Systematic Review: Impact of Interferon-based Therapy on HCV-related Hepatocellular Carcinoma

Hepatitis C virus (HCV) is the leading cause of hepatocellular carcinoma (HCC), and several antiviral agents are available for the treatment of chronic HCV infection. However, the impact of antiviral therapy on the long-term outcomes of HCV-related HCC patients remains inconclusive. We aimed to examine the impact of antiviral therapy on the long-term outcomes of HCV-related HCC patients. We conducted a systematic review using PRISMA guidelines to identify trials and English-language literature from PubMed, Ovid MEDLINE, Scopus and the Cochrane Library database till August 2014. Randomized trials of antiviral treatments examining the effects of antiviral therapy on CHC patients and HCV-related HCC patients were screened and selected. We identified 6 trials evaluated the effectiveness of interferon (IFN)-alfa treatment, 3 studies examined pegylated interferon-alfa treatment, and 2 studies examined IFN-beta treatment. IFN-based therapy may decrease HCC incidence in HCV cirrhotic patients after a >5-year follow-up, improve liver reserve, decrease HCC recurrence rate, and increase survival rate in HCV-related HCC patients after curative HCC therapy. In conclusion, IFN-based therapy is beneficial and may be recommended in the management of HCV-related HCC patients who are IFN eligible.

Hepatitis C virus (HCV) infection is the leading cause of chronic liver diseases, leading to the progression of liver disease from chronic hepatitis to cirrhosis, with the complications of liver failure and hepatocellular carcinoma (HCC) 1,2 . Therefore, eradication of HCV viremia has been recognized as a reasonable way to delay or halt disease progression, hepatic complications and HCC of chronic hepatitis C CHC) patients. Before 2011, interferon (IFN)-based therapy, including dual therapy with pegylated interferon (PEG-IFN) plus ribavirin, was the standard of care of HCV patients in most part of the world. Since 2011, several effective direct antiviral agents (DAAs) have been proven by the U.S. Food and Drug Administration for the treatment of CHC. Although previous studies examined the effectiveness of anti-HCV therapy on disease progression, long-term complications and HCC risk of CHC patients, most of them included non-HCC, highly selected subjects and did not adequately control confounders 3 . Moreover, even fewer studies focused on PEG-IFN based treatment 4,5 , and the impact of DAAs on the long-term complications and HCC risk of CHC patients remain largely unknown 6 .
Because understanding the impact of antiviral regimens on HCV-related HCC is essential for the treatment plan designing of CHC patients with or without HCC, several trials have examined the impact of antiviral agents on HCV-related HCC patients. However, most of these trials differ significantly in the study population, protocols, antiviral regimens, and definitions of outcomes [7][8][9][10][11][12] , whether antiviral regimens will be beneficial for HCV-related HCC patients are inconclusive and deserve further studies.
To this end, we conducted this systematic review to examine the impact of antiviral therapy on the long-term outcomes of HCV-related HCC patients. As the effect of antiviral therapy on the HCC risk of CHC patients remain unconcluded, and antiviral therapy is not likely to affect the long-term outcomes of HCV-related HCC patients when it does not decrease the HCC risk in CHC patients. We investigated the impact of antiviral regimens on the HCC incidence of patients with HCV infection first. Once positive findings were shown in the first part, we then examined the long-term outcomes of HCV-related HCC patients by searching relative references and examining the benefits as well as harms of antiviral agents for HCV-related HCC patients.

Scope.
A review protocol and an analytic framework were developed with input and discussion from the authors to set clear questions, ensure a standardized and repeatable review process between reviewers (Supplementary Figure and Supplementary Protocol), and included the following key questions: 1. What is the effect of antiviral treatment on the HCC incidence of patients with chronic HCV infection? 2. What is the effect of antiviral treatment on HCC outcomes of patients with HCV-related HCC (including the recurrence or progression of HCC)? 3. What is the effect of antiviral treatment on clinical outcomes (including the survival and longterm complications) of patients with HCV-related HCC? 4. What are the comparative harms associated with antiviral treatments in patients with HCV-related HCC?
