Predictors of sustained response to alpha interferon therapy in chronic hepatitis C

Clinical Biochemistry, Vol. 32, No. 7, 537–545, 1999 Copyright © 1999 The Canadian Society of Clinical Chemists Printed in the USA. All rights reserved 0009-9120/99/$–see front matter

PII S0009-9120(99)00053-3

Predictors of Sustained Response to Alpha Interferon Therapy in Chronic Hepatitis C MANUELA G. NEUMAN,1 JEAN-PIERRE BENHAMOU,2 MICHELLE MARTINOT,2 NATHALIE BOYER,2 NEIL H. SHEAR,1 IZABELLA MALKIEWICZ,1 GADY G. KATZ,1 ASHIMA SUNEJA,1 STEVE SINGH,1 and PATRICK MARCELLIN2 1

Division of Clinical Pharmacology, Sunnybrook and Women’s Health Sciences Centre, and Department of Pharmacology, University of Toronto, Toronto, Ontario, Canada, and 2Service d’Hepatologie, INSERM U481 and Hopital Beaujon, Clichy, France Objectives: To utilize cytokine levels to predict sustained response (SR) to alpha interferon (IFN ␣) therapy in chronic hepatitis C patients, and to determine the relationship between serum tumor necrosis factor ␣ (TNF ␣), interleukin (IL) IL 6, IL 8, IL 12, transforming growth factor beta (TGF ␤ 1) and the degree of liver damage as reflected by traditional markers. Design and methods: Serum cytokine levels were assessed using ELISA in 18 patients included in a controlled clinical trial of IFN ␣. Results: Of the 18 patients, 27% were sustained responders (SR), 27% were response and relapse responders (RR), and 46% were non-responders (NR). Multivariate analysis showed that a low serum TNF ␣ level and high serum IL 8 levels were independent factors associated with SR to IFN ␣ therapy. Serum TNF ␣ level highly correlated with viral load and genotype predictive values (p ⬍ 0.001). Therapy lowered the IL 6 and IL 12 profile. TGF ␤ 1 levels in serum are positively correlated with fibrinogenesis. Conclusions: IFN ␣ therapy modulates immune response to hepatits C virus, contributing to sustained response. Copyright © 1999 The Canadian Society of Clinical Chemists

KEY WORDS: chronic hepatitis C; cytokines; IFN therapy; transforming growth factor beta; tumor necrosis factor alpha.

Introduction our of the six identifiable hepatitis viruses are associated with chronic disease. The constant stimulation of cell damage over years, followed by regeneration and repair, ultimately leads to cirrhosis in at least 20% of patients within 20 years. Chronic hepatitis C patients are at an increased risk of developing primary liver cancer, with an annual incidence of about 1– 4% (1). Hepatitis C remains

F

Correspondence: M. G. Neuman Ph.D., Division Clinical Pharmacology, E-240, Sunnybrook and Women’s Health Sciences Centre, 2075 Bayview Avenue, Toronto,Ontario, M4N 3M5, Canada. E-mail: [email protected]. utoronto.ca. Manuscript received July 19, 1999; accepted July 19, 1999. CLINICAL BIOCHEMISTRY, VOLUME 32, OCTOBER 1999

difficult to treat, and treatment with interferon (IFN) has a long-term efficacy below 25% (2). In chronic hepatitis B-D, the rate of sustained response to IFN ␣ therapy is low (3). In about 10% of the patients, therapy is interrupted, mainly because of severe fatigue, thyroid dysfunction, exacerbation of diabetes, or depression (4 –12). The response to treatment is generally assessed in terms of a return to normal clinical status and transaminases activity (4,5). Hepatitis C remains difficult to treat, therefore establishing immunological status and identifying patients with a better response to IFN ␣ may help the physician direct the management of such patients more successfully. The following factors are clearly predictive of the response to IFN ␣: young age, short time since onset of infection, absence of cirrhosis, lower-level viremia, and infection by HCV genotypes other than 1b (3–5). In patients with active hepatitis C virus (HCV) infection, IFN ␣ treatment has a long-term response of only 20 –25% (5,13–20). IFN ␣ therapy during the acute phase of hepatitis C significantly reduces the risk of chronic liver disease (21–23). Interferons (IFNs) are a family of three proteins, IFN ␣, ␤, and ␥, that have the ability to interfere with viral activity. IFN ␣ and ␤ mainly have antiproliferative and antiviral activity, while IFN ␥ has mainly immunomodulatory effects. In the longer term, IFN therapy could prevent the onset of liver cancer in patients with viral cirrhosis (5). The central role of cytokines includes cell-to-cell communication, inflammatory response amplification, and immune response regulation. Cytokines can mediate cell to-cell communication that is separate from or in addition to, contact-mediated cell-cell communication (24). Cytokines can either synergize or antagonize other cytokines. The cytokine interactions lead to a cascade of functions (25). Some 537

