Risk of Hepatocellular Carcinoma Development in Cases of Hepatitis C Treated by Long-Term, Low-Dose PEG-IFNα-2aHagiwara S. · Sakurai T. · Takita M. · Ueshima K. · Minami Y. · Inoue T. · Yada N. · Kitai S. · Nagai T. · Hayaishi S. · Arizumi T. · Nishida N. · Kudo M.
Department of Gastroenterology and Hepatology, Kinki University School of Medicine, Osaka, Japan Corresponding Author
Objective: Increasing evidence suggests the efficacy of maintenance therapy with interferon (IFN) for chronic hepatitis C (CHC) in reducing the risk of hepatocellular carcinoma (HCC). The aim of this study was to determine clinical characteristics on the risk of occurrence of HCC in CHC patients receiving maintenance IFN therapy. Methods: A total of 55 patients were treated in a single center with PEG-IFNα-2a monotherapy for CHC and evaluated for variables predictive of the occurrence of HCC. Results: The cumulative incidences of HCC were 0.092, 0.117 and 0.161 at 3, 5 and 7 years, respectively. Serum ALT level (>40 IU/l) in the 6th month after commencement of IFN therapy and BMI >25 were associated with shorter time-to-HCC emergence using multivariate analysis (relative risk 16.034, p = 0.01 for ALT >40 IU/l; relative risk 6.020, p = 0.026 for BMI >25, respectively). The IL28B SNP was extracted as a significant factor for the occurrence of HCC. Conclusions: Maintenance therapy with the use of long-term low-dose PEG-IFNα-2a is effective for preventing HCC occurrence irrespective of the IL28B SNP, at least for a subset of CHC patients. The initial response of serum ALT levels and BMI provides a prognostic value for determining the risk of developing HCC later in life.
Copyright © 2012 S. Karger AG, Basel
Chronic infection of hepatitis C virus (HCV) is a leading cause of developing liver cirrhosis (LC) and hepatocellular carcinoma (HCC) in many countries, including Japan, the United Sates and some of Western Europe [1,2,3,4,5]. Preventing the development of HCC and liver failure due to advance of LC is a primal goal for treatment of chronic hepatitis C (CHC), and interferon (IFN) treatment has been applied to achieve it.
On the one hand, a number of studies have shown that IFN treatment reduces the risk of HCC emergence and improves the disease-free survival. In the randomized control study of type C LC, Nishiguchi et al.  reported that IFN treatment significantly suppressed the occurrence of HCC after a median follow-up of 4.9 years even in the case with advanced liver fibrosis. In addition, the Inhibition of Hepatocarcinogenesis by Interferon Therapy (IHIT) study, a multicenter collaborative cohort study, also revealed that IFN treatment significantly reduced the occurrence of HCC in CHC patients compared with those without any history of IFN therapy (odds ratio 0.516, p < 0.001), and sustained virological response (SVR) reduced the annual occurrence of HCC regardless of stage of liver fibrosis (F-stage) . Taken together, it is conceivable to believe that achievement of SVR suppresses the occurrence of HCC dramatically.
On the other hand, a subset of CHC is known to be refractory to IFN therapy and hard to achieve SVR, such as the elderly and those with advanced liver fibrosis and background disease. In such cases, long-term low-dose IFN therapy appears to offer an option for treatment failures with advanced disease. It has been used in an attempt to prevent disease progression and HCC rather than viral eradication. Arase et al.  reported that long-term low-dose IFN therapy led to a reduction of serum AFP as well as ALT levels, even in the cases without sufficient viral response. In addition, the incidence of HCC reportedly decreased among the elderly patients of 60 years or older compared to those without treatment. The Hepatitis C Antiviral Long-Term Treatment against Cirrhosis (HALT-C) trial also revealed that administration of 90 µg of PEG-IFNα-2a once weekly significantly suppressed the emergence of HCC after a median follow-up of 6.1 years among non-responder cases on a previous combination therapy of PEG-IFNα-2a and ribavirin (RBV) (HR 0.45, p = 0.01) .
