Interferon therapy prolonged life expectancy among chronic hepatitis C patients

GASTROENTEROLOGY 2002;123:483– 491

Interferon Therapy Prolonged Life Expectancy Among Chronic Hepatitis C Patients HARUHIKO YOSHIDA,* YASUYUKI ARAKAWA,‡ MICHIO SATA,§ SHUHEI NISHIGUCHI,㛳 MICHITAMI YANO,¶ SHIGETOSHI FUJIYAMA,# GOTAROU YAMADA,** OSAMU YOKOSUKA,‡‡ YASUSHI SHIRATORI,* and MASAO OMATA* *Department of Gastroenterology, University of Tokyo Graduate School of Medicine, Tokyo; ‡Third Department of Internal Medicine, Nihon University, Tokyo; §Second Department of Internal Medicine, Kurume University, Fukuoka; 㛳Third Department of Internal Medicine, Osaka City University, Osaka; ¶Department of Clinical Research, National Nagasaki Medical Center, Nagasaki; #Third Department of Internal Medicine, Kumamoto University, Kumamoto; **Center of Liver Diseases, Kawasaki Hospital, Kawasaki Medical School, Okayama; and ‡‡First Department of Internal Medicine, Chiba University, Chiba, Japan

Background & Aims: The effects of interferon therapy in chronic hepatitis C patients on survival are unclear. Our objective was to analyze survival among a large cohort of chronic hepatitis C patients. Methods: We used a retrospective cohort study design in the setting of 7 university hospitals and 1 regional core hospital in Japan. Our study included 2889 patients with histologicalproven chronic hepatitis C: 2430 patients received interferon therapy, and 459 patients were untreated. For intervention, the median dose and duration of interferon administration was 480 million units and 137 days, respectively. For measurements, survival status was confirmed by medical records or direct questionnaires. The effect of interferon therapy on survival was assessed by standardized mortality ratio (SMR) based on published mortality among the Japanese general population and by risk ratio calculated by proportional hazards regression. Results: Thirty of 459 untreated patients, 7 of 817 virologic sustained responders, and 49 of 1613 nonresponders died in 5.4-years follow-up. Fiftyeight (67%) of 86 patient deaths were due to liver diseases (39 to hepatocellular carcinoma). Compared with the general population, overall mortality was high among untreated patients (SMR: 1.9; CI: 1.3–2.8) but not among interferon-treated patients (SMR: 0.9; CI: 0.7–1.1). The risk of death was reduced, compared with untreated patients, among interferon-treated patients (risk ratio for overall death: 0.367; CI: 0.236 – 0.596; for liver-related death: 0.284; CI: 0.164 – 0.494) and among sustained responders (risk ratios: 0.148; CI: 0.064 – 0.343 and 0.050; CI: 0.012– 0.216). The risk of liver-unrelated deaths remained unchanged. Conclusions: Interferon therapy improved survival of chronic hepatitis C patients by preventing liver-related deaths.

epatocellular carcinoma is one of the major complications of chronic hepatitis C.1,2 In Japan, more than 30,000 people die annually because of primary liver

H

cancer, mostly hepatocellular carcinoma. Approximately 80% of cases were caused by hepatitis C virus (HCV) infection. The risk for hepatocellular carcinoma increases with the progression of liver fibrosis, reaching an annual incidence of as high as 8% in cirrhotic patients in Japan.3,4 Fortunately, cirrhosis is no longer a terminal stage of liver diseases. Esophageal varices can be controlled with endoscopic band ligation or sclerotherapy. We have effective treatments also for hepatocellular carcinoma, such as surgical hepatectomy and tumor ablation with alcohol or radio frequency. However, recurrence of hepatocellular carcinoma is frequent because the liver remains at high risk for the tumor. Despite recent advances in treatments,5,6 hepatocellular carcinoma remains one of the leading causes of death among chronic hepatitis C patients. Since 1994, the Inhibition of Hepatocarcinogenesis by Interferon Therapy (IHIT) Study Group has been surveying hepatocellular carcinoma development among a cohort of more than 3000 Japanese patients with histological-proven chronic hepatitis C.4,7 The IHIT Study has previously shown that interferon therapy reduces the risk for hepatocellular carcinoma by half among interferontreated patients, and down to one-fifth among virologicsustained responders, when compared with untreated patients.4 Several recent studies have reported similar effects of interferon therapy.8 –12 Moreover, another project of IHIT recently showed that not only inflammation but also fibrosis in the liver ameliorates significantly following virologic-sustained Abbreviations used in this paper: HCV, hepatitis C virus; IHIT, inhibition of hepatocarcinogenesis by interferon therapy; SMR, standardized mortality ratio. © 2002 by the American Gastroenterological Association 0016-5085/02/$35.00 doi:10.1053/gast.2002.34785

484

YOSHIDA ET AL.