NEUMAN

reports showed that in cells in vitro, detectable IL 2, IL 10, and IFN ␥ production can occur in some HCV patients before or after therapy (or both) (25–29). In patients with chronic hepatitis C, measurement of serum IFN ␣ at baseline might be useful for the selection of patients with higher probability of longterm response (30). Findings suggest that secretion of interleukins by mononuclear cells from control individuals and various patient groups with different degrees of chronic hepatitis may be different, and that cytokines might show different effects on IFN ␥ production by some cells (30 –36). Apart from immunoregulatory and antiviral effect, cytokines may be involved in inflammation (TNF ␣, IL 1, IL 6, and IL 12) (25). Chemokines constitute a superfamily of small inducible, secreted, proinflammatory cytokines involved in a variety of immune responses, acting primarily as chemoattractants and activators of specific types of leukocytes. Great interest has been generated by the finding that viruses use some chemokine receptors as co-receptors to enter the cells. There are also multiple mechanisms of viral interaction with cytokines and host defense (molecular mimicry of IL 10 cytokine and cytokine receptor) (37,38), inhibition of transcriptional activation (IFN), or formation of cellular proteins (39). The objectives of the present study were to measure levels of cytokines in serum of patients with HCV at the basal level and to correlate the levels of these cytokines and other predictive factors for the outcome of the disease (ALD genotype). Another objective was to monitor the influence of IFN ␣ on the status of cytokines during the period of patient treatment. Pretreatment levels of known cytokines may be predictive of likely response to IFN ␣ therapy in HCV patients, leading to more effective and rational patient selection and direction of therapy. Patients and methods Patients (n ⫽ 18) with chronic hepatitis C, defined by the usual biochemical, serological and histological criteria, were enrolled in a controlled trial of IFN ␣ (4). PATIENT

ENROLLMENT

Patients with detectable pretreatment serum HCV RNA by polymerase chain reaction (PCR) were included in this study. All patients received alpha interferon (␣-2b, INTRON A, Schering-Plough, Kenilworth, NJ, USA) at the dose of 3 million units, three times a week (3TIW) for 12 months. All patients tested positive for antibody to HCV, on third-generation ELISA and recombinant immunoblot assay (Ortho Diagnostic Systems, Roissy, France). The criteria for inclusion in the trials were: (i) persistently elevated serum alanine aminotransferase (ALT) levels for more than 6 months before randomization; (ii) no evidence of infection with hepatitis B virus (absence of detectable hepatitis B 538

ET AL.

surface antigen); (iii) exclusion of other causes of chronic liver disease (alcoholism, hepatotoxic drugs, autoimmune chronic hepatitis, hemochromatosis, Wilson’s disease and ␣-1 antitrypsin deficiency); and (iv) liver histologic examination showing lesions characteristic of chronic hepatitis. The criteria for exclusion were: (i) a previous course of interferon; (ii) pregnant females and females not adequately protected against pregnancy during the course of the study; (iii) history of decompensated cirrhosis (ascites, bleeding esophageal varices or hepatic encephalopathy); (iv) history of depressive illness; (v) presence of anti-HIV antibodies; (vi) prothrombin activity less than 50% of normal, hemoglobin less than 110 g/L, neutrophil count below 1.5 ⫻ 109/L or platelet count less than 100 ⫻ 109/L. The trial was approved by the Ethics Committee of the Faculty of Medicine Xavier Bichat, France. Informed consent was obtained from all patients. Patient characteristics are given in Table 1. CONTROLS Serum levels of the proinflammatory cytokines were measured at base line and during IFN ␣ therapy. The levels of TNF ␣, IL 6, IL 8, and IL 12 in the entire HCV group (18 patients) were compared to 80 normal, healthy volunteers, 90 individuals diagnosed as hypersensitivity reactions (HSR) to sulphonamide antibiotics or to aromatic anticonvulsants, and 10 patients with alcoholic liver disease (ALD). Patients with HSR or ALD were negative for HCV infection. HSR patients were recruited from the Glaxo-Wellcome–Sunnybrook Drug Safety Clinic. This is a tertiary care clinic where patients with histories suggestive of drug reactions are evaluated. HSRs have been defined by the triad of fever, skin rash, and single or multiple internal organ involvement that starts within 12 weeks after initiation of therapy. We chose to compare serum levels of IL 6, IL 8 (Figure 1), IL 12, and TNF ␣ (Figure 2) in chronically HCV-infected individuals and in patients with HSR, as some of the HCV patients may have also had an autoimmune component to their liver inflammation, compatible with HSR. The comparison with ALD patients was chosen because ALD is known to produce liver inflammation and to release high serum levels of proinflammatory cytokines in serum. DEFINITION

OF RESPONSE TO THERAPY

Patients were separated into three groups according to the response. Patients with sustained response (SR) were defined by the normalization of serum ALT and the absence of detectable serum HCV RNA during treatment and 6 months posttreatment. Patients with response and relapse (RR) were defined by the normalization of serum ALT and non-detectable serum HCV RNA at the end of the treatment but with an increase of serum ALT and presence of HCV RNA occurring at the 6-month CLINICAL BIOCHEMISTRY, VOLUME 32, OCTOBER 1999

INFLAMMATORY CYTOKINES IN CHRONIC HEPATITIS C

TABLE 1 Comparison of the Controls and HCV Groups at Entry

Age (years) Sex (F/M) Histology (%) Chronic persistent hepatitis Chronic active hepatitis ALT (IU/L) (normal range, 12–32) Sources of HCV infection (%) Blood transfusion or surgery Drug abuse Unknown Duration of HCV infection (years)

Control

HCV

32 ⫾ 14 15/65 Not performed

39 ⫾ 7a 4/14

44 ⫾ 24a 1/4

42 ⫾ 18a 60/30

38 62 91 ⫾ 9

40 60 82 ⫾ 28 Not performed

Not performed

18 ⫾ 12* None

15 ⫾ 2

ALD

HSR

28 ⫾ 18* None

39 33 28

Hypersensitivity syndrome reactions have been defined by the triad of fever, skin rash and single or multiple internal organ involvement that starts within 12 weeks after initiation of therapy with sulfonamide antibiotics or aromatic anticonvulsant. Groups: Control-healthy volunteers; ALD-alcoholic liver disease; HCV-chronic hepatitis C infected individuals; HSRhypersensitivity syndrome reaction. Values represent mean ⫾ SD. *p ⬍ 0.001 lower versus HCV-infected individuals. a Difference was not significant versus non-infected individuals.

follow-up. Patients with no response (NR) were defined by elevated serum ALT and the presence of HCV RNA at the end of the treatment.