Given the results of several published reports, maintenance IFN therapy suppresses the occurrence of HCC to a certain extent, however the carcinogenic risk factor among the cases treated with low-dose IFN is still unknown. In this study, we clearly focus on this gap of knowledge. Long-term, low-dose PEG-IFNα-2a monotherapy was performed in 55 cases with CHC and risk factors for tumor development were assessed retrospectively. Here, we report that the initial response of serum ALT levels and BMI was a significant contributor for developing HCC among the cases who underwent maintenance IFN therapy, however IL28B single nucleotide polymorphism (SNP) did not affect the occurrence of HCC later in life.
Seventy patients with CHC were treated with PEG-IFNα-2a monotherapy between February 2005 and December 2010 at Kinki University Hospital. Among them, 55 were selected for this study based on the following criteria: (1) positive for anti-HCV and for HCV RNA using RT-PCR; (2) infected with HCV genotype 1; (3) naive to IFN therapy; (4) chronic hepatitis or cirrhosis confirmed by liver biopsy within 6 months prior to initiation of IFN therapy, and (5) leukocyte count >2,500/mm3, platelets >50,000/mm3 and serum bilirubin level <2.0 mg/ml. The exclusion criteria were: (1) prior history of HCC; (2) positive for hepatitis B surface antigen (HBsAg); (3) coinfection with human immunodeficiency virus, and (4) cases with hemochromatosis, Wilson’s disease, alcoholic liver disease, autoimmune liver disease and primary biliary cirrhosis. HCV genotype was determined by polymerase chain reaction (PCR) amplification of the core region of the HCV genome using genotype-specific PCR primers. Serum HCV-RNA level was determined by TaqMan PCR assay (Cobas Amplicor HCV Monitor; Roche Molecular Systems, Pleasanton, Calif., USA). SNP near the IL28B gene was determined for all cases enrolled in this study. Samples for the genome-wide association survey were genotyped using the Illumina HumanHap610-Quad Genotyping BeadChip. Genotyping data were subjected to quality control before the data analysis. Genotyping for replication and fine mapping was performed using the Invader assay, TaqMan assay, or direct sequencing, as described previously [10,11]. In this study, genetic variation near the IL28B gene (rs8099917), reported as a pretreatment predictor of treatment efficacy in Japanese patients, was investigated [12,13].
Written informed consent was obtained from each patient for this study. The study protocol was approved by the Ethics Committee at Kinki University Hospital and performed in compliance with the Helsinki Declaration.
All cases received subcutaneous injections of PEG-IFNα-2a weekly with an initial dose of 90 µg/week. The duration of IFN therapy ranged from 48 to 267 weeks (median 70).
Follow-up of the patients began on the first day of IFN treatment with monthly evaluation of general condition, biochemical and hematological tests until death or the last hospital visit. For screening of emergence of HCC, ultrasonography, dynamic contrast-enhanced computed tomography (CT), dynamic contrast-enhanced magnetic resonance imaging (MRI), and/or measurement of tumor markers (AFP and DCP) were performed for all patients every 3–6 months. Serum AFP concentration was measured by chemiluminescent enzyme immunoassay (Lumipalse AFP-N; Fujirebio, Tokyo, Japan). Plasma DCP concentrations were measured by an electrochemiluminescent immunoassay (Picolumi PIVKA2; Eisai, Tokyo, Japan). The follow-up periods of all cases ranged from 1.8 to 7.4 years (median 4.6).
Liver biopsy specimens were obtained percutaneously, fixed in 10% formalin, digested with diastase and stained with hematoxylin. Histopathological diagnosis was made independently by an experienced liver pathologist who had no clinical information. All specimens for examination contained six or more portal areas. Baseline liver histology was diagnosed according to the scoring system of Desmet et al. .
Among the 55 patients without HCC at the initiation of IFN, the prognostic factors involved in the development of HCC were compared by univariate analysis using the log-rank test between patients who developed HCC and those without emerging HCC during follow-up periods. Subsequently, multivariate analysis using a Cox regression model was performed and variables with significant differences determined. In addition, for the factors with significant differences determined by multivariate analysis, the hazard ratio (RR) for emergence of HCC was calculated. p < 0.05 was regarded as significant. All the analyses described above were performed using SPSS version 11.5 (SPSS, Inc., Chicago, Ill., USA).