response to interferon,7 underlying the improvement of liver function among the responders. Other studies showed that interferon therapy prevented the progression to liver failure in cirrhotic patients.13–15 Thus, interferon therapy may prolong life expectancy among chronic hepatitis C patients by preventing death from liver failure, as well as from hepatocellular carcinoma. However, a reduction in the risk of hepatocellular carcinoma does not necessarily indicate improvement of overall survival. Whereas cirrhotic patients are at high risk of hepatocellular carcinoma and hepatic failure, interferon therapy is less effective among them than among noncirrhotic patients.4 Indeed, several studies failed to confirm the effects of interferon therapy on hepatocellular carcinoma prevention among cirrhotic patients.15–18 In addition, because cirrhotic patients tend to be older, liver-unrelated mortality may not be negligible, obscuring a potential beneficial effect of interferon therapy on life expectancy. However, chronic hepatitis C may not be immediately life threatening for noncirrhotic patients because those patients are at low risk for hepatocellular carcinoma3,4 and the progression to cirrhosis is usually slow.7,19 –21 Because few studies have evaluated the effect of interferon therapy on overall survival among chronic hepatitis C patients, the IHIT Study Group started survival analysis among its cohort in 1996. The results, described in this article, showed a beneficial effect of interferon therapy on overall survival as well as survival from liverrelated deaths.

Materials and Methods Patients The IHIT Study Group consecutively enrolled 3223 patients who met the following inclusion and exclusion criteria.4 All patients were positive for HCV antibody by the second or third generation assay and underwent liver biopsy at 1 of the 8 participating institutions between June 1986, and February 1998. Exclusion criteria were the presence at the time of liver biopsy of hepatocellular carcinoma and the complication of other hepatic conditions such as infection with hepatitis B virus as determined by hepatitis B surface (HBs) antigen seropositivity, alcoholic liver disease, autoimmune hepatitis, or primary biliary cirrhosis. By surveying this cohort, we previously showed that interferon therapy reduced the risk for hepatocellular carcinoma4 and improved liver histology.7 Among the 3223 patients, 334 (10.3%) were referred to hospitals other than the participating institutions immediately after the liver biopsy. We were not able to confirm present survival status in 30% of those patients. To avoid possible biases, all 334 patients were excluded from the current analysis, which was thus based on the annual survey among the

GASTROENTEROLOGY Vol. 123, No. 2

remaining 2889 patients performed by the IHIT Study Group in 1999.

Histological Evaluation Liver biopsy specimens were assessed by a representative pathologist at each institution according to the criteria of Desmet et al.,22 where the staging of fibrosis was defined as F0 (no fibrosis), F1 (mild fibrosis), F2 (moderate fibrosis), F3 (severe fibrosis), or F4 (cirrhosis). We had confirmed interobserver concordance of histological evaluation previously.7

Interferon Therapy Two thousand four hundred thirty (84.1%) patients received interferon therapy, which was started within 1 year after liver biopsy. The other 459 patients remained untreated. Eighty-four percent of treated patients received interferon-␣; 14%, interferon-␤; and 2%, both. The median dose and duration of interferon administration were 480 mouse units (MU) (first quartile, 324 MU; third quartile, 600 MU) and 137 days (first quartile, 83 days; third quartile, 168 days), respectively. No patients received ribavirin. A sustained virologic response, defined as HCV RNA negativity more than 6 months after the termination of interferon therapy, was found in 817 (33.6%) patients. Positivity at the same time point, found in 1613 (66.4%) patients, was considered a nonsustained response.23,24 Four hundred fifty-nine patients did not receive interferon because of anxiety over adverse effects of interferon (11%), being busy during the scheduled treatment (22%), being not covered by health insurance (55%), or physician’s judgment on patient’s tendency toward depression (8%), severe diabetes mellitus (2%), pulmonary diseases (1%), or combined cardiovascular diseases (1%). Demographic characteristics of patients at entry were summarized in Table 1.

Survival Status and Cause of Death Because the IHIT Study Group protocol required that patients undergo biochemical blood tests every 1 to 3 months and receive abdominal ultrasonography at least every 6 months,4 the recent survival status of most patients was confirmed by checking medical records at each participating institution. If a patient skipped visits for more than 6 months, we made contact with the patient by letter or by telephone. Follow-up period was defined as the time between liver biopsy and either death or the latest confirmation of survival. We thus obtained an accumulated observation of 15,624 person-years, amounting to 91.5% of the total potential follow-up. During an average follow-up of 5.4 ⫾ 2.4 years, 86 patients died. Autopsy was performed in 26 (30%) patients, and the cause of death was noted accordingly. The other 31 patients died at the participating institutions, and the remaining 29 patients died elsewhere. Their cause of death was investigated from death certificates. Causes of death were divided into liver related and unrelated. The former category included hepatocellular carcinoma, liver failure, and gastrointestinal bleeding because of portal hypertension such as esophageal variceal bleeding; the latter category included extrahepatic malignancy, heart dis-

August 2002

INTERFERON PROLONGED LIFE EXPECTANCY

485

Table 1. Demographic Characteristics of Patients Interferon-treated

Characteristic

Untreated

All

Sustained response

Nonsustained response

Pa

Pb

Patients, n Age, yr Men/Women, n/n Biochemical variables ALT, U/L AST, U/L Albumin, g/L Bilirubin, ␮mol/L Platelet, ⫻10/L Fibrosis staging, n F0/F1 F2 F3 F4