Figure 1—Serum levels of TNF ␣ and IL 12 in normal individuals (control), and in patients with hypersensitivity reaction (HSR) to drugs, alcohol liver disease (ALD), or chronic hepatitis C (HepC). The levels of TNF ␣ and IL 12 in the entire HCV group (18 patients) were compared to 80 normal, healthy volunteers, 90 individuals diagnosed as presenting hypersensitivity reactions (HSR) to sulphonamide antibiotics or to aromatic anticonvulsant, and in 10 patients with alcoholic liver disease (ALD). The level of cytokines was assessed as described in materials and methods. The results are presented in picograms/mL ⫾ standard deviation (S.D.). There were significant differences between the level of TNF ␣ in chronic hepatitis C patients; (p ⬍ 0.001) when compared to controls and HSR patients, and p ⬍ 0.05 when compared to ALD patients. IL 12 levels are higher than control and HSR (p ⬍ 0.001), in chronic hepatitis C patients. No significant difference was found between levels in chronic hepatitis C patients and ALD patients. CLINICAL BIOCHEMISTRY, VOLUME 32, OCTOBER 1999

CHARACTERISTICS

STUDIED

The following characteristics were compared between the three groups of patients: gender, age, source of HCV infection, duration of HCV infection, pretreatment levels of serum ALT, liver histology, type of IFN, pretreatment serum HCV RNA level, HCV genotype. In patients who became infected by blood transfusion or intravenous drug use, the duration of HCV infection was estimated as the interval between the date of transfusion or the date of the

Figure 2—Serum levels of IL 6 and IL 8 in normal individuals (control), and in patients with hypersensitivity reaction (HSR) to drugs, alcohol liver disease (ALD), or chronic hepatitis C (HepC). In chronic hepatitis C patients, IL 6 is significantly higher (p ⬍ 0.001) when compared to the levels found in control and HSR. No significant difference in serum IL 6 level was found between chronic hepatitis C patients and ALD. IL 8 levels are higher in chronic hepatitis C patients than the values in control and HSR (p ⬍ 0.001), and when compared with ALD patients (p ⬍ 0.05). 539

NEUMAN

onset of intravenous drug use and the date of initiation of treatment. Patients with unknown source of infection were not considered for this variable. Liver biopsy specimens were obtained from all patients within the 6 months before initiation of treatment. The histological preparations were examined in a blinded manner and scored. For statistical analysis, activity and fibrosis scores were defined as low if they were 0 or 1 and high if they were more than 1 as defined by Marcellin et al. (4). LABORATORY

TESTS

Patients were sampled at time 0 (base line), at the end of the treatment, and a 6 months post-treatment follow-up. Patient serum specimens were kept at 4° C immediately after collection, centrifuged after formation of the clot through a paraffin plug, and frozen at ⫺80° C within 2 h of being drawn, until the day of quantitation and genotyping of cytokine measurements. This management provided the optimal condition for reliable results (4). DETECTION

OF SERUM HEPATITIS

C

VIRUS

RNA

Serum HCV RNA was detected using reversetranscription polymerase chain reaction with primers located in the 5⬘ noncoding region of the HCV genome (AMPLICOR HCV Amplification Kit, Roche Diagnostics, Neuilly, France). Serum HCV RNA detection with PCR was performed at initiation of the treatment for HCV genotyping, and 6 months after treatment in all the patients with a biochemical sustained response (4). QUANTITATION

OF SERUM HEPATITIS

C

VIRUS

RNA

Serum HCV RNA quantitation was done with the improved quantitative branched DNA signal amplification assay (Quantiplex™ HCV RNA 2.0 Chiron Diagnostics, Eragny sur Oise, France). The assay, based on specific hybridization of synthetic oligonucleotides located in the 5⬘ untranslated region (UR) of the HCV genome, incorporates refined oligonucleotide probe sets based on sequence variation of disparate HCV isolates. This assay is more efficient for HCV RNA quantitation of HCV genotypes 2 and 3, and gives an equivalent quantitation for HCV genotypes 1 to 6. The quantification cut-off of the assay is 0.2 ⫻ 106 Eq genomes per milliliter. All the samples were run in duplicate. GENOTYPING

OF HEPATITIS

C

VIRUS

HCV genotyping was performed on serum collected at initiation of treatment in the 5⬘ UR of the HCVgenome, using reverse hybridization with the line probe assay (LiPA), (InGeN, Rungis, France). During the “nested” PCR, the product is biotinylated. The amplification products obtained are hybridized to oligonucleotides directed against the variable region of the 5⬘ UR, immobilized as parallel 540

ET AL.

lines on membrane strips, and then incubation with streptavidin labeled with alkaline phosphatase allowed detection of hybrids. The HCV line-probe assay contained 15 probe lines, allowing identification of HCV types 1 to 5 and subtypes 1a, 1b, 2a, 2b, 3a, 3b, 4a, and 5a (40). CYTOKINE

