In vitro effect of indomethacin and interferon-α on Th1 and Th2 cytokine synthesis in patients with chronic hepatitis C

www.elsevier.com/locate/jnlabr/ycyto Cytokine 26 (2004) 95e101

In vitro effect of indomethacin and interferon-a on Th1 and Th2 cytokine synthesis in patients with chronic hepatitis C Pietro Andreone), Annagiulia Gramenzi, Elisabetta Loggi, Laura Favarelli, Carmela Cursaro, Marzia Margotti, Maurizio Biselli, Stefania Lorenzini, Mauro Bernardi Semeiotica Medica, Dipartimento di Medicina Interna, Cardioangiologia ed Epatologia, Universita` di Bologna, Via Massarenti, 9, 40138 Bologna, Italy Received 10 April 2003; received in revised form 16 August 2003; accepted 25 August 2003

Abstract Current evidences suggest that non-steroidal anti-inflammatory drugs could enhance the antiviral activity of interferon-a in chronic HCV infection. In this study, we investigated the effect of indomethacin, a non-steroidal anti-inflammatory drug, and interferon-a on cytokine production by peripheral blood mononuclear cells from 12 untreated patients with chronic hepatitis C. We evaluated the effect of incubation with indomethacin, interferon-a or both on synthesis of Th1- (interleukin-2, interferon-g) and Th2-associated cytokines (interleukin-4, interleukin-10), and of the antiviral protein 2#,5#-oligoadenylate synthetase. Interferon-a induced a significant increase in production of interleukin-2. Smaller increases were also seen in the presence of indomethacin, while incubation with both indomethacin and interferon-a leads to a synergistic effect. Incubation with indomethacin decreased both interleukin-4 and interleukin-10, whereas interferon-a increased these cytokines. The addition of indomethacin to interferon-a significantly reversed this interferon-induced increase. Finally, both indomethacin and the association interferon-a plus indomethacin determined a significant increase in 2#,5#-oligoadenylate synthetase production compared to both baseline and interferon-a alone. In conclusion, indomethacin was able to enhance the antiviral activity of interferon-a and to modulate the interferon-induced Th1 and Th2 cytokine response by increasing the Th1-response, fundamental for sustained clearance of HCV, and by decreasing the Th-2 type response, associated with HCV persistence. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Hepatitis C; Immune response; Indomethacin; Interferon-a; Th1; Th2

1. Introduction Infection by hepatitis C virus (HCV) is very common and can induce insidious and progressive liver damage in many chronically infected patients. Disease evolution is extremely variable, from persistent mild liver injury to rapid progression to cirrhosis [1]. The pathogenetic mechanisms responsible for viral persistence and liver damage are currently unknown, although recent studies

) Corresponding author. Tel.: +39-051-6363618; fax: +39-051340877. E-mail address: [email protected] (P. Andreone). 1043-4666/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.cyto.2003.08.014

suggested that the host immune response might play an important role. The host immune response is a complex event, which requires the coordinated activity of different effector arms. One of them involves the specific profile of Th1 and Th2 lymphocyte polarization, which is characterized by distinct and mutually exclusive patterns of cytokine production with different functions. Th1 cells secrete interleukin-2 (IL-2) and interferon-g (IFN-g), which promote cell-mediated immunity and are implicated in the protection against intracellular pathogens [2]. Th2 cells produce IL-4, IL-5 and IL-10, which regulate antibody production by B cells and play a prominent role in anti-helminth and allergic responses [2].

96

P. Andreone et al. / Cytokine 26 (2004) 95e101

It has been recently suggested that an imbalance between Th1 and Th2 response could influence the clinical outcome of HCV infection by modulating hepatocellular damage and disease progression. Several lines of evidence [3e5] indicate that viral clearance is associated with a prevalent Th1 response, whereas chronicization of HCV infection and disease progression are more likely in cases with a prevalent Th2 profile. Pegylated interferon (IFN) plus ribavirin [6] is currently the treatment of choice for chronic HCV infection. However, this regimen fails to obtain a long-term response in 50e60% of patients. The molecular and immunological basis for IFN-a resistance is unknown at present. To increase the efficacy of IFN-a, alternative treatments for chronic hepatitis C have been proposed, including the association of IFN-a with non-steroidal anti-inflammatory drugs (NSAIDs). However, a rational approach to combination therapies needs a detailed knowledge of the mechanism through which the different drugs affect the IFN-a intracellular signalling and mechanism of action. Some in vitro and in vivo studies showed that NSAIDs could enhance the antiviral activity of IFN-a [7e11] by potentiating IFN-dependent gene activation via the inhibition of arachidonic acid metabolism. In this study, we investigated the effects of indomethacin, an NSAID that selectively blocks cyclooxygenase 1 and 2, and IFN-a, both alone and in combination, on the synthesis of 2#,5#-oligoadenylate synthetase (2#,5#-OAS, an IFN-induced protein with antiviral activity), and the production of Th1 and Th2 cytokines by peripheral blood mononuclear cells (PBMCs) from naive patients with chronic hepatitis C. 2. Materials and methods 2.1. Patients Twelve patients (nine males, median age 32.5 years, range 21e50) with histologically proven chronic hepa-