459 54.6 ⫾ 11.1 237/223

2430 49.5 ⫾ 11.4 1542/888

817 47.5 ⫾ 11.9 570/247

1613 50.6 ⫾ 11.0 972/641

0.0001c 0.001d

0.0000d 0.001d

72 ⫾ 56 66 ⫾ 44 40 ⫾ 4 14 ⫾ 6 157 ⫾ 60

106 ⫾ 85 76 ⫾ 69 42 ⫾ 4 13 ⫾ 6 168 ⫾ 56

116 ⫾ 105 78 ⫾ 69 43 ⫾ 4 12 ⫾ 5 177 ⫾ 58

101 ⫾ 73 76 ⫾ 54 42 ⫾ 4 13 ⫾ 6 164 ⫾ 55

0.0005c 0.3161c 0.0000c 0.0001c 0.0001c

0.0001c 0.0001c 0.0000c 0.0146c 0.0002c

153 161 55 90

737 910 553 230

280 322 162 53

457 588 391 177

0.001d

0.001d

aDifference

between virologic sustained responders and nonresponders. between untreated patients and interferon-treated patients. cStudent t test. d␹2 test. bDifference

eases, cerebrovascular accidents, and pulmonary disease (Table 2). No patients underwent liver transplantation.

Results Cause of Death

Statistical Analysis Statistical analysis was performed with SAS software version 6.12 (SAS Institute, Inc., Cary, NC). All P values were two-tailed and considered significant when less than 0.05. The values of the confidence interval (CI) indicated the lower and upper 95% confidence limits. Continuous variables were expressed as mean ⫾ standard deviation unless otherwise indicated. Difference of means was evaluated by using Student t test. The ␹2 test was used to evaluate frequency distribution. Cumulative survival curves were determined with the KaplanMeier method. The number of expected deaths and the expected survival probability were calculated based on the gender- and age (5-year)-ranked mortality among the Japanese general population in 1994 published by the Statistics and Information Department, Japan Ministry of Health and Welfare.25 The standardized mortality ratio (SMR) was calculated by dividing the observed number of deaths by the expected number of deaths. The standard error and the 95% CI of SMR were estimated by assuming Poisson’s distribution,26 and changes in mortality of the study cohort from general population were considered as significant if the CI did not include unity. Survival was also analyzed by using Cox proportional hazards regression controlling for the age and gender of patients, and for the stages of liver fibrosis, used as strata of regression, in some calculations. Risk ratios attributed to each of virologicsustained and nonsustained responses were calculated as against no treatment by using dummy variables.

Thirty (6.5%) of 459 untreated patients and 56 (2.3%) of 2430 interferon-treated patients died during an average follow-up of 5.4 ⫾ 2.4 years. The causes of deaths were summarized in Table 2. Liver-related causes were responsible for 23 (77%) of 30 deaths among the untreated patients and 35 (63%) of 56 deaths among the interferon-treated patients. Hepatocellular carcinoma was the leading cause of deaths, claiming 39 victims (45% of all deaths and 67% of liver-related deaths). Malignancies other than hepatocellular carcinoma were the second leading cause of death, including 4 cases of colon cancer and 3 cases each of lung cancer and of stomach cancer and 1 case each of uterine cancer, of bile duct cancer, and of malignant lymphoma. No death directly related to interferon therapy, such as suicide, was reported. Cumulative Survival We considered it inappropriate to compare survival directly between the interferon-treated and untreated patients because male-to-female ratios and the mean age differed (Table 1). Thus, the observed cumulative survival among each group was respectively compared with the expected survival probability calculated based on the mortality in age- and gender-matched

486

YOSHIDA ET AL.

GASTROENTEROLOGY Vol. 123, No. 2

Table 2. Cause of Death Interferon-treated

Deaths, n Liver related, n Hepatocellular carcinoma Liver failure GI bleeding Liver unrelated, n Malignanciesa Heart diseases Cerebrovascular diseases Pulmonary disease Injury Unknown aExcluding

Untreated (n ⫽ 459)

All (n ⫽ 2430)

Sustained response (n ⫽ 817)

Nonsustained response (n ⫽ 1613)

30 23 (77%) 14 8 1 7 (23%) 2 2 1 1 0 1b

56 35 (63%) 25 4 6 21 (37%) 11 5 3 1 1 0

7 2 (29%) 1 0 1 5 (71%) 4 0 1 0 0 0

49 33 (67%) 24 4 5 16 (33%) 7 5 2 1 1 0

hepatocellular carcinoma. intravascular coagulation with unknown etiology.

bDisseminated

general population. Whereas survival among the untreated patients was apparently reduced compared with the expected survival (Figure 1A), that among the interferon-treated patients was not (Figure 1B). When inter-

feron-treated patients were divided into virologic-sustained responders and nonresponders, the survival of the former patients (Figure 1C), but not of the latter ones (Figure 1D), was better than the expected survival.

Figure 1. Cumulative survival (A) among untreated chronic hepatitis C patients (30 deaths among 459 patients), (B) among interferon-treated chronic hepatitis C patients (56 deaths among 2430 patients), (C) among virologic sustained responders to interferon (7 deaths among 817 patients), and (D) among nonsustained responders (49 deaths among 1613 patients). Survival curves were determined by the Kaplan-Meier method. Dotted lines indicate the upper and lower 95% confidence limits. Circles represent the expected survival probability based on the mortality in gender- and age-matched general population in Japan.