MEASUREMENT

Cytoscreen™, Immunoassay Kits, Human IL 6, IL 8, IL 12, TGF ␤ and TNF ␣, Enzyme-Linked-Immuno-Sorbent Assay (ELISA) (Biosource International, Camarillo, CA, USA) were used for the quantitative determination of cytokines in serum as previously described (26). The assay is designed to recognize both natural human and recombinant human cytokines. The principle of the method is using a solid phase assay. The wells of a 96-microtiter plate were coated with antibody specific for each human cytokine. The readings were done at 450 nm. The correlation coefficient was linear (r ⫽ 0.989) in a concentration range between 2 and 500 pg/mL. The samples having higher concentrations were diluted. Each specimen was analyzed in triplicate with a sensitivity of 95% and a specificity of 98%. We used standards and reference reagents available from the National Institute for Biological Standards and Controls (NIBSC, Herts., UK). The method is standardized in our laboratory according to the procedures described by Ruiz Arguelles (41). Statistical analysis We compared each of the parameters at base-line versus the other points in time of the treatment by using SPSS 7.0 for Windows 95. Normality of data was tested by means of Shapiro and Wilk’s W-test. Most of the data was in a normal distribution. To test the differences between groups, we compared the mean ⫾ standard deviation (S.D.) of each one of the parameters using either parametric or non-parametric tests. The initial histological lesions were evaluated by the non-parametric rank correlation for each parameter. Modifications of variables during the therapy were compared using a Chi-square test. Some quantitative variables were compared using the Student paired t-test. Mann-Whitney and Wilcoxon rank–sum tests were used to compare values of continuous variables. Correlation between variables was analyzed calculating the Spearman rank correlation coefficient. The details for specific statistical tests will be given in legends and figures. To determine the independent prognostic value of the selected characteristics, a logistic regression model was used. Results BIOCHEMICAL

AND VIROLOGICAL RESPONSE TO THE

TREATMENT

Five of 18 patients (28%) were SR to IFN ␣ therapy, while five other patients (28%) were RR 6 month after cessation of therapy. CLINICAL BIOCHEMISTRY, VOLUME 32, OCTOBER 1999

INFLAMMATORY CYTOKINES IN CHRONIC HEPATITIS C

TABLE 2 Clinical and Biochemical Parameters Correlated to Serum HCV RNA Level Pretreatment characteristics Number Mean age (years) Source of HCV infection (%) Blood transfusion Drug addiction Unknown Duration of HCV infection (years) Serum ALT (IU/L) HCV genotype (%) 1a 1b 2a 3a Others

Serum HCV RNA (⫻106 Eq. Genomes/mL) Low (⬍0.20)

Medium (0.2 to 3.0)

High (⬎3.0)

p

7 39 ⫾ 5

5 42 ⫾ 6

0.01

22a 23 55b

40 28 32

38 49b 33

0.05 0.05 0.05

10 ⫾ 6 54 ⫾ 13a

13 ⫾ 7 83 ⫾ 18

13 ⫾ 9 114 ⫾ 45

ns 0.05

4 2a 4 4 2

19 11 6 2 2

28b 10 4 0 0

0.05

6 35 ⫾ 5

ns

Data are given in mean ⫾ SD. a Significantly lower compared to the 2 other HCV RNA genome titer groups. b Significantly higher compared to the other groups.

The clinical and biochemical data were grouped and presented as a function of HCV titer. A correlation between the level of serum ALT, HCV titer and the source of infection was observed (Table 2). Patients infected via blood transfusion or intravenous drug use, had higher ALT levels and higher HCV titers when compared with the patients whose source of infection was unknown (p ⬍ 0.05). There was no correlation between the duration of the infection, ALT levels, and HCV titer. Patients with genotype 1 demonstrated a significantly higher HCV RNA level (p ⬍ 0.05) compared with other genotypes. There were significantly fewer patients (p ⬍ 0.05) with genotype 1 b, who presented with lower levels of HCV RNA pre-treatment. HCV RNA titer was correlated positively with ALT levels. SERUM

CYTOKINE LEVEL

Each one of the groups was found to have a distinct cytokine profile. HCV patients had significantly higher levels of cytokines versus control and HSR patients (p ⬍ 0.001). Only IL 8 and TNF ␣ levels were significantly higher when compared to the ALD patients (p ⬍ 0.05). TGF ␤ levels (ng/mL) (53.5 ⫾ 6.5) were increased significantly in the ALD group when compared to the HCV group (29.9⫾ 7.0) (p ⬍ 0.05). Prior to treatment, the TGF ␤ levels were not significantly different between the HCV sub-groups SR (34.85 ⫾ 6.5), RR (24.85 ⫾ 8.5), and NR (33.67 ⫾ 4.9). After the treatment, TGF ␤ values were significantly higher in NR (52.38 ⫾ 5.8) versus RR (35.0 ⫾ 2.5) or SR (25.0 ⫾ 3.5) subjects. There was a significant increase in TGF ␤ in the NR sub-group after completion of IFN ␣ treatment, while the RR and SR levels did not significantly change, either at the CLINICAL BIOCHEMISTRY, VOLUME 32, OCTOBER 1999

beginning of the therapy or during the therapy (p ⬍ 0.05). The correlation between the differences (end of treatment and base-line) of TGF ␤ values with the Knodell score was significant (r ⫽ 0.247, p ⫽ 0.05). The changes within 2 years in TGF ␤ are also significantly correlated (r ⫽ 0.258, p ⫽ 0.05) with the level of serum transaminases. There were statistically significant differences observed at the beginning of the treatment for TNF ␣ levels, between the patients belonging to SR when compared to RR or NR as shown in Figure 3. The rate of change of TNF ␣ changed with the treatment differently for the HCV subgroups. In untreated HCV patients the level in all patients was above normal (Figure 2); however, in SR patients the levels were significantly falling even further than the levels in the other 2 subgroups (p ⬍ 0.05). Interestingly, in the SR subgroup, TNF ␣ levels at the end of the treatment were significantly lower than at baseline and further reduced during the follow-up period. The TNF ␣ levels in the RR subgroup dropped drastically during treatment but at the follow-up the level returned to the initial or even higher levels (p ⬍ 0.001). The TNF ␣ levels in NR subgroup did not change significantly neither with treatment nor after the cessation of therapy. A correlation (p ⫽ 0.039) was found between the ALT (Table 3) and TNF ␣ levels. Also the rate of change of interleukins had a different profile for the different groups. IL 6 levels (Figure 4) decreased significantly only in the SR group and only at the 6 months follow-up point versus base line. IL 8 levels in SR group (Figure 5) increased after treatment, and remained high at follow-up (p ⬍ 0.05). In the RR group the initial IL 8 level was higher than the other groups, but decreased significantly during the treatment (p ⬍ 541

NEUMAN

ET AL.