Data Sources and Searches. We followed the PRISMA recommendations for systematic literature analysis, which is an evidence-based set of items for reporting in systematic reviews and meta-analyses, and focuses on ways to ensure the transparent and complete reporting 13 . We performed computer-based searches of the PubMed, Ovid MEDLINE, Scopus, and the Cochrane Library database for clinical trials in original articles between 1947 and August 2014. We searched using combinations of the following terms: "interferon" or "antiviral therapy" or "pegylated interferon" or "ribavirin" or "DAA" or "hepatitis C" or "chronic hepatitis C" or "HCV" and "hepatocellular carcinoma" or "HCC" in the title or abstract fields. We screened each abstract resulting from these searches for eligibility. We also examined reference lists of each selected original article that might meet our eligibility requirements.

Study Selection.
Two reviewers independently evaluated studies for inclusion. We included randomized trials on HCV-related HCC patients that investigated IFN-based therapy, including mono-therapy with IFN or PEG-IFN alfa; dual therapy with IFN plus ribavirin or PEG-IFN alfa plus ribavirin; and triple therapy with PEG-IFN (alfa−2a or −2b), ribavirin, and either telaprevir or boceprevir or newer DAAs. Clinical outcomes were mortality, cirrhosis, hepatic decompensation, hepatocellular carcinoma, and need for transplantation. HCC outcomes were the recurrence and progression of HCC. Sustained virological response (SVR) was defined as the absence of detectable HCV RNA in the serum 6 months after the end of a course of therapy. Biological response (BR) was defined as a decrease in serum ALT to within the normal range after treatment. Harms included withdrawals due to adverse events, serious adverse events, neutropenia, anemia, psychological adverse events, influenza-like symptoms, and rash. We restricted the inclusion to English-language articles. Article selection was determined a priori using the following inclusion criteria: (i) only adults were included in the trial; (ii) patients with chronic HCV infection were included in the study examining effects of antiviral treatments on HCC incidence; and patients with HCV-related HCC were included in the study examining effects of antiviral treatments on HCC outcomes; (iii) the study used IFN-or PEG-IFN-based therapy combined with or without other antiviral agents. We excluded: (i) review articles, (ii) editorials; (iii) letters to the editor (iv) case reports; (v) research protocols; (vi) duplicated records; (vii) sub-analyses of the same clinical trial; if the final study results had been published; (ix) interim reports if the final results had been published; (x) regimens with antiviral drugs not approved by the U.S. Food and Drug Administration for HCV infection. Data Abstraction. One investigator abstracted details of the study, and a second investigator reviewed data for accuracy. The methodological quality of each study was assessed using the Cochrane Collaboration's tool for assessing risk of bias in randomized trials (Table 1) 14 . Two investigators independently reviewed all studies and assigned a value of "low risk, " "high risk, " or "unclear" to the following: a) random sequence generation, b) allocation concealment, c) blinding of participants and personnel, d) adequate assessment of each outcome, e) selective outcome reporting avoided, and f) Other bias (reports for adverse effects of intervention). Discrepancies were resolved through consensus. The following data were extracted from each report: publication type (original article or conference abstract); number of patients; antiviral regimens; status of liver cirrhosis or liver fibrosis stage; percentage of patients with Scientific RepoRts | 5:09954 | DOi: 10.1038/srep09954 hepatitis B or hepatitis C; period of follow-up; incidence of HCC; recurrence rate of HCC; overall survival (OS); and adverse effects of antiviral agents.