August 2002

INTERFERON PROLONGED LIFE EXPECTANCY

487

Table 3. Standardized Mortality Ratio Overall deaths Observed Expected Untreated Interferon treated All Sustained response Nonsustained response

Liver-related deaths SMR

Observed Expected

SMR

Liver-unrelated deaths Observed Expected

SMR

30

15.5

1.9 (1.3–2.8)

23

1.7

13.5 (8.6–20.3)

7

14.0

0.5 (0.2–1.0)

56 7 49

63.3 19.4 43.9

0.9 (0.7–1.1) 0.4 (0.1–0.7) 1.1 (0.8–1.5)

35 2 33

7.5 2.4 5.1

4.7 (3.3–6.5) 0.8 (0.1–3.0) 6.5 (4.5–9.1)

21 5 16

56.1 17.1 38.9

0.4 (0.2–0.6) 0.3 (0.1–0.7) 0.4 (0.2–0.7)

NOTE. Difference from the expected number of deaths was considered significant if 95% confidence interval of SMR (in parentheses) did not include unity.

SMR

5.4 years on average, only a few of the enrolled patients acquired those diseases afresh and died. We also calculated SMRs separately among the cirrhotic and noncirrhotic patients (Table 4). Overall mortality was elevated, as compared with the general population, among the untreated cirrhotic patients (SMR: 3.7; CI: 1.9 – 6.0) but not among the untreated noncirrhotic patients (SMR: 1.4; CI: 0.8 –2.3). However, liver-related mortality was high among not only cirrhotic patients but also noncirrhotic ones, and interferon therapy decreased liver-related mortality among cirrhotic and noncirrhotic patients. Next, SMRs were calculated with patients being stratified by age (Table 5). Untreated patients between 40 and 49 years of age showed the highest SMR for overall death (SMR: 6.0; CI: 1.9 –14.0), which was reduced by interferon therapy to one-sixth (SMR: 1.0; CI: 0.5–2.0). Untreated patients older than those patients showed lower SMRs for overall and liver-related deaths, although the mortality was still higher than that in the general population and reduced by interferon therapy.

Changes in mortality among interferon-treated and untreated patients were further assessed by calculating SMR, the ratio of the observed number of deaths to the expected number (Table 3). As compared with gender- and age-matched general population, overall mortality was high among the untreated patients (SMR: 1.9; CI: 1.3–2.8) but not among the interferon-treated patients (SMR: 0.9; CI: 0.7–1.1). Liver-related mortality was high among the untreated patients (SMR: 13.5; CI: 8.6 –20.3) and also among the interferon-treated patients to a lesser degree (SMR: 4.7; CI: 3.3– 6.5). However, the virologic-sustained responders had a liver-related mortality comparable with the general population (SMR: 0.8; CI: 0.1–3.0). SMR for liver-unrelated deaths was less than unity among both interferon-treated and untreated patients (Table 3). This may represent a selection bias introduced by the performance of liver biopsy as an inclusion criterion, which virtually excluded those patients with malignancy, severe heart diseases, and cerebrovascular diseases. These diseases occupy 70%– 80% of causes of deaths among the Japanese general population of the matched age. Because the follow-up was relatively short,

Multivariate Analysis The effect of interferon therapy on the risk of death was assessed by Cox proportional hazard regression

Table 4. Standardized Mortality Ratio: Cirrhotic and Noncirrhotic Patients Overall deaths Observed Expected Cirrhotic patients Untreated Interferon-treated All Sustained response Nonsustained response Noncirrhotic patients Untreated Interferon-treated All Sustained response Nonsustained response

Liver-related deaths SMR

Observed Expected

SMR

Liver-unrelated deaths Observed Expected

SMR

13

3.7

3.5 (1.9–6.0)

10

0.4

24.4 (11.7–44.9)

3

3.3

0.9 (0.2–2.7)

22 1 21

8.1 1.8 6.3

2.7 (1.7–4.1) 0.6 (0.0–3.1) 3.3 (2.1–5.1)

16 1 15

1.0 0.2 0.7

16.8 (9.6–27.4) 4.5 (0.1–25.3) 20.3 (11.3–33.4)

6 0 6

7.2 1.6 5.6

0.8 (0.3–1.8) 0.0 (0.0–2.3) 1.1 (0.4–2.3)

17

11.9

1.4 (0.8–2.3)

13

1.3

10.2 (5.5–17.5)

4

10.7

0.4 (0.1–1.0)

34 6 28

55.2 17.6 37.6

0.6 (0.4–0.9) 0.3 (0.1–0.7) 0.7 (0.5–1.1)

19 1 18

6.6 2.2 4.4

2.9 (1.7–4.5) 0.5 (0.0–2.6) 4.1 (2.4–6.5)

15 5 10

49.0 15.6 33.4

0.3 (0.1–1.0) 0.3 (0.1–0.7) 0.3 (0.1–0.6)

488

YOSHIDA ET AL.