TABLE 3 Changes in ALT During the Treatment Time SR (5) (patients) Baseline 80.2 ⫾ 12.6 End of 19.2 ⫾ 4.3* treatment Follow up 29.8 ⫾ 17.2*

RR (5)

NR (8)

86.8 ⫾ 11.7 18.2 ⫾ 1.8**

101.0 ⫾ 17.9 80.1 ⫾ 15.9

98.0 ⫾ 17.1†

69.3 ⫾ 13.1‡

Values are given in mean ⫾ standard error. SR-sustained responders: *p ⬍ 0.05 lower than baseline. RR-relapse responders: **p ⬍ 0.05 lower than base line, †p ⬍ 0.05 higher than the value at the end of the treatment. NR-non-responders: ‡p ⫽ 0.039 vs. baseline.

Figure 3—Effect of interferon ␣ (IFN ␣) therapy on TNF ␣ levels in chronic hepatitis C patients. The 18 chronic hepatitis C patients were separated into three groups according to the response: patients with sustained response (SR) (line with square on graph) defined by the normalization of serum ALT and the absence of detectable serum HCV RNA during the treatment and after 6-months post-treatment follow-up period; patients with response and relapse (RR) (line with circle on graph) defined by the normalization of serum ALT and nondetectable serum HCV RNA at the end of the treatment but with an increase of serum ALT and presence of HCV RNA occurring at the follow up point (6 months after the end of treatment); and patients with no response (NR) (line with pentagon on graph) defined by elevated serum ALT at the end of the treatment. Patients were sampled before the treatment (time 0), at the end of the treatment (time 1), and at the follow-up point, 6 months after treatment (time 2). The values of cytokines are given in picograms/mL ⫾ S.D. The profile of TNF ␣ was characteristic for each one of the groups. At time 0, the SRs had significantly lower levels than NRs and RRs (p ⬍ 0.05). The level of TNF ␣ in the SRs was lowered significantly (at time 1) versus the baseline value (p ⬍ 0.05), and continued to drop 6 months after the treatment (p ⬍ 0.05). At the end of treatment, TNF ␣ levels in the RR group, were lower (p ⬍ 0.05) than at the pretreatment period. At the follow-up time, however, they came almost to the pretreatment values, the levels being (p ⬍ 0.05) higher than at the end of treatment. TNF ␣ in NR remained unchanged during the entire period of treatment and follow-up.

0.05). No significant changes were observed in NRs. IL 12 levels (Figure 6) in the SR group decreased at the end of treatment and remained significantly decreased versus entry levels (p ⬍ 0.05). Discussion In the present study we found increased levels of proinflammatory cytokines in the serum of noncirrhotic patients with chronic hepatitis C compared to normal healthy volunteers and patients with hypersensitivity syndrome reactions. Inflammatory cytokines are known to be higher in HCV infected 542

individuals than in normal individuals (42– 47). Also the patients with alcoholic liver disease are known to have high levels of proinflammatory cytokines (48,49). In the present study when compared to ALD subjects, HCV infected subjects showed higher level of the chemokine IL 8 and TNF ␣, but no significant differences in IL 6 or IL 12 (Figures 1 and 2). To our knowledge, this is the first time that a comparison of cytokine profiles of chronic hepatitis patients, not known to be alcoholics, has been recorded and compared to patients with known alcoholism (ALD but not HCV infected). An interesting finding is that the

Figure 4 —Effect of interferon ␣ (IFN ␣) therapy on IL 6 levels in chronic hepatitis C patients. The chronic hepatitis C patients were separated into three groups according to the response: SR (line with square on graph), RR (line with circle on graph), and NR patients (line with pentagon on graph). Patients were sampled before the treatment (time 0), at the end of the treatment (time 1), and at the follow-up point, 6 months after treatment (time 2). The values of cytokines are given in picograms/mL ⫾ S.D. The profile of IL 6 was characteristic for each one of the groups. No significant changes between IL 6 levels at time 0 were seen between the groups (SR, NR, and RR). At the end of treatment, IL-6 levels in the SR group were lower (p ⬍ 0.05) than at the pretreatment period, and remained unchanged at the follow-up. In RR and NR patients, the values did not change during the entire treatment and follow-up period. CLINICAL BIOCHEMISTRY, VOLUME 32, OCTOBER 1999

INFLAMMATORY CYTOKINES IN CHRONIC HEPATITIS C

Figure 5—Effect of interferon ␣ (IFN ␣) therapy on IL 8 levels in chronic hepatitis C patients. Patients were separated into three groups according to the response to IFN ␣: SR (line with square on graph), RR (line with circle on graph), and NR patients (line with pentagon on graph). Patients were sampled before the treatment (time 0), at the end of the treatment (time 1), and at the follow-up point, 6 months after treatment (time 2). The values of cytokines are given in picograms/mL ⫾ S.D. At time 0, IL 8 levels were significantly higher in the RR population when compared to the other sub-groups (NR and SR). At the end of treatment, IL 8 levels in SR group was higher than time 0, and remained at the same level at the follow-up time. In RR patients IL-8 levels lowered at the end of the treatment and continued to remain significantly lower (p ⬍ 0.05). In NR the levels of IL 8 remained unchanged during the entire period.