titis C, attending the outpatient clinic at our department, were enrolled in this study. Inclusion criteria were: abnormal alanine aminotransferase (ALT) levels; antiHCV antibodies detected by third generation ELISA (Ortho Diagnostic Systems; Raritan, NJ, USA); detectable serum HCV-RNA by a polymerase chain reaction (PCR) assay (Cobas Amplicor HCV Test, Roche Diagnostic System, Italy); and liver biopsy findings compatible with chronic hepatitis C. All patients were seronegative for hepatitis B virus or HIV markers. Other possible causes of chronic hepatitis (i.e. drugs, alcohol abuse, autoimmunity and metabolic disorders) were excluded by a detailed patient history and appropriate laboratory tests. Patients with intravenous drug use within one year of study entry were also excluded. All patients had never been treated with IFN-a and/or with any other antiviral treatment. The study was carried out according to the Helsinki Declaration and informed consent was obtained from each individual. The patient characteristics are listed in Table 1. 2.2. Preparation of cell culture Peripheral venous blood was collected from each patient into sterile heparinized tubes. All the samples were diluted 1 : 1 with normal saline; each suspension was layered on Histopaque-1077 (Sigma; St Louis, MO, USA) and then centrifuged at 750 G for 15 min at room temperature. The mononuclear cells at the interface between the plasma and the Histopaque were harvested and centrifuged for 10 min at 500 G after the addition of 20 mL of RPMI containing 10% heat inactivated fetal calf serum (Gibco BRL, Life Technologies; European Division, Milan, Italy) and glutamine (200 mM, Gibco BRL, Life Technologies; European Division, Milan, Italy), and tested for viability with trypan blue. The pellet was diluted with 1 mL of culture medium and the mononuclear cells were counted in a Neubauer counting

Table 1 Main characteristics and laboratory features of patients under study Case

Sex

Age

Source of infection

ALT (U/L)

HCV-RNA (!106 genEq/mL)

Genotype

1 2 3 4 5 6 7 8 9 10 11 12

F F M M M M M M M F M M

33 50 36 29 23 25 33 21 32 22 49 37

Blood transfusion Community acquired IVDU Community acquired IVDU IVDU IVDU IVDU Community acquired IVDU Community acquired Community acquired

65 57 274 51 219 217 142 107 94 100 98 213

1.05 1 6.60 0.50 2.50 12.37 3.80 0.23 15.06 0.5 13.1 3.87

1b 1b 1a 2a 4 1a 3 3a 1b 1b 2a 1b

IVDU: past history of intravenous drug use; HCV-RNA was quantified by the branched DNA signal amplification assay (Quantiplex 2.0; Chiron Corporation, Emeryville, CA, USA).

P. Andreone et al. / Cytokine 26 (2004) 95e101

chamber with a reverse phase-contrast microscope (Zeiss Munich, Germany). The cells were transferred to sterile test tubes at a concentration of 1!106 =mL to obtain liquid cultures. PBMCs were cultured in 24-well plates at the concentration of 1!106 =mL in RPMI 1640 supplemented with 10% heat-inactivated fetal calf serum. The cells from each patient were subjected to eight different experimental conditions: four in the presence and four in the absence of concanavalin A (ConA) at 10 mg/mL (Sigma; St Louis, MO, USA). As we previously reported [12], concanavalin A was used as a polyclonal stimulus for triggering cytokine production. The incubation conditions were as follows: culture medium alone (RPMI), condition defined as baseline; culture medium containing 500 U/mL of IFN-a (Wellferon, Glaxo Wellcome; Verona, Italy); culture medium containing indomethacin (Liometacen, Chiesi; Parma, Italy) at the concentration of 5 mmol/L; culture medium containing both IFN-a and indomethacin. The cell cultures were stored for 24 h at 37 (C in a moist chamber with an atmosphere containing 5% carbon dioxide. After 24 h, the supernatants were harvested, centrifuged at 500 G for 10 min, separated from the cells, divided into aliquots, and immediately preserved at
80 (C for subsequent testing. The production of IL-2, IFN-g, IL-4, and IL-10 was determined by the immunoadsorbent assay ELISA (for IFN-g, IL-4, IL-10, R & D System; Minneapolis, MN, USA, and for IL-2, Endogen; Woburn, MA, USA) using the supernatants harvested from the cell cultures stimulated with ConA. Each sample was tested in duplicate and the values determined from the standard curve. The values were accepted only if they plotted into the standard curves. In the other cases, they were diluted according to the manufacturer’s instruction and the allowed tolerance was !10%.