GASTROENTEROLOGY Vol. 123, No. 2

Table 5. Standardized Mortality Ratio: Patients Stratified by Age at Entry Overall deaths Age (yr) Untreated patients ⱕ39 40-49 50-59 ⱖ60 Interferon-treated patients ⱕ39 40-49 50-59 ⱖ60

Observed Expected

Liver-related deaths SMR

Observed Expected

Liver-unrelated deaths

SMR

Observed Expected

SMR

0 5 9 16

0.1 0.8 3.5 11.2

0.0 (0.0–36.9) 6.0 (1.9–14.0) 2.5 (1.2–4.8) 1.4 (0.8–2.3)

0 3 7 13

0.0 0.1 0.4 1.2

0.0 (0.0–878.3) 46.2 (9.5–134.9) 17.1 (6.9–35.2) 10.8 (5.8–18.5)

0 2 2 3

0.1 0.6 3.2 10.1

0.0 (0.0–30.7) 3.1 (0.4–11.2) 0.6 (0.1–2.3) 0.3 (0.1–0.9)

0 9 24 23

1.9 8.6 23.6 31.1

0.0 (0.0–1.9) 1.0 (0.5–2.0) 1.0 (0.7–1.5) 0.7 (0.5–1.1)

0 6 19 10

0.1 0.7 2.9 3.9

0.0 (0.0–52.7) 8.5 (3.1–18.4) 6.6 (4.0–10.3) 2.6 (1.2–4.8)

0 3 5 13

1.9 6.0 20.8 27.6

0.0 (0.0–2.0) 0.5 (0.1–1.5) 0.2 (0.1–0.6) 0.5 (0.3–0.8)

controlling for the gender and age of patients. Interferon therapy significantly reduced the risk of overall death with a risk ratio of 0.375, as compared with no treatment (Table 6). When stratified into virologic-sustained and nonsustained responders, the virologic-sustained responders showed a further reduction in the risk of overall death (risk ratio: 0.148 compared with untreated patients), and the nonsustained responders had a higher but still significantly reduced risk (risk ratio: 0.472). Then, the risks of liver-related deaths (n ⫽ 58) and of liverunrelated deaths (n ⫽ 28) were separately evaluated by using Cox proportional hazards regression, where deaths because of irrelevant causes were treated as censored (Table 6). The reduction in the risk of liver-related death among the interferon-treated patients, as compared with untreated patients, was more prominent than that of overall death. In contrast, neither interferon therapy nor virologic-sustained response had significant effects on the risk of liver-unrelated deaths (Table 6). Because the majority of deaths were related to liver diseases and the risk of liver-related death was likely to change according to the stage of liver fibrosis, we performed multivariate analysis controlling also for the stages of fibrosis by using the stages as strata in regression. This modification scarcely changed the values of

risk ratios, attributing a risk ratio for overall deaths of 0.367 (CI: 0.227– 0.593; P ⫽ 0.0001) to interferon therapy. Then, we performed survival analysis among cirrhotic (n ⫽ 320; 230 treated) and noncirrhotic patients (n ⫽ 2569; 2197 treated) separately (Table 7). Interferon therapy was associated with a reduced risk of overall deaths among the noncirrhotic patients but not among the cirrhotic patients. However, the risk was significantly reduced among the virologic-sustained responders with cirrhosis as well as among those without cirrhosis (Table 7). Laboratory data, i.e., serum concentrations of albumin and bilirubin, levels of transaminases, and platelet count, at entry did not retain significance when they were included in the multivariate analysis as candidate explanatory factors for survival, to be selected by the stepwise method.

Discussion We previously demonstrated that chronic hepatitis C patients, especially those with cirrhosis, constitute a high-risk group for hepatocellular carcinoma development.4 In this study, we showed that chronic hepatitis C patients had higher overall and liver-related mortality

Table 6. Risk of Death Overall deaths Factors Men (vs. women) Age (by every 1 yr) Interferon therapy (vs. no treatment) All Sustained responsea Nonsustained responsea

Risk Ratio (95% CI)

Liver-related deaths P

Risk Ratio (95% CI)

Liver-unrelated deaths P

Risk Ratio (95% CI)

P

3.871 (2.214–6.770) 1.076 (1.037–1.096)

0.0001 4.503 (2.207–9.189) 0.0001 1.064 (1.032–1.096)

0.0001 2.930 (1.184–7.246) 0.0001 1.103 (1.050–1.157)

0.0200 0.0001

0.367 (0.236–0.596) 0.148 (0.064–0.343) 0.472 (0.294–0.757)

0.0001 0.284 (0.164–0.494) 0.0001 0.050 (0.012–0.216) 0.0018 0.388 (0.222–0.676)

0.0001 0.694 (0.287–1.681) 0.0001 0.549 (0.169–1.787) 0.0008 0.752 (0.302–1.873)

0.4186 0.3198 0.5403

NOTE. 86 deaths (58 liver-related and 28 unrelated) among 2889 patients were analyzed by using Cox’s proportional hazards regression. ratios attributed to gender and age changed slightly (not shown) when the 2 virologic responses, instead of interferon therapy as a whole, were separately evaluated. aRisk

August 2002

INTERFERON PROLONGED LIFE EXPECTANCY

489

Table 7. Risk of Overall Death: Cirrhotic and Noncirrhotic Patients

Death/Patients

Cirrhotic patients

Noncirrhotic patients

35/320

51/2569

Factors

Risk ratio (95% CI)

P value

Risk ratio (95% CI)

P value

Men (vs. women) Age (by every 1 yr) Interferon therapy (vs. no treatment) All Sustained responsea Nonsustained responsea

3.592 (1.549–8.330) 1.040 (0.996–1.086)

0.0029 0.0763

4.192 (1.969–8.924) 1.085 (1.052–1.126)

