without cirrhosis. The occurrence of HCC in noncirrhotic patients with chronic hepatitis C in general, and in sustained virologic responders in particular, is rare. This observation questions whether sustained virologic response really represents a 'cure' of the disease and calls for long-term follow up of successfully treated patients.
The occurrence of HCC was observed in up to 4.6% of sustained virologic responders to treatment with interferon monotherapy from Asia [15,17,18,24-26]. Most of these studies are retrospective and have a potential selection bias. Furthermore most of the patients were treated with antiviral therapies, which are not the standard of care today. Also the virologic tests to document response to therapy were less sensitive than those used today. The incidence of HCC in transient responders was as low as in sustained virologic responders, but was 7.9-fold higher in nonresponders [15]. The risk factors for HCC in patients with sustained virologic response to interferon therapy were the stage of fibrosis (F3/F4), age (>60 years) and excessive alcohol (>50 g/day) consumption [20]. Similar observations were made in Italian patients followed for up to 10 years after interferon treatment [19].
In our study, three of the five patients were noncirrhotic, both before antiviral therapy and at diagnosis of HCC. Only one similar case has been reported so far [27]. In one patient no primary HCC was detected, in spite of bilateral adrenal metastases. Adrenal metastasis as first presentation of HCC has been reported in a few patients with chronic hepatitis [28] but to our knowledge not in an HCV patient after successful antiviral treatment. One apparent common factor in our patients was that they had prolonged treatment exposures: three were relapsers to the first therapy and became sustained virologic responders after retreatment (patients 1-3) and one patient (no. 4) was treated with lower doses of peginterferon and ribavirin for 2 years. Since this is an observational study we do not know the true incidence of HCC in sustained virologic responders long term. The two reported US cases were referrals and do not allow any estimation of the incidence of HCC. In the Austrian cohort at least we can estimate the incidence of HCC to be around 1% within five years after successful antiviral treatment. One (patient no. 1; then considered as a 'long time sustained virologic responder') of the 187 sustained virologic responders with long-term follow up published by us [11] developed HCC. Since then we have followed a total of about 300 patients including the three cases reported here. Nevertheless, since this was not a prospective study and the long-term outcome of some of the patients is unknown, this figure remains uncertain. In contrast to the Italian study [19], most sustained virologic responders in the Austrian cohort were noncirrhotic, thus the HCC incidence was expected to be lower. Whether the risk for HCC in virologic non-responders can be lowered by long-term interferon therapy is currently being investigated in two prospective trials (HALT-C and EPIC [29,30]).
The precise mechanism by which HCV infection causes HCC is not known. Important factors are non-specific inflammation-related processes which may favour protumour growth factor signalling [31] and facilitate angiogenesis [32]. The development of cirrhosis is usually the cause of HCC in HCV infection. However, three of our five patients had no cirrhosis, even at the time when HCC was detected. All were HCV-RNA negative using a sensitive PCR test throughout follow up, including the time of diagnosis of HCC. In two patients, both the liver and tumour tissue were tested for the presence of the virus and no virus was detected.
Although there was no evidence of viral persistence in our patients, it is conceivable that sustained virologic response may not result in complete elimination of the virus and that the virus is sequestered in small quantities in hepatic and non-hepatic sites. Several recent observations report that HCV can be detected in various compartments in consequently HCV-RNA negative patients, including hepatocytes, peripheral blood mononuclear cells, lymphocytes and macrophages. Continuous viral presence could result in persistent humoural and cellular immunity for many years after therapy and could present a potential risk for infection reactivation [33]. Another important observation is the detection of occult hepatitis C infection by highly sensitive reverse transcriptase-polymerse chain reaction (RT-PCR) and in situ hybridisation in hepatocytes and peripheral blood mononuclear cells of 57 of 100 patients with persistently abnormal liver enzymes but no markers of HCV infection by commercial assays [34]. Patients with occult HCV infection were more likely to have necroinflammatory activity and fibrosis than patients without intrahepatic HCV RNA. The impact of the presence, quantity, and location of replicating virus on clinical outcome remains to be studied. Viral persistence in the liver in some patients may play a minor role in sustained responders but successful antiviral therapy was shown to decrease inflammation even in patients with 'occult' HCV infection [35,36]. Thus, the presence of minute amounts of non-replicating viruses is unlikely to contribute to carcinogenesis and HCV, unlike HBV, does not integrate into the host genome [15]. Occult HBV infection indicated by detectable anti-HBc in serum in interferon treated HCV sustained responders may play a role in the development of HCC [37,38]. Only one of our patients was anti-HBc positive.
A second possibility would be that premalignant foci were already present at the time when antiviral therapy was started. Transition of premalignant foci to HCC may happen even without the presence of the virus [39]. In most Japanese studies HCC was diagnosed shortly after achieving SVR, an observation consistent with this possibility. The incidence of preneoplastic lesions such as foci of altered hepatocytes with small-cell changes was higher in cirrhotic livers than that without HCC [40]. The presence of large liver cell dysplasia in patients with liver cirrhosis due to viral hepatitis is an independent risk factor for HCC [41]. Thus, liver biopsy may identify a subgroup of patients at higher risk of developing HCC [41].