Table 1 shows the characteristics of the 55 patients who were enrolled and underwent PEG-IFNα-2a monotherapy. Median age was 65 years (range 44–78) and median BMI was 22.4 (range 16.2–29.8). 29 liver biopsies (53% of the total cases) revealed LC whereas, 26 were diagnosed as chronic hepatitis of F1–F3. The baseline level (before IFN therapy) of median serum AFP was 8 ng/ml (range 1–209). 40 cases carried the major allele (TT) of IL28B and 15 showed minor (TG or GG).
|Table 1. Demographics of the enrolled patients (n = 55)|
During the follow-up periods (median 4.6 years), HCC was diagnosed in 8 (14.5%) patients. The cumulative incidences of HCC were 0.092, 0.117 and 0.161 at 3, 5 and 7 years, respectively (fig. 1).
|Fig. 1. Incidence of HCC. HCC was diagnosed in 8 (14.5%) patients during follow-up. The cumulative incidences of HCC were 0.092, 0.117 and 0.161 at 3, 5 and 7 years, respectively.|
Predictive factors for the occurrence of HCC were analyzed in all 55 patients using the log-rank test (table 2). Univariate analysis showed that BMI >25 (p = 0.02) and baseline AFP level >10 ng/ml (p = 0.04) were associated with emergence of HCC during follow-up periods. In addition, regarding the AFP and ALT levels during the IFN therapy, higher serum AFP (>10 ng/ml) and serum ALT measured at 6 months after commencement of IFN therapy (>40 IU/l) showed a significant correlation with HCC occurrence (p = 0.01 for serum AFP of post-IFN and p = 0.0007 for serum ALT of post-IFN, respectively). However, interestingly, neither the IL28B SNP nor an undetectable level of HCV RNA at 6 months after commencement of IFN therapy was extracted as a significant factor.
|Table 2. Comparison between patients with and without developing HCC|
To determine the independent variables and narrow down the risks for emergence of HCC after maintenance IFN therapy, we subsequently performed multivariate analysis using a Cox regression model, including the four variables detected as a significant by univariate analysis (table 3). Among them, ALT >40 IU/l at 6 months after commencement of IFN therapy (relative risk 16.034, p = 0.01) and BMI >25 (relative risk 6.020, p = 0.026) were significantly and independently associated with a higher risk of HCC emergence.
|Table 3. Summary of stepwise multiple Cox regression analysis of the occurrence of HCC|
Figure 2 shows Kaplan-Meier survival curves of patients classified based on the subgroups listed in table 3. The cumulative incidences of HCC in patients with ALT >40 IU/l at 6 months after commencement of IFN were much higher than those with ALT ≤40 IU/l throughout the follow-up periods (cumulative incidences of HCC in patients with ALT >40 IU/l were 0.056, 0.264 and 0.501 at 3, 5 and 7 years, and those in patients with ALT ≤40 IU/l were 0, 0.05 and 0.05 at 3, 5, and 7 years, respectively, p = 0.0007 by log-rank test; fig. 2a). Similarly, the cumulative incidences of HCC in patients with BMI >25 were 0, 0.524 and 0.524 at 3, 5 and 7 years, whereas those with BMI ≤25 were 0.022, 0.052 and 0.187 at 3, 5 and 7 years, respectively (p = 0.02 by log-rank test; fig. 2b).
