| |
Study Finds Interferon Reduces Cancer Risk in HCV+ Cirrhotics
|
| |
| |
"Antiviral Therapy for Cirrhotic Hepatitis C: Association with Reduced Hepatocellular Carcinoma Development and Improved Survival"
Annals of Internal Medicine, Jan 18, 2005
Yasushi Shiratori, MD; Yoichi Ito, MHlth Sc; Osamu Yokosuka, MD; Fumio Imazeki, MD; Ryo Nakata, MD; Naohide Tanaka, MD; Yasuyuki Arakawa, MD; Etsuko Hashimoto, MD; Katsutaro Hirota, MD; Haruhiko Yoshida, MD; Yasuo Ohashi, PhD; and Masao Omata, MD, for the Tokyo-Chiba Hepatitis Research Group*
"...We conclude that antiviral treatment improved the natural course in cirrhotic patients with HCV infection, especially cirrhotic patients with viral eradication. Because the efficacy of antiviral therapy for patients with chronic hepatitis C, even cirrhotic patients, reportedly increases with combined interferon--ribavirin or pegylated interferon--ribavirin therapies, combination therapy for cirrhotic patients may lead to a favorable outcome. Trials studying combination therapy in a large number of cirrhotic patients are being conducted in Japan. The survival advantage of interferon treatment may not be as great in western cohorts because the incidence of hepatocellular carcinoma may not be as high in such groups as in Japan. However, with higher rates of sustained virologic response obtained through use of pegylated interferon, the survival advantage may be similar...
"...Despite the contradictory findings in previous studies and the built-in biases in this study, our current study clearly showed that the incidence of hepatocellular carcinoma development was significantly reduced in the interferon-treated group compared with the untreated group, even after adjustment for age...
"...In this study, the annual incidence of hepatocellular carcinoma in untreated patients was 6% to 8%; in interferon-treated patients, the incidence was 4% to 5%. Analysis of interferon response indicated that the annual incidence of tumor development in sustained responders was markedly decreased to 2.5% (3% to 4% in the initial 2 years and 2% thereafter), in contrast to the 5% in nonsustained responders (4% in the initial 2 to 3 years and 5% to 8% thereafter). Furthermore, normalization of serum ALT level (biochemical response) in nonsustained responders did not contribute to the reduction in development of hepatocellular carcinoma (data not shown)...
... In addition to hepatocellular carcinoma development, we also analyzed mortality rates among patients with chronic hepatitis C and cirrhosis. The survival results turned out to be more important than the incident results because it is theoretically possible that tumors that arise after interferon treatment have a different prognosis or treatment response than those that arise without viral suppression. In the end, only the survival advantage justified the use of preventive treatment. Of the overall deaths in the interferon-treated and untreated groups, liver-related death (hepatocellular carcinoma and liver failure) occurred in 71% and 79%, respectively. Liver failure caused 40% of these deaths in the untreated group and 20% of those in the interferon-treated group.
...According to an age-adjusted analysis, survival among interferon-treated patients was significantly better than that among untreated patients. Furthermore, sustained negativity of HCV RNA after interferon therapy was associated with a reduced risk for death. Treatment tactics for hepatocellular carcinoma in the interferon-treated and untreated groups were similar, and the groups had a similar survival rate after the development of hepatocellular carcinoma (P = 0.15; data not shown). In the ad hoc analysis, the untreated group and nonresponders had even greater similarity (P > 0.2; data not shown)..."
Chronic hepatitis C is a common disease that progresses slowly to cirrhosis and eventually may lead to hepatocellular carcinoma. The annual incidence of hepatocellular carcinoma and mortality rate were 1.4% to 3.3% and 1.9% to 5.5%, respectively, in retrospective series of white patients with hepatitis C virus (HCV)--related compensated cirrhosis; in Japan, the annual incidence of hepatocellular carcinoma was 5% to 7%. Risk factors for hepatocellular carcinoma include age older than 50 to 60 years, male sex, advanced fibrosis stage, high histologic activity score, and high alanine aminotransferase (ALT) levels.