Results
The flow chart of the systematic review is shown in Fig. 1. Our initial search yielded 3835 articles. Ultimately, 11 randomized controlled trials (RCTs) were chosen for review, including 5 investigated antiviral treatments and HCC incidence in CHC patients, and 6 examined antiviral treatments and HCC outcomes in HCV-related HCC patients who had received curative therapy for HCC. Six trials evaluated effectiveness of IFN-alfa based treatment, 3 examined the effect of pegylated interferon-alfa, and 2 studies examined the effect of IFN-beta treatment. No RCT ever investigated the effects of antiviral treatments on HCV-related HCC patients without curative HCC therapy, and none evaluated the effectiveness of IFN plus ribavirin or DAAs.
Four studies reported the improvement of serum alanine aminotransferase (ALT) levels, and decreased serum HCV RNA level in treated subjects. A biochemical end-of-therapy response was observed in 5/38 (13%) IFN-beta treated patients [2/23 (9%) in controls], and 6/47 (12.8%) IFN-alfa treated patients [0/39 in controls] 16,17 . Although Nishiguchi et al. did not identify a significant difference in the normalization of ALT level between groups, they found that ALT deceased to <80 IU/L in 19/45(42.2%) IFN-alfa treated patients [9/45 (20%) in controls]. Moreover, they also found lower AFP and higher serum albumin levels in IFN-treated patients. For PEG-IFN-alfa 2a treatment, HALT-C study showed a 0.45-fold decline of serum ALT levels among treated patients at 1.5 years (0.21 times among control patients), a 0.47-fold decline among treated patients at 3.5 years (0.19 times among control patients), and a significant difference of 0.28 times the upper limit of the normal range (95% CI, 0.12 to 0.44; P < 0.001).
Two studies (Nishiguchi et al. & HALT-C) 4,18 reported the improvement of liver histology in treated subjects, and 1 study (EPIC3) 5 Table 1. Quality of assessment of each included studies. NOTE. "Low" is "low risk of bias", that means if bias is present, is unlikely to alter the results. "Unclear" is "unclear risk of bias", that means a risk of bias that raises some doubt about the results. "High risk" is "high risk of bias", that means bias may alter the results seriously.
A decreased incidence of HCC was observed in 2 trials with a follow-up period more than 5 years (Nishiguchi et al. & HALT-C) 15,18 , but not in trials with a follow-up less than 5 years 5,15-17 . Therefore, IFN-treated CHC patients had a better liver function tests, liver histology, and slower liver disease progression than non-IFN-treated controls. Moreover, IFN-treated CHC cirrhotic patients had a lower HCC incidence than non-IFN-treated controls after a >5-year of follow-up.
The impact of antiviral treatment on the clinical outcomes of patients with HCV-related HCC. Six RCTs were identified to examine the impact of antiviral agents for patients with HCV-related HCC after curative HCC therapy. A total of 374 HCV-related HCC patients, including 201 IFN-treated patients and 173 controls, are summarized in Table 3 [7][8][9][10][11][12] . All trials included HCV-related HCC patients who received curative HCC therapy, including surgical resection in 3 8,10,11 , percutaneous ethanol injection(PEI) or resection in 1 9 , PEI in 1 7 and transcatheter arterial chemoembolization(TACE) plus PEI in 1 12 . One study used IFN-beta 9 , 5 used IFN-alfa 7,8,10,12 , and one used PEG-IFN alfa−2b plus ribavirin 11 for the treatment of HCV-related HCC patients after curative HCC treatment.