profile of each one of the cytokines was different for the HCV patients during the time of the disease and in response to IFN therapy. Our results on TGF ␤ 1 are consistent with those observed in a semi-quantitative assessment of the expression of TGF ␤ 1 in liver biopsy specimens from patients with chronic hepatitis C, after treatment with IFN ␣-2b (3 MU TTW). This study reported decreased expression of TGF ␤ 1 in biopsy specimens from six of seven patients (50). The reduction in TGF ␤ significantly correlated with the degree of inflammation and necrosis in both their experience and in our study. The serum levels of TGF ␤ 1 correlated with ALT levels in our study, while this was not seen in some other studies (50,51). Numerous studies have looked at the influence of IFN ␣ monotherapy on cytokine levels (52–58). Analysis of the disease progression in the cohort of patients presented in this study is undoubtedly influenced by the homogeneous nature (Caucasian, urban, good nutritional status) and the narrow histological spread of the disease severity to date (noncirrhotic with moderate to mild histologic activity index) (59,60). The apparent conflict with some of the previous reported studies may in part reflect: (i) the heterogeneity of patient populations with respect to age, CLINICAL BIOCHEMISTRY, VOLUME 32, OCTOBER 1999

Figure 6 —Effect of interferon ␣ (IFN ␣) therapy on IL 12 levels in chronic hepatitis C patients. Patients were separated into three groups according to the response to IFN ␣: SR (line with square on graph), RR (line with circle on graph), and NR patients (line with pentagon on graph). Patients were sampled before the treatment (time 0), at the end of the treatment (time 1), and at the follow-up point, 6 months after treatment (time 2). The values of cytokines are given in picograms/mL ⫾ S.D. At time 0, there was no significant differences between IL 12 levels in all groups of patients. At follow-up, SR was significantly lower (p ⬍ 0.05) than the baseline levels, when compared to RRs and Nrs.

gender, severity of the disease, viral type/subtype; (ii) the use of different quantitative methods and kits to assess serum cytokines; and (iii) the use of different methodologies to collect and to store samples until complete analysis. In chronic hepatitis C patients, the predictive value of TNF ␣ levels assessment for the outcome of therapy has been reported previously by us (46). This study clearly reports that in our population, lower levels of TNF ␣ at pretreatment have a high predictive value for sustained response. Higher levels of TNF ␣ correlate with a response rate to the IFN ␣ therapy in these patients. As shown in Figure 3, the TNF ␣ profile has a different rate of change for each response. The rate of change of TNF ␣ in all 3 groups correlates significantly with the ALT rate of change (Table 3). Introducing our observation into clinical practice might change the way to administer therapy. In the patients with a high initial TNF ␣ levels that drops with the therapy (RR patients), continuing the therapy for a longer period might shift their response to SR. However, in the patients that showed high initial TNF ␣ levels and no changes during the therapy (NR), the chances of success if treatment is continued are lower than 5%, therefore the TNF ␣ profile in these patients might indicate that an alternative treatment should be considered. The profile of IL 6 (Figure 4) and of IL 12 (Figure 6) following IFN ␣ treatment in SR suggests they are good markers of liver inflammation and appear to be involved in termination of the viral infection. 543

NEUMAN

The rather static serum cytokine profile observed in NRs in this study, may be explained by the failure of INF ␣ to modulate their response to aid the eradication of HCV chronic infection. Early treatment in the acute phase of disease with agents that stimulate cytotoxic immune type 1 responses may lead to eradication of HCV infection as suggested also by Shapiro et al. (57). Our present findings show the importance of IL 8 assessment in establishing the cytokine profile characteristic for different responses to therapy. The high IL 8 levels in SR patients (Figure 5) may suggest that the higher the chemokine attractant properties, the lower the virus-infected cells, and, therefore, the higher efficacy of treatment. Therefore, we conclude that measuring TNF ␣ levels at base-line has a high predictive value for the outcome of IFN␣ therapy, and we suggest that cytokine measurement should be used as a management tool in deciding with whom to continue standard treatment, whom to adjust it, or whom to consider other therapeutic options. References 1. Sherlock S. Antiviral therapy for chronic hepatitis C viral infection. J Hepatol 1995; 23(Suppl 2): 3–7. 2. Fried MW, Hoofnagle JH. Therapy of hepatitis C. Semin Liver Dis 1995; 15: 82–91. 3. Marcellin PB. Treatment of chronic viral hepatitis. Clin Gastroenterol 1994; 8: 233–53. 4. Marcellin P, Martinot M, Boyer N, Castelnau C, Benhamou JP, Erlinger S. Factors of response to treatment with interferon-alpha in patients with chronic hepatitis C. Therapeutic perspectives. Gastroenterol Clin Biol 1995; 19: 991–1000. 5. EASL International Consensus Conference on Hepatitis C (Paris 26 –28 February 1999) Consensus Statement. J Hepatol 1999; 30: 956 – 61. 6. Christian MD, Sachez RL, Soloway RD. Injection site vasculitis in patient receiving interferon alpha for chronic hepatitis C. J Am Acad Dermatol 1997; 37: 118 –20. 7. Coroneos E, Truong L, Olivero J. Fibrillary glomerulonephritis associated with hepatitis C viral infection. Am J Kidney Dis 1997; 29: 132–5. 8. Czaja AJ. Extrahepatic immunologic features of chronic viral hepatitis. Dig Dis 1997; 15: 125– 44. 9. Falaschi P, D’Urso R, Proietti A, Martocchia A, Pastore R, Angelucci L. Effect of r-interferon alpha administration on hypothalamus-pituitary-thyroid axis in chronic hepatitis. Life Sci 1997; 60: 43–50. 10. Hadziyannis SK. Diabetes mellitus and chronic hepatitis C virus infection. Hepatol 1999; 29: 604 –5. 11. Hadziyannis SJ. The spectrum of extrahepatic manifestations in hepatitis C virus infection. J Viral Hepat 1997; 4: 9 –28. 12. Henrion JM, Colin L, Schapira M, Delannoy A, Heller FR. Hypoxic hepatitis caused by acute exacerbation of chronic respiratory failure: a case-controlled, hemodynamic study of 17 consecutive cases. Hepatology 1999; 29: 427–33. 13. Cooreman MP, Stadhouders P. Optimal antiviral therapy of chronic hepatitis caused by hepatitis C virus. Orv Hetil 1997; 138(Suppl 1): 1476 – 82. 14. Di Bisceglie AM, Hoofnagle JH, Krawczynski K. 544

ET AL.