Several retrospective studies have shown inhibition of hepatocellular carcinoma development after interferon therapy. This inhibitory effect was seen in patients with moderate fibrosis for whom antiviral therapy was effective. However, the inhibitory effect in patients with liver cirrhosis was not statistically significant, possibly because of the low efficacy of interferon therapy in cirrhotic patients. Other retrospective and prospective studies that had small patient samples indicated that interferon therapy reduced the development of hepatocellular carcinoma.
Because cirrhosis is a major risk factor for hepatocellular carcinoma, a prospective study is needed to determine whether interferon therapy benefits cirrhotic patients. We previously performed 2 prospective studies on the efficacy of interferon treatment in cirrhotic patients. During enrollment, many cirrhotic patients who fulfilled the inclusion criteria were enrolled as controls to clarify the long-term effect of interferon therapy on development of liver tumors. We conducted a 7-year study on the inhibition of hepatocellular carcinoma development in the previous cohorts of our multicenter, prospective study.
ABSTRACT
Background: Although cirrhosis is a major risk factor for development of hepatocellular carcinoma, no definitive prospective analyses have assessed the long-term efficacy of antiviral therapy in cirrhotic patients.
Objective: To elucidate the role of antiviral therapy in the suppression of liver tumors and survival over a long-term follow-up period.
Design: Prospective cohort study.
Setting: 25 clinical centers.
Patients: 345 patients with chronic hepatitis C and cirrhosis enrolled in previous trials.
Intervention: 271 patients received 6 to 9 million U of interferon 3 times weekly for 26 to 88 weeks; 74 received no treatment.
A total of 157 patients received 9 million units (MU) of interferon-{alpha}2a (Nippon Roche KK, Tokyo, Japan) by intramuscular injection 3 times a week for 32 to 88 weeks.
Patients negative for HCV RNA more than 24 weeks after the completion of interferon therapy were considered to show a sustained virologic response, while patients positive for HCV RNA more than 24 weeks after the completion of interferon therapy were considered to show a nonsustained response.
Measurements: Blood tests and abdominal ultrasonography were done regularly to detect hepatocellular carcinoma.
Results:
Hepatocellular carcinoma was detected in 119 patients during 6.8 year follow-up: 84 (31.0%) of the 271 interferon-treated patients and 35 (47.3%) of the 74 untreated patients. Of the 64 patients with a sustained virologic response and 207 with a nonsustained response, hepatocellular carcinoma developed in 11 (17.2%) and 73 (35.3%) patients, respectively.
Cumulative incidence of hepatocellular carcinoma among interferon-treated patients was significantly lower than in untreated patients (Cox model: age-adjusted hazard ratio, 0.65 [95% CI, 0.43 to 0.97]; P = 0.03), especially sustained virologic responders. A total of 69 patients died during follow-up: 45 (17%) in the treated group and 24 (32%) in the untreated group. Interferon-treated patients had a better chance of survival than the untreated group (Cox model: age-adjusted hazard ratio, 0.54 [CI, 0.33 to 0.89]; P = 0.02). This was especially evident in sustained virologic responders.
Limitation: This was not a randomized, controlled study. Patients enrolled in the control group had declined to receive interferon treatment even though they were eligible for treatment.
Conclusion: Interferon therapy for cirrhotic patients with chronic hepatitis C, especially those in whom the infection had been cured, inhibited the development of hepatocellular carcinoma and improved survival.