Four studies reported SVR rates in IFN-treated patients 7,8,10,12 , and improved liver function tests in IFN-treated HCV-related HCC patients 7,8,10,11 . Patients receiving IFN-alfa treatment had SVR rate of 7-50% and sustained biological response (SBR) rate of 7-43% 7,8,10,12 . Although IFN-beta did not affect the SVR and SBR rates in HCV-related HCC patients, serum transaminases tended to decrease at the 12th month in IFN-beta treated group 9 . Moreover, increased serum albumin level was observed in HCV-related HCC patients receiving PEG-IFN alfa-2b plus ribavirin therapy, especially in SVR patients 11 . Six studies reported the overall survival difference between groups, including 4 reported a better survival rate in IFN-treated group (P < 0.05) 7,8,11,12 , one reported a marginal higher survival rate in the IFN-treated group (IFN vs. control: 24.3% vs. 5.8%; P = 0.49) 10 , and one reported no mortality until the end of the observation period 9 . In the study of Shiratori et al., patients receiving IFN-alfa treatment had survival rates of 98%, 82%, 68%, and 53% at 1, 3, 5, and 7 years, respectively; untreated patients had survival rates of 96%, 84%, 48%, and 23% at 1, 3, 5, and 7 years, respectively. In addition, survival rates in SVR patients were 86%, 78%, and 68% at 3, 5, and 7 years, respectively; whereas survival rates in non-SVR patients were 80%, 65%, and 47% at 3, 5, and 7 years, respectively 7 . In patients receiving PEG-IFN-alfa−2b plus ribavirin treatment 11 cumulative survival rates among HCV-related HCC patients were 100% after 1 year and 90.2% after 3 years in the IFN-treated subjects compared to 96.0% and 61.2% in the non-treated controls 11 . Therefore, IFN-treated patients who had received curative HCC therapy, had a better biochemical response, virological response, liver function reserve and survival rate than non-IFN-treated controls. The impact of antiviral treatment on the recurrence of HCC after patients receiving curative HCC therapy. Six studies reported the recurrence rate of HCC in HCV-related HCC patients after curative HCC therapy [7][8][9][10][11][12] (Table 3). Five studies reported a lower recurrence rate of HCC in  HCV-related HCC patients receiving IFN-based treatment [7][8][9][10]12,19,20 , while the study investigating the impact of PEG-IFN-alfa-2b plus ribavirin treatment did not identify a significant difference in cumulative first recurrence rates of HCC 11 . Because tumor recurrence after resection of HCC can occur early (<2 years) or late (≧2 years) as metastases or de novo tumors, four studies also examined the recurrence rates in HCV-related HCC patients in the second year or later, and found that IFN-based treatment was associated with a lower late recurrence rate 7,9,10,12 . Therefore, IFN-treated patients who had received curative HCC therapy had a lower recurrence rate, especially a lower late recurrence rate of HCC than non-IFN-treated controls.
Adverse events associated with antiviral treatments in patients with HCV-related HCC. There was no available data for a direct comparison between HCV infected patients and HCV-related HCC patients in the adverse events after antiviral therapy. However, five studies investigating IFN-based therapy reported the associated adverse events, and most were expected symptoms such as fever, chills, myalgia, headaches fatigue, anorexia, weight loss, irritability, concentration, difficulty, sleep disturbance, depression, severe malaise, emotional liability, thrombocytopenia, leukocytopenia, hyperthyroidism, and deterioration in Child's score.

Discussion
Antiviral regimens have been documented to result in decreasing risks of HCC and liver cirrhosis complication in CHC patients 3 . Although understanding the impact of antiviral agents on HCV-related HCC patients is critical for making treatment decision in such patients, whether antiviral regimens are beneficial for HCV-related HCC remain inconclusive. In this systematic review, we concluded that IFN-based treatment may decrease HCC incidence in HCV-related cirrhotic patients. Moreover, IFN-based therapy may improve liver reserve, significantly increase the long-term survival and decrease the HCC recurrence rate of HCV-related HCC patients after curative HCC therapy. Therefore, IFN-based therapy is beneficial, and may be recommended in the management of HCV-related HCC patients who are IFN eligible.