15.

16. 17.

18. 19.

20.

21.

22.

23. 24. 25. 26.

27.

28.

29.

30.

31. 32.

Changes in hepatitis C virus antigen in liver with antiviral therapy. Gastroenterology 1993; 105: 858 – 62. Di Marco V, Lo Iacono O, Camma C, et al. A randomized controlled trial of high-dose maintenance interferon therapy in chronic hepatitis C. J Med Virol 1997; 51: 17–24. Dusheiko GM, Zuckerman AJ. Treatment of chronic viral hepatitis C. J Antimicrob Chemother 1993; 32(Suppl A): 107–20. Fargion S, Fracanzani AL, Sampietro M, et al. Liver iron influences response to interferon alpha therapy in chronic hepatitis C. Eur J Gastroenterol 1997; 9: 497–503. Ferenci P. Historical treatment of chronic hepatitis B and chronic hepatitis C. Gut 1993; 34(Suppl 2): S69 – S73. Imai Y, Kawata S, Tamura S, et al. Recombinant interferon-alpha-2a for treatment of chronic hepatitis C: results of a multicenter randomized controlled dose study. Liver 1997; 17: 88 –92. Piresi M, Fabris C, Toniutto P, et al. Endogenous interferon-alpha concentration and outcome of interferon treatment inpatients with chronic hepatitis C. Dig Dis Sci 1997; 42: 767–71. Bonis PA, Cappelleri JC, Kaplan MM, Lau J. Correlation of biochemical response to interferon alpha with histological improvement in hepatitis C: A metaanalysis of diagnostic test characteristics. Hepatology 1997; 26: 1035– 44. Camma C, Giunta M, Linea C, Pagliaro L. The effect of interferon on the liver in chronic hepatitis C: a quantitative evaluation of histology by meta-analysis. J Hepatol 1997; 26: 1187–99. Schalm SW. Antiviral therapy of hepatitis C. Scand J Gastroenterol 1997; Suppl 223: 46 –9. Simpson KJ, Lukacs NW, Coletti L, Strieter RM, Kunkel SL. Cytokines in the liver. J Hepatol 1997; 27: 1120 –32. Neuman MG. Cytokines and Inflamed Liver. Clin Invest Med 1998; 23: 1– 6. Neuman MG, Shear NH, Bellentani S, Tiribelli C. Role of cytokines in ethanol-induced hepatocytotoxicity in Hep G2 cells. Gastroenterology 1998; 114: 157– 69. Zhang ZX, Milich DR, Peterson DL, et al. Interferonalpha treatment induces delayed CD4 proliferative responses to the hepatitis C virus nonstructural protein 3 regardless of the outcome of therapy. J Infect Dis, 1997; 175: 1294 –301. Iacovacci SM, Barca S, Sargiacomo M, et al. Molecular characterization and dynamics of hepatitis C virus replication in human fetal hepatocyets infected in vitro. Hepatology 1997; 26: 1328 –37. Imada KF, Koyama Y, Nakano I, Yamada M, Katano Y, Hayakawa T. Naive and memory T cell infiltrates in chronic hepatitis C: phenotypic changes with interferon treatment. Clin Exp Immunol 1997; 109: 59 – 66. Cacciarelli TV, Martinez OM, Gish RG, Villanueva JC, Krams SM. Immunoregulatory cytokines in chronic hepatitis C virus infection: pre- and posttreatment with interferon alfa. Hepatology 1996; 24: 6 –9. Itoh Y. Monokine production by peripheral whole blood in chronic hepatitis C patients treated with interferon. Dig Dis Sci 1995; 40: 2423–30. Carreno V, Parra A, Navas S, Quiroga JA. Granulocyte-macrophage colony-stimulating factor as adjuvant therapy for factor as adjuvant therapy for inter-

CLINICAL BIOCHEMISTRY, VOLUME 32, OCTOBER 1999

INFLAMMATORY CYTOKINES IN CHRONIC HEPATITIS C

33.

34.

35.

36.

37.

38.

39.

40.

41. 42.

43.

44.

45.

46.

47.