|Fig. 2. Incidence of HCC in relation to predictive factors. a The cumulative incidences of HCC in patients with ALT concentrations >40 IU/l at 6 months after commencement of IFN therapy were 0.056, 0.264 and 0.501 at 3, 5 and 7 years, respectively. On the other hand, the cumulative incidences of HCC in patients with ALT concentrations ≤40 IU/l at 6 months after commencement of IFN therapy were 0, 0.05 and 0.05 at 3, 5 and 7 years, respectively. The log-rank test gave a value of p = 0.0007, indicating that the incidence of HCC was significantly higher in the group with ALT concentrations >40 IU/l at 6 months after commencement of IFN therapy. b The cumulative incidences of HCC in patients with a BMI >25 were 0, 0.524 and 0.524 at 3, 5 and 7 years, respectively. On the other hand, the cumulative incidences of HCC in patients with a BMI ≤25 were 0.022, 0.052 and 0.187 at 3, 5 and 7 years, respectively. The log-rank test gave a value of p = 0.02, indicating that the incidence of HCC was significantly higher in the group with a BMI >25. c The cumulative incidences of HCC in IL28B SNP (TT) were 0.101, 0.101 and 0.157 at 3, 5 and 7 years, respectively. On the other hand, the cumulative incidences of HCC in patients with the IL28B SNP (TG or GG) were 0.067, 0.17 and 0.17 at 3, 5 and 7 years, respectively. The log-rank test gave a value of p = 0.91, indicating that the incidence of HCC did not differ significantly between TT and TG or G.|
On the contrary, the genetic variant of the IL28b gene, which was known as a predictive factor for effectiveness of IFN therapy, did not affect the incidence of HCC during follow-up periods of maintenance IFN therapy (the cumulative incidences of HCC in patients with the IL28B SNP (TT) were 0.101, 0.101 and 0.157 at 3, 5 and 7 years, and those with the IL28B SNP (TG or GG) were 0.067, 0.17 and 0.17 at 3, 5 and 7 years, respectively (p = 0.91; fig. 2c). Similarly, the HCV-RNA level at 6 months after commencement of IFN therapy did not show any effects on emergence of HCC.
Post-Treatment ALT Reduction Rates, Virological Response Rates and Rates of Occurrence of HCC in Relation to the IL28B SNP (table 4)
The median ALT reduction rate (= baseline ALT – ALT after IFN/baseline ALT) after treatment in cases with IL28B (TT) was 0.395 (range –1.00 to 0.925). On the other hand, in cases with IL28B (TG or GG), the median ALT reduction rate was 0.208 (range –0.405 to 0.813) and there was no significant difference from the TT cases (p = 0.15).
|Table 4. Relationship of IL28B SNP with reduction in ALT after IFN therapy, virological response, and occurrence of HCC|
The virological response rate was 0.50 in cases with IL28B (TT) and 0.20 in cases with IL28B (TG or GG). Although there was no significant difference (p = 0.07), the rate tended to be lower in cases with IL28B (TG, GG).
The rate of HCC occurrence was 0.15 in cases with IL28B (TT) and comparable with 0.133 in cases with IL28B (TG or GG) (p = 1).
Thanks to the recent advancement of treatment options including IFN, CHC is becoming a curative disease, at least in the majority of cases. However, there are still refractory cases in terms of elimination of HCV, such as the cases with advanced liver fibrosis, elderly cases, and with minor allele of the IL28B gene. These cases should be at higher risk of HCC because of a contentious inflammation due to ongoing HCV infection. From this point of view, reduction of HCC emergence is a top priority for these refractory patients. In this study, we intensively analyze the effect of long-term IFN monotherapy on refractory CHC cases in the context of emergence of HCC. Here, we clearly showed that IFN monotherapy suppressed the emergence of HCC, especially in cases with normal BMI and showing a decrease of ALT at the initial phase of IFN therapy regardless of the genetic background of IL28B.
Several findings of this study have direct implications for long-term PEG-IFNα-2a treatment of patients with chronic hepatitis or cirrhosis. In the present cohort, the cumulative rate of development of HCC in patients whose serum concentrations of ALT were within normal limits (≤40 IU/l) after initiation of PEG-IFNα-2a therapy was lower than that of patients with high concentrations of ALT of >40 IU/l (fig. 2a). This suggests that ALT could be a valuable indicator for predicting effectiveness of long-term PEG-IFNα-2a therapy in terms of HCC emergence. In other words, maintenance of low-dose IFN cannot be recommended if a sufficient decrease of ALT is not achieved within 6 months after the initiation of IFN. This is of importance for both clinical and financial viewpoints because the ineffective use of IFN should be avoided properly.
Previous reports have shown that a biological response suppresses the occurrence of HCC. Imai et al.  compared the rates of HCC occurrence in 419 cases with IFN treatment and 144 control cases and reported that the incidence of HCC decreased even among cases showing a biological response (p = 0.04) after an average follow-up period of 4 years. Okanoue et al.  also compared 1,370 cases treated with IFN and 54 untreated cases and reported that the rate of emergence of HCC was lower in biological response cases than in NR cases. On the other hand, long-term, low-dose therapy reportedly led to a decrease in ALT levels as well . Therefore, it could be speculated that continuous inflammation, necrosis and regeneration of hepatocytes induced genetic injury in hepatocytes and clonal expansion of cells carrying a mutation in cancer-related genes. From this point of view, long-term low-dose IFN might control liver inflammation and suppress the occurrence of HCC.