Development of Hepatocellular Carcinoma
Hepatocellular carcinoma was detected in 119 patients during follow-up: 84 (31.0%) of the 271 interferon-treated patients and 35 (47.3%) of the 74 untreated patients. Of the 64 patients with a sustained virologic response and 207 with a nonsustained response, hepatocellular carcinoma developed in 11 (17.2%) and 73 (35.3%) patients, respectively. The median tumor nodule diameter was 2 cm (range, 1 to 10 cm). The median number of tumors was 1 (range, 1 to 6); 75 patients (62%) had 1 nodule, and 44 patients had 2 or more nodules. The size and number of tumors did not differ between the interferon-treated and untreated groups. All patients with hepatocellular carcinoma had dynamic CT or MRI and arteriography, as well as abdominal ultrasonography. One hundred thirteen patients (95%) showed a typical pattern of hyperattenuation in the early phase and hypoattenuation in the late phase on dynamic CT or MRI. Tumor blush was confirmed in these patients. Six patients were found to have atypical features (hypoattenuation in the early phase and late phase) on dynamic CT. Hepatocellular carcinoma was histologically confirmed for 83 patients (70%), including the 6 patients who had atypical features on dynamic CT. Hepatocellular carcinoma in the remaining 36 patients was diagnosed by the typical pattern on dynamic CT or MRI, and these tumors were characterized by larger size (median diameter, 3 cm), multiple tumor nodules (>=3 nodules in 12 patients), and high serum {alpha}-fetoprotein (AFP) levels (>=100 ng/mL in 13 patients).
Although we had not initially determined how often serum AFP levels would be measured during follow-up, serum AFP was measured in 79 patients at the time of hepatocellular carcinoma diagnosis; 21 patients had an AFP level of at least 100 ng/mL, 15 had a level of 50 to 99 ng/mL, and 43 had a level less than 49 ng/mL.
We analyzed the cumulative incidence of tumor development by using the Kaplan--Meier method. The cumulative incidence of hepatocellular carcinoma was lower in interferon-treated patients than in untreated patients, and the hazard ratio for treatment was similar between the models stratified by 3 age groups and the numeric age-adjusted model (age-adjusted hazard ratio, 0.65 [95% CI, 0.43 to 0.97]); P = 0.03).
Of the 119 patients with hepatocellular carcinoma, 7 had surgical hepatectomy, 103 had local ablation therapy using percutaneous ethanol injection or radiofrequency, and 9 had transarterial chemoembolization. No patients underwent liver transplantation. Of the 84 interferon-treated patients, surgical hepatectomy was performed in 6 (7.1%), local ablation therapy in 72 (85.7%), and transarterial chemoembolization in 6 (7.1%). Of the 35 control patients, 1 (2.9%) had hepatectomy, 31 (88.6) had local ablation, and 3 (8.6%) had a transarterial procedure. Treatment methods did not differ between the interferon-treated and untreated groups (P > 0.2).
We used a Cox proportional hazards model to analyze the predictors of hepatocellular carcinoma. In the univariate analysis, age, albumin, platelet count, and interferon treatment were significant factors in the inhibition of hepatocellular carcinoma development; HCV RNA load, HCV genotype, ALT level, the difference of the preceding trial, and total interferon dose were not significant. In a multivariate analysis, age, treatment, and albumin level were significant predictors.
When we stratified interferon-treated patients according to sustained response state, cumulative incidence of hepatocellular carcinoma among sustained responders was lower than in untreated patients (age-adjusted hazard ratio, 0.31 [CI, 0.16 to 0.61]; P < 0.001), although a nonsustained response was not significant compared with untreated controls (age-adjusted hazard ratio, 0.77 [CI, 0.51 to 1.16]; P > 0.2).
Mortality and Survival
Causes of Death
Sixty-nine patients died during the follow-up period: 45 patients (17%) from the interferon-treated group and 24 (32%) from the untreated group. Thirty-two of the interferon-treated patients died of liver-related diseases: 25 of hepatocellular carcinoma, 6 of liver failure, and 1 of varix rupture. Of the 13 remaining interferon-treated patients, 2 died of pancreatic cancer, 4 of cerebral hemorrhage, 3 of pneumonia or pulmonary interstitial fibrosis, 1 of heart disease, 1 of peritonitis, 1 of suicide, and 1 of unknown causes. Of the 24 untreated patients, 19 died of liver-related diseases: 11 of hepatocellular carcinoma and 8 of liver failure. The other 5 patients died of neoplasms in the stomach and biliary tract, cerebral hemorrhage, heart failure, and peritonitis, respectively. Deaths from liver disease (hepatocellular carcinoma, hepatic failure, or varix rupture) occurred in none of 64 sustained responders, 32 (15%) of 207 nonsustained responders, and 19 (26%) of 74 untreated patients (P < 0.001 and P = 0.08, respectively).