In summary these observations underline the importance of long-term follow up in patients with SVR. The frequency and methods for long-term surveillance of noncirrhotic patients remain to be established. The reason for HCC development in non-cirrhotic patients with SVR is still unknown, but could be explained by a direct oncogenic effect of HCV before antiviral therapy and the development of microcarcinomas, which may take a long time to cause symptoms. As time goes by and we have had the opportunity to follow large cohorts of sustained virologic responders, particularly those with advanced fibrosis/cirrhosis, we are likely to get a better idea of the incidence of HCC in such populations.
Case Reports
The key characteristics of the patients are shown in Table 1.
Case 1 (Austria)
Chronic non-A, non-B infection was diagnosed in a male patient in 1987 (45 years old). He was presumed to be infected with HCV when he received about 50 U of blood following severe upper gastrointestinal bleeding in 1982. He was moderately obese (BMI=27) and was on treatment with metformin 850 mg/day for type 2 diabetes. Subsequently he was diagnosed with HCV infection (Genotype 3a). From October 1997 to August 1998 he was treated with 5 MU TIW interferon-_n1 (Wellferon; Borroughs Welcome, London, UK) for a total of 10 months and achieved an end of treatment response but relapsed 3 months later. Liver ultrasonography 2 years later revealed hepatomegaly with diffuse parenchymal damage. A liver biopsy showed stage 2 fibrosis (according to Metavir Score [21]) with 50% steatosis and a moderate iron deposition (grade II). A year later he participated in a prospective therapy trial [22] and received 1200 mg of ribavirin/day in combination with 5 MU interferon-alpha 2 b/day (Intron A; AESCA-Schering Plough, Traiskirchen, Austria) for 14 weeks followed by 5 MU/every other day for the following 24 weeks. He achieved SVR in December 2000. Thereafter he was followed at yearly intervals. Up until 2005, hepatic biochemical tests were normal, HCV-RNA (COBAS AMPLICOR; Roche-Diagnostics, Pleasanton, CA, USA, detection limit <50 IU/mL) was consistently undetectable, and AFP levels were normal. In December 2005 routine ultrasonography showed a normal liver, but a tumour in the left adrenal gland (0 11 cm). After confirming this finding by magnetic resonance imaging (MRI) and computer tomography (CT) the left kidney and the left adrenal gland were removed laparoscopically in April 2006. Histological evaluation revealed adrenal metastasis from HCC (Fig. 1). Immunohistochemistry using monoclonal antibodies against CK8 and hepatocytes respectively (Dako, Glostrup, Denmark) revealed strong reactivity with the tumour cells, whereas the surrounding adrenal tissue showed consistently negative results, proving that the tumour cells were of hepatocytic origin. No primary cancer was detected in the liver by inspection during surgery, CT-scan and MRI. A lesion in the right adrenal gland was also detected and was considered to be benign. Four months later, a tumour was detected in the right adrenal (0 7-8 cm) and a right adrenalectomy was performed. At surgery, the liver appeared macroscopically normal and intraoperative ultrasonography revealed no hepatic lesions. The histology of the right adrenal gland tumour was identical to the left one. After surgery the patient was doing well on hormonal substitution therapy. HCV-RNA remained undetectable (AmpliPrep/COBAS TaqMan HCV Test; Roche Diagnostics, limit of detection <10 IU/mL) before and after surgery. During follow up neither a primary liver tumour nor recurrent metastases were detected.
Case 2 (Austria)
Chronic hepatitis C (genotype 1 b) was diagnosed in a female patient in 1999 (60 years old). In 1990 she underwent local excision of a cancer in her left breast, followed by chemotherapy and radiation. During treatment, she received blood transfusions. In 2000, liver biopsy showed fibrosis stage 3 (Metavir). A 48-week treatment with 5 MU/QD interferon alpha-2b and 1.2 g/day of ribavirin was initiated in June 2000. On treatment and at end of treatment, HCV-RNA became undetectable by qualitative polymerase chain reaction (PCR). One month after end of treatment she relapsed. After retreatment with 180 μg pegylated interferon alpha-2a/week (PEGASYS; Roche, Basel, Switzerland) and 1200 mg ribavirin/day (COPEGUS; Roche) for another 48 weeks (July 2001-June 2002) the patient achieved SVR. From then onwards at yearly follow up examinations (including December 2006) HCV-RNA (by AmpliPrep/COBAS TaqMan HCV Test) was undetectable in serum. In September 2005 an elevated serum AFP value (46.8 ng/mL, normal: -8.5) was found; liver sonography did not show any pathological findings. By October 2006 the AFP level had risen to 86 ng/mL. Sonography and a CT scan of the liver revealed a 5 cm tumour in segment 5/6. A CT-guided biopsy showed HCC, grade 2 (Edmondson and Steiner classification [23]). No HCV-RNA was detectable by PCR in the tumour tissue. In January 2007 she underwent a resection of segment 5/6. The resected specimen revealed fibrosis stage 2 (Fig. 2). At the last visit in February 2008 she was well and a CT-scan showed no evidence of recurrent tumour.