0.0001 0.0001

0.557 (0.276–1.134) 0.098 (0.012–0.766) 0.685 (0.337–1.394)

0.1066 0.0268 0.2970

0.363 (0.197–0.667) 0.198 (0.077–0.514) 0.437 (0.233–0.819)

0.0011 0.0009 0.0098

aRisk

ratios attributable to gender and age changed slightly (not shown) when the 2 virologic responses, were separately evaluated.

than the general population and that hepatocellular carcinoma was the leading cause of deaths among them, being responsible for 39 (45%) of the observed 86 deaths. Furthermore, we demonstrated that interferon therapy reduced overall and liver-related mortality, whereas liver-unrelated mortality remained unchanged. Our data were based on vital statistics in Japan, where HCV-related hepatocellular carcinoma is the major cause of liver-related deaths, and may not be directly applicable in other countries. We have no data concerning the other effective treatments received in both groups: prevention of bleeding with ␤-blockers, sclerotherapy, or band ligation; treatment of hepatocellular carcinoma by surgery, ethanol, or radio frequency. The effect of the interferon cannot be adjusted on these factors, which is a limitation of the study. However, the data clearly demonstrated the effect of interferon therapy on survival. There were only a few previous studies that assessed the effects of interferon therapy on survival. Niederau et al.15 showed a reduction in the risk of death or liver transplantation by interferon therapy using 838 patients, but they found no changes in the risk for hepatocellular carcinoma. Their observation differed from ours in several aspects, such as virologic-sustained response rate (6.2% vs. 33.6%), annual incidence of hepatocellular carcinoma among cirrhotic patients (2.5% vs. 7.9%), and the proportion of deaths because of hepatocellular carcinoma and liver failure (respectively, 18% and 41% vs. 45% and 14%). Although the reason for these differences is yet to be clarified, they may have led to different conclusions regarding the effect of interferon therapy on survival. Fattovich et al.21 found no beneficial effects of interferon therapy on survival, with a virologic-sustained response rate of 7%. In contrast, Ikeda et al.11 showed improved survival of patients, as well as reduced risk of hepatocellular carcinoma by interferon therapy among a cohort of Japanese patients. Their study resembled ours in the high incidence of hepatocellular carcinoma and high-virologic-sustained response rate when compared

with Western studies. However, they did not adjust survival analysis for confounding factors such as gender and age. Because the IHIT Study is a retrospective cohort study, it may be liable to several biases. First, interferontreated and untreated patients had different demographic characteristics, including age and gender. To adjust for those confounding factors, we compared the observed mortality with the expected mortality in matched general population by calculating SMR and performed multivariate analysis with proportional hazards regression. Both analyses showed beneficial effects of interferon therapy on the overall survival of patients. Second, 2034 patients (70.4%) were consecutively but retrospectively enrolled when the IHIT Study was started in 1994, and the remaining 865 patients were prospectively enrolled afterwards. Sixty-five (2.2%) patients dropped out before 1994 and 160 (5.5%) after 1994. We were not able to obtain information on their current survival status because of relocation, and they were censored as alive at the date of last confirmation of survival. The analysis may be biased if some of them have actually died. However, the proportion of dropouts was greater in the untreated group (56 of 459, 12%) than in the interferon-treated group (169 of 2430, 7%), and the proportion of cirrhotic patients among the dropouts was greater in the untreated group (5 of 56 vs. 8 of 169). Thus, the untreated group has a higher possibility of unrecorded liver-related deaths, and the current analysis, censoring all dropouts as alive, may be underestimating the effect of interferon therapy on survival. Several patients did not receive interferon therapy because of extrahepatic complications, which may have affected their survival. However, SMR for liver-unrelated death was similar in the untreated and interferon-treated groups (Table 3), suggesting that extrahepatic complications in untreated patients at entry did not significantly affect their survival.

490

YOSHIDA ET AL.

Heavy alcohol consumption is known to worsen survival of chronic hepatitis C patients.15 However, we do not have sufficient data to allow evaluation of interferon therapy in face of heavy alcohol consumption because less than 2% of patients drank more than 80 g of alcohol daily during the follow-up. Although low-alcohol consumption (20 to 80 g/day) was reported by 25% of patients at entry, 90% of these patients stopped drinking alcohol after undergoing liver biopsy. The results of this study suggest that patients with certain characteristics are more likely to receive beneficial effects of interferon therapy. For example, beneficial effects of interferon therapy on overall survival were validated among noncirrhotic patients but not among cirrhotic ones. Although virologic-sustained responders with cirrhosis showed a significant reduction in the risk of overall death (Table 5), this was obscured when evaluated among the interferon-treated cirrhotic patients as a whole by low–virologic-sustained response rate of interferon monotherapy. Although not licensed in Japan, a combination of interferon and ribavirin is recently reported to be more efficient than interferon alone,27,28 and the combination therapy may prove to be beneficial even to cirrhotic patients. SMR for overall deaths among untreated patients decreased in accordance with the age of patients (Table 5). Although the risk of liver-related deaths, as well as liver-unrelated deaths, increased with the age of patients, the risk ratio attributed to the age was greater for the liver-unrelated deaths (risk ratios, 1.103 vs. 1.064, Table 6), indicating that the importance of liver-related deaths relative to overall mortality became lower in older patients. In this sense, interferon therapy may be recommended, especially for younger patients. In conclusion, we showed in this study not only that chronic hepatitis C patients were at a higher risk of overall death than the general population but also that interferon therapy improved life expectancy by preventing liver-related deaths. The beneficial effect of interferon therapy on survival was much stronger among virologic-sustained responders than among nonsustained responders. We can expect further improvement of life expectancy with more effective antiviral therapy.