Survival
The median duration of follow-up for the survivors was 8.0 years (25th and 75th percentiles, 5.8 and 9.3 years). We analyzed survival by using the Kaplan--Meier method with an age-adjusted analysis. We used a Cox proportional hazards model to analyze predictors. Of the variables tested, only age and treatment were significant predictors in the multivariate analysis. Interferon-treated patients had a better chance of survival than did the untreated patients (Cox model: age-adjusted hazard ratio, 0.54 [CI, 0.33 to 0.89]; P = 0.02). Because the survival curves in the interferon-treated and untreated patients crossed each other, we used a piecewise Cox model in which the hazard ratio changed over time. The age-adjusted hazard ratio for survival in the interferon group was 1.03 (CI, 0.46 to 2.29; P > 0.2) before the 6-year follow-up and 0.32 (CI, 0.16 to 0.62; P = 0.001) after the 6-year follow-up. Furthermore, sustained virologic response was associated with a better chance of survival (age-adjusted hazard ratio, 0.05 [CI, 0.006 to 0.34]; P = 0.003 compared with the untreated group) than was nonsustained virologic response (age-adjusted hazard ratio, 0.71 [CI, 0.43 to 1.18]; P = 0.19 compared with the untreated group).
AUTHOR DISCUSSION
Cirrhosis is a major risk factor for hepatocellular carcinoma and death. Antiviral therapy may be the most beneficial form of treatment for patients with chronic hepatitis C and cirrhosis, although this idea remains controversial. Retrospective cohort studies involving many patients with chronic hepatitis C showed that antiviral therapy reduces the risk for hepatocellular carcinoma in patients who have mild to moderate fibrosis. Nishiguchi and colleagues showed that interferon therapy can reduce the incidence of hepatocellular carcinoma in cirrhotic patients, regardless of interferon response. However, these investigators examined only 90 patients and reported a low rate of response to interferon therapy (sustained response, 16%). Cohort studies by Fattovich and Niederau and their colleagues showed that interferon therapy does not reduce the risk for hepatocellular carcinoma, perhaps because of the lower rate of liver tumor development (1% to 2% annually), low response rate (7% to 12%), and short follow-up period (4 to 5 years). A retrospective study by Serfaty and coworkers showed that interferon treatment induces a favorable outcome for hepatocellular carcinoma development or decompensation; however, their study included only 103 patients. Because of the contradictory findings and various built-in biases of previous studies, we designed a prospective study to clarify the exact role of interferon therapy in reducing the rate of hepatocellular carcinoma over a 7-year observation period.
Our study participants consisted of interferon-treated and untreated patients who were enrolled in our 2 preceding prospective trials. Those trials had been conducted to clarify the efficacy of interferon therapy in patients with chronic hepatitis C and cirrhosis. One limitation we encountered was that while untreated patients fulfilled the inclusion criteria for the trials, interferon-treated patients and untreated patients differed with regard to age and serum ALT levels. This difference may be due to greater anxiety in older patients or patients with mildly elevated ALT levels stemming from reports in the popular press in Japan in the 1990s about the high prevalence of adverse effects associated with interferon. Other than age and serum ALT levels, the interferon-treated and untreated groups were very similar and the untreated group was not sicker. Because only age was favorable to the outcome for interferon-treated patients, we accounted for age in the statistical analyses. However, abnormal ALT, aspartate aminotransferase, and albumin levels did not affect the hazard ratio of interferon use.
In addition to the preceding limitation, ours was not truly a randomized trial because we did not randomly choose the patients in the untreated groups. Instead, these groups consisted of patients who were offered but declined treatment. This situation introduces a possible bias because not only were the patients older and otherwise different from the treated group at baseline, but they may have also differed from the treated patients in other ways. Follow-up rate and treatment tactics for patients with hepatocellular carcinoma did not differ between the 2 groups; thus, we believe that these biases were small.