In this study, BMI >25 was also extracted as a responsible factor for the occurrence of HCC. In all HCC cases with BMI >25, serum HCV-RNA decreased to an undetectable level. Under the condition of undetectable serum HCV-RNA, obesity rather than chronic HCV infection seemed to accelerate the development of HCC. Recently, a relationship between obesity and the development of HCC has been reported in a number of studies. According to the large-scale prospective study, the relative risk of development of HCC in obese cases (BMI >35) was 4.52 in males and 1.6 in females . Muto et al.  investigated the risk of HCC occurrence in 622 cases with decompensated LC and concluded that obesity was associated with the development of HCC. It is well known that obesity is accompanied by insulin resistance. When adipose tissue is enlarged due to obesity, adiponectin secretion is reduced and secretion of adipokain, such as TNF-α and IL-6, increased, inducing insulin resistance [19,20,21]. Because insulin is known as a growth factor for hepatocytes, hyperinsulinemia due to insulin resistance could be an important process for hepatocarcinogenesis [22,23]. The results of this study suggested that, even in cases that achieved viral eradiation through IFN therapy, the risk of HCC was high in obese patients.
On the other hand, the IL28B SNP is a potent predictive factor for treatment response in combination therapy of PEG-IFN and RBV [12,24,25]. As the clinical response for elimination of HCV depends on the IL28B SNP, we speculated that this genetic variant might also influence serum ALT levels after treatment. Therefore, the relationships between the status of IL28B SNP and several serological markers, such as ALT, were also investigated in this study (table 4). However, the rate of reduction in ALT concentrations was 0.395 in TT cases and 0.208 in TG or GG cases, which yielded no significant difference (p = 0.15). The low-dose IFN therapy also showed no significant effect on the virological response rate, which was defined as an undetectable level of HCV RNA at 6 months after commencement of IFN therapy (0.50 in TT cases and 0.15 in TG and GG cases, respectively, p = 0.07). Similarly, the incidence of HCC during follow-up period did not show any difference (0.15 in TT cases and 0.133 in TG or GG cases, respectively). According to the data presented here, it might be attractive to speculate that a decrease of ALT is important to suppress HCC carcinogenesis regardless of SNP of IL28B, as both patients carrying TT allele and TG/GG allele show a similar degree of reduction of ALT and also a similar incidence of HCC.
In this study, we clearly demonstrated that the ALT level ≤40 IU/l and BMI <25 at 6 months after commencement of IFN therapy were extracted as favorite factors for HCC emergence in cases receiving long-term low-dose PEG-IFNα-2a therapy, regardless of SNP in IL28B. Therefore, cases with a decrease of ALT at the initial phase of IFN and normal BMI should be good candidates for an aggressive treatment of receiving low-dose PEG-IFNα-2a. On the other hand, if a decrease of ALT could not be achieved, especially in obese patients, other treatments, such as administration of SNMC and/or UDCA, should be considered. These are the unique and important points of this study because we can predict the emergence of HCC simply using ALT and BMI. Additional prospective studies with larger numbers of patients are required to validate the significance of these findings.
The authors have no conflicts of interest to disclose.
Masatoshi Kudo, MD, PhD
Department of Gastroenterology and Hepatology
Kinki University School of Medicine, 377-2 Ohno-Higashi
Osaka-Sayama, Osaka 589-8511 (Japan)
Published online: December 13, 2012
Number of Print Pages : 7
Number of Figures : 2, Number of Tables : 4, Number of References : 25
Digestive Diseases (Clinical Reviews)
Vol. 30, No. 6, Year 2012 (Cover Date: December 2012)
Journal Editor: Malfertheiner P. (Magdeburg)
ISSN: 0257-2753 (Print), eISSN: 1421-9875 (Online)
For additional information: http://www.karger.com/DDI