In lieu of direct evidence on long-term clinical outcomes of antiviral agents, SVR rate is the primary outcome measurement for the effectiveness of antiviral regimens 21 . However, most of the selected trials investigated the effects of IFN-alfa-or beta-based therapy that are known to have a lower efficacy than PEG-IFN or DAAs-based regimens 22 . Therefore, studies investigating IFN regimens usually have a modest virological and biochemical response after treatment, and a longer observational period is needed for significant differences been detected in the target population. That is why a significant decline of HCC incidence in CHC patients receiving IFN treatment was observed in studies with a period of follow-up more than 5 years 15,18 , but not in ones with a shorter period of follow-up. Interestingly, even the differences in terms of virological response and improvement in liver reserve are modest (Table 3), IFN-based therapy may significantly reduce HCC recurrence rate in HCV-related HCC patients after curative HCC therapy. This may be attributed to the additional inhibitory effects of IFN on liver carcinogenesis 23,24 .
Compared to untreated controls, 8 studies showed a biochemical response with serum ALT decline, and 2 showed a better liver reserve in CHC patients after receiving IFN-based therapy 11,18 . Although 2 studies showed a potential benefit of IFN-based treatments for improving liver histology (Nishiguchi et al. & HALT-C) 15,18 , and 1 study (EPIC3) demonstrated a decreasing incidence of liver cirrhosis complications, including ascites, encephalopathy, and varices bleeding, in CHC patients receiving IFN-based treatments 5 . There is no study investigating the effects of antiviral regimens on liver histology and complications of liver cirrhosis in HCV-related HCC patients. Moreover, although two large RCTs (HALT-C & EPIC3) 4,5 have examined the impact of PEG-IFN treatment on CHC cirrhotic patients, most of the enrolled patients failed initial antiviral treatments and received half-dose regimens, the long-term outcomes after PEG-IFN-based therapy in treatment-naive HCV cirrhotic patients remain unclear. Therefore, future studies examining the impact of PEG-IFN-based regimens on long-term outcomes in HCV-related HCC patients and treatment-naive HCV cirrhotic patients are needed. In addition, this review is limited by the lack of information regarding DAA regimens. As DAAs have become the backbones of standard of care for HCV infection, their impact on HCC risk and long-term outcomes in CHC patients with and without HCC warrants future studies.
Our study had a few limitations. First, a positive "publication" bias that tends to publish research with a positive outcome more frequently may be introduced into this study 25 . Moreover, because some inclusion criteria were used to select studies, a few studies might be excluded and missed during the literature search. Second, there exists substantially clinical heterogeneity in the selected randomized controlled trials. For examples, kinds of IFN, duration and dose of antiviral agents, and race that may be different in the prevalence of IL28B major allele, and all of these may hamper the validity of the pooled estimate and affect the ability to perform a metaanalysis 26 . Therefore, we performed descriptive evaluations for each study rather than combined the data together. However, descriptive evaluations for each study have shown consistent results among studies. Third, the time-to-event outcomes are likely affected by censoring and the duration of follow-up that are different among individual trials 27 , and the unavailability of individual data may limit the analysis of the clinical outcomes in patients with HCV-related HCC. Last, based on the sub-group analyses of patients with compensated cirrhosis in several large clinical trials [28][29][30]   of Liver Diseases and the Infectious Diseases Society of America does not recommend mono or dual therapy with IFN for the treatment patients with HCV-related cirrhosis, and has recommended the use of highly potent DAA oral combination regimens for HCV infected patients. However, the long-term effects of the use of DAAs, including its effect on HCC, remain unclear and need be further examined.
Considering the additional anti-tumorigenic effect of IFN, IFN-based therapy might be recommended in HCV-related HCC patients who are IFN eligible. However, future comparative studies are needed to confirm this speculation.

Conclusions
This systematic review supports the beneficial effects of IFN-based treatment for decreasing HCC incidence in HCV-related cirrhotic patients, and HCC recurrence rate in HCV-related HCC patients after curative HCC therapy. Moreover, IFN-based therapy is associated with the improvement of serum ALT levels and better liver reserve in HCV-related HCC patients after curative HCC therapy. However, future studies are required to clarify the impact of DAAs on HCC risk and cirrhotic complication in CHC patients and HCV-related HCC patients after curative HCC therapy.