feron alpha treatment of chronic hepatitis C. Cytokine 1996; 8: 318 –22. Carrerno V. Biological properties of interleukin-12 and its therapeutic use in persistent hepatitis B virus and hepatitis C virus infection. J Viral Hepatol 1997; 4(Suppl 2): 83– 6. Kakumu SO, Ishikawa T, Iwata K, Yano M, Yoshioka K. Production of interleukins 10 and 12 by peripheral blood mononuclear cells (PBMC) in chronic hepatitis C virus (HCV) infection. Clin Exp Immunol 1997; 108: 138 – 43. Kishihara Y, Hayashi J, Yoshimura E, Yamaji K, Nakashima K, Kashiwagi S. IL-1 beta and TNF-alpha produced by peripheral blood mononuclear cells before and during interferon therapy in patients with chronic hepatitis C. Dig Dis Sci 1996; 41: 315–21. Kishihara Y. IL-1 beta an dTNF-alpha produced by peripheral blood mononuclear cells before and during interferon therapy in patients with chronic hepatitis C. Dig Dis Sci 1996; 41: 315–21. Kuzushita N. High levels of serum interlukin-10 are associated with a poor response to interferon treatment in patients with chronic hepatitis C. Scand J Gastroenterol 1997; 32: 169 –74. Quiroga JA. Serum levels of soluble immune factors and pathogenesis of chronic hepatitis C, and their relation to therapeutic response to interferon-alpha. Dig Dis Sci 1994; 39: 2485–96. Lohr HF. Liver-infiltrating and circulating CD4⫹ T cells in chronic hepatitis C: immunodominant epitopes, HLA-restriction and functional significance. Liver 1996; 16: 174 – 82. Hino K, Sainokaimi S, Shimoda K, et al. Genotypes and titers of hepatitis C virus for predicting response to interferon therapy in patients with chronic hepatitis C. J Med Virol 1994; 42: 229 –305. Ruiz-Arguelles GJ. Laboratory measurement of human cytokines. JIFCC 1995; 7: 12–15. Malaguarnera M, DiFazio I, Romeo MA, Restuccia S, Laurino A, Trovato BA. Elevation of IL6 levels in patients with chronic hepatitis due to hepatitis C virus. J Gastroenterol 1997; 32: 211–15. Neuman MG, Benhamou JP, Shear NH, et al. Interleukine 6 normalization during interferon-ribavirin combination therapy in chronic hepatitis C. J Hepatol 1999c; 30(Suppl 1) (PC06/49): 242. Martin J. Recombinant human granulocyte colonystimulating factor reduces hepatitis C virus replication in mononuclear cells from chronic hepatitis C patients. Cytokine 1996; 8: 313–17. Quiroga JA. Induction of interlekin-12 production in chronic hepatis C virus infection correlates with the hepatocellular damage. J Infect Dis 19998; 178: 247– 51. Neuman MG, Benhamou JP, Shear NH, et al. Serum tumor necrosis factor as a predictor of sustain response to alpha-interferon-ribavirin combination therapy in chronic hepatitis C. J Hepatol 1999; 30(Suppl 1) (PC06/109): 257. Zylberberg HR, Pol S, Masson A, et al. Soluble tumor necrosis factors in chronic hepatitis C: a correlation with histological fibrosis and activity. J Hepatol 1999; 30: 185–91.

CLINICAL BIOCHEMISTRY, VOLUME 32, OCTOBER 1999

48. Khoruts A, Stanke L, McClain CF, Logan G, Allen JI. Circulating tumor necrosis factor, interleukin-1 and interleukin 6 concentrations in chronic alcoholic patients. Hepatology 1991; 13: 267–76. 49. Sheron N, Bird G, Goka J, Alexander G, Wiliiams IL. Elevated plasma interleukin-6 and increased severity and mortality in alcoholic hepatitis. Clin Exp Immunol 1991; 84: 453–99. 50. Bedossa P, Poynard T, Mathurin P, Lemaigre G, Chaput JC. Transforming growth factor beta 1: in situ expression in the liver of patients with chronic hepatitis C treated with alpha interferon. Gut 1993; 34: S147– 8. 51. Roulot D, Durand H, Coste T, Rautureau J, Strosberg AD, Benarous R, Marullo S. Quantitative analysis of transforming growth factor beta 1 messenger RNA in the liver of patients with chronic hepatitis C: absence of correlation between high levels and severity of disease. Hepatology 1995; 21: 298 –304. 52. Kishihara Y, Hayashi J, Yoshimura E, Yamaji K, Nakashima K, Kashiwagi S. IL-1 beta and TNF-alpha produced by peripheral blood mononuclear cells before and during interferon therapy in patients with chronic hepatitis C. Dig Dis Sci 1996; 41: 315–21. 53. Fabris C, Soardo G, Falleti E, et al. Relationship among hepatic inflammatory changes, circulating levels of cytokines, and response to IFN-alpha in chronic hepatitis C. J Interferon Cytokine Res 1998; 18: 705–9. 54. Fukuda R, Ishimura N, Ishihara S, et al. Intrahepatic expression of pro-inflammatory cytokine mRNAs and interferon efficacy in chronic hepatitis C. Liver 1996; 16: 330 –9. 55. Larrea E, Garcia N, Qian C, Civeira MP, Prieto J. Tumor necrosis factor alpha gene expression and the response to interferon in chronic hepatitis C. Hepatology 1996; 23: 210 –17. 56. Malaguarnera M, De Fazio I, Laurino A, Ferlito L, Romano M, Trovato BA. Serum interleukin 6 concentrations in chronic hepatitis C patients before and after interferon-alpha treatment. Int J Clin Pharmacol Ther 1997; 35: 385– 8. 57. Shapiro S, Gershtein V, Elias N, Zuckerman E, Salman N, Lahat N. mRNA cytokine profile in peripheral blood cells from chronic hepatitis C virus (HCV)infected patients: effects of interferon-alpha (IFNalpha) treatment. Clin Exp Immunol 1998; 114: 55– 60. 58. Davis GL, Esteban-Mur R, Rustgi V, et al. Interferon alfa-2b alone or in combination with ribavirin for the treatment of relapse of chronic hepatitis C. N Engl J Med 1998; 339: 1493–9. 59. Kobayashi K, Ishii M, Igarashi T, et al. Profiles of cytokines produced by CD4-positive T lymphocytes stimulated by anti-CD3 antibody in patients with chronic hepatitis C. J Gastroenterol 1998; 33: 500 –7. 60. Tsushima H, Kawata S, Tamura S, et al. Reduced plasma transforming growth factor-beta1 levels in patients with chronic hepatitis C after interferonalpha therapy: association with regression of hepatic fibrosis. J Hepatol 1999; 30: 1–7.

545