Despite the contradictory findings in previous studies and the built-in biases in this study, our current study clearly showed that the incidence of hepatocellular carcinoma development was significantly reduced in the interferon-treated group compared with the untreated group, even after adjustment for age. Although the difference in tumor development reduction was larger in patients younger than 50 years of age than in those 50 to 59 years of age and those 60 years of age or older, both stratified and numeric age-adjusted analyses showed a significant reduction in the interferon-treated group. Furthermore, the cumulative incidence of hepatocellular carcinoma was markedly decreased, especially in sustained virologic responders when compared with nonresponders and untreated patients. However, transient virologic response during interferon treatment followed by relapse was not associated with the outcome (data not shown). Although baseline serum ALT levels differed between the 2 groups, this variable was not associated with the inhibition of hepatocellular carcinoma. Furthermore, while the patients enrolled in this study also participated in the 2 preceding trials (which used a different treatment schedule), this participation was not associated with tumor inhibition. In this study, sex, total dose of interferon, viral load at baseline, and HCV genotype were not associated with outcome, although these variables are known to be related to interferon response. The present results may be explained by the fact that 72% of the patients in the natural interferon trial had a low viral load and that most of the enrolled patients in the preceding trials were male.
In this study, the annual incidence of hepatocellular carcinoma in untreated patients was 6% to 8%; in interferon-treated patients, the incidence was 4% to 5%. Analysis of interferon response indicated that the annual incidence of tumor development in sustained responders was markedly decreased to 2.5% (3% to 4% in the initial 2 years and 2% thereafter), in contrast to the 5% in nonsustained responders (4% in the initial 2 to 3 years and 5% to 8% thereafter). Furthermore, normalization of serum ALT level (biochemical response) in nonsustained responders did not contribute to the reduction in development of hepatocellular carcinoma (data not shown).
In addition to hepatocellular carcinoma development, we also analyzed mortality rates among patients with chronic hepatitis C and cirrhosis. The survival results turned out to be more important than the incident results because it is theoretically possible that tumors that arise after interferon treatment have a different prognosis or treatment response than those that arise without viral suppression. In the end, only the survival advantage justified the use of preventive treatment. Of the overall deaths in the interferon-treated and untreated groups, liver-related death (hepatocellular carcinoma and liver failure) occurred in 71% and 79%, respectively. Liver failure caused 40% of these deaths in the untreated group and 20% of those in the interferon-treated group.
According to an age-adjusted analysis, survival among interferon-treated patients was significantly better than that among untreated patients. Furthermore, sustained negativity of HCV RNA after interferon therapy was associated with a reduced risk for death. Treatment tactics for hepatocellular carcinoma in the interferon-treated and untreated groups were similar, and the groups had a similar survival rate after the development of hepatocellular carcinoma (P = 0.15; data not shown). In the ad hoc analysis, the untreated group and nonresponders had even greater similarity (P > 0.2; data not shown).
This study found a large decline in survival at 6 years. This finding may be related to aspects of the protocol (extension of follow-up period); changes in follow-up procedure for the last 10 years, such as increased ease of CT and MRI use; and a change in tumor treatment from percutaneous ethanol injection therapy to radiofrequency ablation. Thus, the survival data should be considered circumspectly. In the ad hoc analysis, the large decline in survival was due not only to hepatocellular carcinoma but also to liver failure that occurred after 6-year follow-up, especially in untreated patients (data not shown). These data coincide with the results of our latest cohort analysis of 2889 patients with chronic hepatitis C. In that study, patients treated with interferon, especially sustained virologic responders, had improved survival.
We conclude that antiviral treatment improved the natural course in cirrhotic patients with HCV infection, especially cirrhotic patients with viral eradication. Because the efficacy of antiviral therapy for patients with chronic hepatitis C, even cirrhotic patients, reportedly increases with combined interferon--ribavirin or pegylated interferon--ribavirin therapies, combination therapy for cirrhotic patients may lead to a favorable outcome. Trials studying combination therapy in a large number of cirrhotic patients are being conducted in Japan. The survival advantage of interferon treatment may not be as great in western cohorts because the incidence of hepatocellular carcinoma may not be as high in such groups as in Japan. However, with higher rates of sustained virologic response obtained through use of pegylated interferon, the survival advantage may be similar.
|
|
| |
| |
|
|
|