Case 3 (Austria)
Hepatitis C genotype 1b was diagnosed in a 44-year-old woman during routine evaluation in August 2000. She had no known risk factors for HCV infection. Liver biopsy revealed fibrosis stage 4 (Metavir). She was treated with 180 μg peginterferon alpha-2a weekly and 800 mg ribavirin daily from September 2000 to September 2002. She relapsed after 3 months. After retreatment (initially with peginterferon alpha-2a monotherapy for 23 weeks followed by additional 47 weeks in combination with 1000 mg ribavirin) she achieved a sustained virologic response in January 2005. She was followed up at six monthly intervals and had normal transaminases and AFP levels until the end of 2006. In April 2007 a routine blood test showed abnormal hepatic biochemical tests and an AFP level of 4442 kIU/L. Sonography and MRI showed an 8 _ 20 cm lesion in the left lobe of the liver and many 5 cm diameter lesions in the right lobe and in the caudate lobe. Biopsy of a lesion demonstrated HCC stage 2 (Edmonson and Steiner), Furthermore, metastatic lesions were detected in the vertebrae. HCV-RNA was not detected in the tumour and in adjacent non-tumour tissue specimens by PCR. She died shortly after diagnosis was made.
Case 4 (USA)
A 53-year-old man was referred in 2006 after the detection of a large liver mass following the development of right upper quadrant abdominal pain and persistent fever. He also noted jaundice with dark urine and light coloured stools. The patient was previously diagnosed with hepatitis C infection (genotype 1a) which epidemiologically was related to a blood transfusion 27years previously. He had also consumed significant amounts of alcohol between the ages 20 and 21 followed by abstinence.
In 2001 the patient was initiated on a prolonged course of a modified dose of pegylated Interferon alfa-2b (PegIntron; Schering-Plough, Kenilworth, NJ, USA; 60 μg/week) and ribavirin (Rebetol, Schering-Plough; 800 mg/day) and he was treated for 2 years. The liver biopsy performed prior to the treatment revealed liver cirrhosis. This therapy resulted in SVR with HCV RNA of less than 10 IU/mL yearly after cessation of treatment and also 5 months prior to the detection of the liver mass.
On physical examination he was mildly icteric, had a few spider angiomata and had some ascites. He did not have any hepatomegaly or splenomegaly. HCV-RNA was negative and alpha-fetoprotein was 2 kIU/L (normal: 0-5). MRI of the abdomen showed a 14.6 x 9.0 x 5.7 cm lesion in the left lobe of the liver with heterogeneous uptake and a central area of low attenuation. His spleen was normal. There was enlargement of the left adrenal gland without any definite mass. The liver was felt to show features consistent with cirrhosis.
He underwent a percutaneous liver biopsy which revealed features of moderately differentiated HCC with extensive infarction. Due to the advanced nature of his tumour, he was not considered to be a candidate for surgical resection or liver transplantation and was placed on chemotherapy. The patient is doing well with stability in the size of the lesion.
Case 5 (USA)
A 52-year-old African-American man was referred in 2006 with a large mass lesion which was detected after he developed right upper quadrant abdominal pain and weight loss of 14 pounds over 4 months. He had been diagnosed with hepatitis C infection (Genotype 1b) 3 years previously. His social history was significant for intravenous drug use 25 years previously. Following the diagnosis of HCV infection, he underwent a liver biopsy which showed grade 1 to 2 inflammation and stage 0 fibrosis (Metavir). He was then treated for 1 year with 1.5 μg/kg pegylated interferon alfa-2b and 1200 mg/day ribavirin and achieved SVR with negative HCV RNA by a quantitative assay.
Physical examination was remarkable for hepatomegaly. His abdominal CT scan and MRI showed a 15 cm mass in the right lobe of the liver with a central area of decreased CT attenuation compatible with necrosis or a scar. He also underwent a CT guided biopsy which revealed viable and poorly differentiated tumour cells. Immunohistochemical staining showed focal immunoreactivity with the biliary differentiation markers AE1 and AE 3 but tested negative for other antigens. Given the possibility of metastatic disease the patient underwent an upper endoscopy and colonoscopy which showed a small hiatal hernia and a single polyp in the ascending colon, respectively.
He underwent a staging laparoscopy prior to any attempt at resection of this lesion (Fig. 3). It revealed a non-cirrhotic liver but there were tiny peritoneal implants on the right side superiorly in the right diaphragmatic area of the peritoneum. Biopsies confirmed metastatic carcinoma with morphology consistent with origin from a primary HCC. Therefore, he was placed on chemotherapy including bevacizumab, oxaliplatin and capecitabine.
|
|
| |
| |
|
|
|