References 1. Tong MJ, el-Farra NS, Reikes AR, Co RL. Clinical outcomes after transfusion-associated hepatitis C. N Engl J Med 1995;332: 1463–1466. 2. Serfaty L, Aumaitre H, Chazouilleres O, Bonnand AM, Rosmorduc O, Poupon RE, Poupon R. Determinants of outcome of compensated hepatitis C virus-related cirrhosis. Hepatology 1998;27: 1435–1440. 3. Takano S, Yokosuka O, Imazeki F, Tagawa M, Omata M. Incidence of hepatocellular carcinoma in chronic hepatitis B and C: a

GASTROENTEROLOGY Vol. 123, No. 2

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

prospective study of 251 patients. Hepatology 1995;21:650 – 655. Yoshida H, Shiratori Y, Moriyama M, Arakawa Y, Ide T, Sata M, Inoue O, Yano M, Tanaka M, Fujiyama S, Nishiguchi S, Kuroki T, Imazeki F, Yokosuka O, Kinoyama S, Yamada G, Omata M. Interferon therapy reduces the risk for hepatocellular carcinoma: national surveillance program of cirrhotic and noncirrhotic patients with chronic hepatitis C in Japan. Ann Intern Med 1999;131: 174 –181. Khan KN, Yatsuhashi H, Yamasaki K, Yamasaki M, Inoue O, Koga M, Yano M. Prospective analysis of risk factors for early intrahepatic recurrence of hepatocellular carcinoma following ethanol injection. J Hepatol 2000;32:269 –278. Figueras J, Jaurrieta E, Valls C, Benasco C, Rafecas A, Xiol X, Fabregat J, Casanovas T, Torras J, Baliellas C, Ibanez L, Moreno P, Casais L. Survival after liver transplantation in cirrhotic patients with and without hepatocellular carcinoma: a comparative study. Hepatology 1997;25:1485–1489. Shiratori Y, Imazeki F, Moriyama M, Yano M, Arakawa Y, Yokosuka O, Kuroki T, Nishiguchi S, Sata M, Yamada G, Fujiyama S, Yoshida H, Omata M. Histologic improvement of fibrosis in patients with hepatitis C who have sustained response to interferon therapy. Ann Intern Med 2000;132:517–524. Mazzella G, Accogli E, Sottili S, Festi D, Orsini M, Salzetta A, Novelli V, Cipolla A, Fabbri C, Pezzoli A, Roda E. Alpha interferon treatment may prevent hepatocellular carcinoma in HCV-related liver cirrhosis. J Hepatol 1996;24:141–147. Kasahara A, Hayashi N, Mochizuki K, Takayanagi M, Yoshioka K, Kakumu S, Iijima A, Urushihara A, Kiyosawa K, Okuda M, Hino K, Okita K. Risk factors for hepatocellular carcinoma and its incidence after interferon treatment in patients with chronic hepatitis C. Osaka Liver Disease Study Group. Hepatology 1998;27: 1394 –1402. Imai Y, Kawata S, Tamura S, Yabuchi I, Noda S, Inada M, Maeda Y, Shirai Y, Fukuzaki T, Kaji I, Ishikawa H, Matsuda Y, Nishikawa M, Seki K, Matsuzawa Y. Relation of interferon therapy and hepatocellular carcinoma in patients with chronic hepatitis C. Osaka Hepatocellular Carcinoma Prevention Study Group. Ann Intern Med 1998;129:94 –99. Ikeda K, Saitoh S, Arase Y, Chayama K, Suzuki Y, Kobayashi M, Tsubota A, Nakamura I, Murashima N, Kumada H, Kawanishi M. Effect of interferon therapy on hepatocellular carcinogenesis in patients with chronic hepatitis type C: A long-term observation study of 1,643 patients using statistical bias correction with proportional hazard analysis. Hepatology 1999;29:1124 –1130. Okanoue T, Itoh Y, Minami M, Sakamoto S, Yasui K, Sakamoto M, Nishioji K, Murakami Y, Kashima K. Interferon therapy lowers the rate of progression to hepatocellular carcinoma in chronic hepatitis C but not significantly in an advanced stage: a retrospective study in 1148 patients. Viral Hepatitis Therapy Study Group. J Hepatol 1999;30:653– 659. Nishiguchi S, Kuroki T, Nakatani S, Morimoto H, Takeda T, Nakajima S, Shiomi S, Seki S, Kobayashi K, Otani S. Randomised trial of effects of interferon-alpha on incidence of hepatocellular carcinoma in chronic active hepatitis C with cirrhosis. Lancet 1995;346:1051–1055. Benvegnu L, Chemello L, Noventa F, Fattovich G, Pontisso P, Alberti A. Retrospective analysis of the effect of interferon therapy on the clinical outcome of patients with viral cirrhosis. Cancer 1998;83:901–909. Niederau C, Lange S, Heintges T, Erhardt A, Buschkamp M, Hurter D, Nawrocki M, Kruska L, Hensel F, Petry W, Haussinger D. Prognosis of chronic hepatitis C: results of a large, prospective cohort study. Hepatology 1998;28:1687–1695. Fattovich G, Giustina G, Degos F, Diodati G, Tremolada F, Nevens F, Almasio P, Solinas A, Brouwer JT, Thomas H, Realdi G, Corrocher R, Schalm SW. Effectiveness of interferon alfa on incidence

August 2002

17.

18.

19.

20.

21.

22.

23.

of hepatocellular carcinoma and decompensation in cirrhosis type C. European Concerted Action on Viral Hepatitis (EUROHEP). J Hepatol 1997;27:201–205. Valla DC, Chevallier M, Marcellin P, Payen JL, Trepo C, Fonck M, Bourliere M, Boucher E, Miguet JP, Parlier D, Lemonnier C, Opolon P. Treatment of hepatitis C virus-related cirrhosis: a randomized, controlled trial of interferon ␣-2b versus no treatment. Hepatology 1999;29:1870 –1875. Hu KQ, Tong MJ. The long-term outcomes of patients with compensated hepatitis C virus-related cirrhosis and history of parenteral exposure in the United States. Hepatology 1999;29:1311– 1316. Poynard T, Bedossa P, Opolon P. Natural history of liver fibrosis progression in patients with chronic hepatitis C. The OBSVIRC, METAVIR, CLINIVIR, and DOSVIRC groups. Lancet 1997;349: 825– 832. Seeff LB, Buskell-Bales Z, Wright EC, Durako SJ, Alter HJ, Iber FL, Hollinger FB, Gitnick G, Knodell RG, Perrillo RP. Long-term mortality after transfusion-associated non-A, non-B hepatitis. The National Heart, Lung, and Blood Institute Study Group. N Engl J Med 1992;327:1906 –1911. Fattovich G, Giustina G, Degos F, Tremolada F, Diodati G, Almasio P, Nevens F, Solinas A, Mura D, Brouwer JT, Thomas H, Njapoum C, Casarin C, Bonetti P, Fuschi P, Basho J, Tocco A, Bhalla A, Galassini R, Noventa F, Schalm SW, Realdi G. Morbidity and mortality in compensated cirrhosis type C: a retrospective follow-up study of 384 patients. Gastroenterology 1997;112: 463– 472. Desmet VJ, Gerber M, Hoofnagle JH, Manns M, Scheuer PJ. Classification of chronic hepatitis: diagnosis, grading, and staging. Hepatology 1994;19:1513–1520. Reichard O, Glaumann H, Fryden A, Norkrans G, Schvarcz R, Sonnerborg A, Yun ZB, Weiland O. Two-year biochemical, virological, and histological follow-up in patients with chronic hepatitis C responding in a sustained fashion to interferon ␣-2b treatment. Hepatology 1995;21:918 –922.

INTERFERON PROLONGED LIFE EXPECTANCY

491

24. Shiratori Y, Kato N, Yokosuka O, Imazeki F, Hashimoto E, Hayashi N, Nakamura A, Asada M, Kuroda H, Tanaka N, Arakawa Y, Omata M. Predictors of the efficacy of interferon therapy in chronic hepatitis C virus infection. Tokyo-Chiba Hepatitis Research Group. Gastroenterology 1997;113:558 –566. 25. Statistics and Information Department, Japan Ministry of Health and Welfare. Vital statistics in Japan (in Japanese). Tokyo: Health and Welfare Statistics Association, 1996. 26. Di Marco V, Lo Lacono O, Camma C, Vacccaro A, Giunta M, Martorana G, Fuschi P, Almasio PL. The long-term course of chronic hepatitis B. Hepatology 1999;30:257–263. 27. Schalm SW, Fattovich G, Brouwer JT. Therapy of hepatitis C: patients with cirrhosis. Hepatology 1997;26:S128S–S132. 28. McHutchison JG, Poynard T. Combination therapy with interferon plus ribavirin for the initial treatment of chronic hepatitis C. Semin Liver Dis 1999;19(suppl 1):57– 65. 29. Reddy KR, Wright TL, Pockros PJ, Shiffman M, Everson G, Reindollar R, Fried MW, Purdum PP III, Jensen D, Smith C, Lee WM, Boyer TD, Lin A, Pedder S, DePamphilis J. Efficacy and safety of pegylated (40-kd) interferon ␣-2a compared with interferon ␣-2a in noncirrhotic patients with chronic hepatitis C. Hepatology 2001;33:433– 438.

Received May 13, 2001. Accepted May 9, 2002. Address requests for reprints to: Haruhiko Yoshida, M.D., Department of Gastroenterology, University of Tokyo Graduate School of Medicine, 7-3-1 Hongo, Bunkyo, Tokyo 113-8655, Japan. e-mail: [email protected]; fax: (81) 3-3814-0021. Supported in part by the Ministry of Health, Labour, and Welfare of Japan as one of the Comprehensive 10-Year Strategy for Cancer Control Projects and of Medical Frontier Strategy Researches and in part by the Ministry of Education, Culture, Sports, Science, and Technology of Japan. The authors thank Dr. Tadao Kakizoe for support.