97

The limits of sensitivity of the assay were from 6 to 2000 pg/mL for IL-2; from 3 to 1000 pg/mL for IFN-g; from 2 to 500 pg/mL for IL-10 and from 1.5 to 500 pg/mL for IL-4. The activity of 2#,5#-OAS was determined in the supernatants of the cell cultures incubated in the absence of ConA, using a radioimmunological assay (Eiken Chemical, Tokyo, Japan). The sensitivity of the method was 10e810 pmol/dL. All reagents used for culture in the present study were lipopolysaccharide-free. 2.3. Statistical analysis The data, expressed as mean G standard error of the mean were analysed using non-parametric statistical tests. In particular, the comparison between the average concentrations of cytokines was evaluated by Wilcoxon’s test for paired data. The correlation between variables was calculated using Spearman’s correlation coefficient. A value of p!0:05 was considered statistically significant. Data processing was carried out with the SPSS statistical package for Windows.

3. Results 3.1. Th1 cytokines The production of IL-2 from PBMCs is shown in Fig. 1. Compared to the baseline conditions, the addiction of IFN-a significantly increased the synthesis of IL-2 (175G43 vs 237G41 pg/mL, p ¼ 0:03). The addition of indomethacin also induced an increase in IL-2 synthesis (198G34 pg/mL), although not significantly different compared to baseline. The association between IFN-a and indomethacin produced a synergistic effect on IL-2 production. This further increase was significantly

pg/mL x 106 cells

900

600

300

0

BASELINE

IFN

INDO

IFN+INDO

Fig. 1. IL-2 production by PBMCs. The bold line is the median of values (baseline vs IFN: p ¼ 0:03; baseline vs IFN + INDO: p ¼ 0:004; IFN vs IFN + INDO: p ¼ 0:004; INDO vs IFN + INDO: p ¼ 0:002).

98

P. Andreone et al. / Cytokine 26 (2004) 95e101

7500

pg/mL x 106 cells

6000 4500 3000 1500 0

BASELINE

IFN

INDO

IFN+INDO

Fig. 2. IFN-g production by PBMCs. The bold line is the median of values (baseline vs IFN: p ¼ 0:006; IFN vs IFN + INDO: p ¼ 0:01).

higher not only compared to baseline (568G66 vs 175G43 pg/mL, p ¼ 0:004) but also to both IFN-a alone (p ¼ 0:004) and indomethacin alone (p ¼ 0:002). We found a significant negative correlation between ALT levels and IL-2 production by PBMCs only at baseline (r ¼
0:63; p ¼ 0:03). No correlation was found between HCV-RNA serum levels and IL-2 cytokine synthesis. Compared to baseline, a significant increase in IFN-g production was observed following the addition of IFN-a (2818G466 vs 3573G330 pg/mL; p ¼ 0:006). Although both indomethacin and the association of IFN-a plus indomethacin also induced an increase in IFN-g production (indomethacin: 3107G547 pg/mL; IFN-a plus indomethacin: 3062G416 pg/mL) the difference with respect to baseline was not statistically significant in either case (Fig. 2). No correlation was found between ALT or HCV-RNA serum levels and IFN-g production. 3.2. Th2 cytokine Fig. 3 shows the changes in the synthesis of IL-4. Compared to baseline, only the addition of IFN-a

determined a significant increase of IL-4 synthesis (76G16 vs 104G16 pg/mL, p ¼ 0:03), while neither indomethacin nor IFN-a plus indomethacin led to significant changes (indomethacin: 76G13 pg/mL; IFN-a plus indomethacin: 96G20 pg/mL). However, the levels of IL-4 in the presence of IFN-a were significantly higher if compared with those observed in the presence of indomethacin (104G16 vs 76G13 pg/mL, p ¼ 0:02). The production of IL10 induced by IFN-a (1764G 189 pg/mL) was significantly higher compared to all the other experimental conditions (Fig. 4). Indomethacin alone induced a significant decrease of IL-10 production compared to both baseline (954G100 vs 1262G150 pg/ mL, p ¼ 0:02) and IFN-a alone (954G100 vs 1764G 189 pg/mL; p ¼ 0:004). The association of IFN-a plus indomethacin had no effect compared to both baseline (1218G136 vs 1262G150 pg/mL; p ¼ not significant) and indomethacin (1218G136 vs 954G100 pg/mL; p ¼ not significant), but was significantly lower compared to IFN-a alone (1218G136 vs 1764G189 pg/mL; p ¼ 0:02).

250

pg/mL x 106 cells

200 150 100 50 0

BASELINE

IFN

INDO

IFN+INDO

Fig. 3. IL-4 production by PBMCs. The bold line is the median of values (baseline vs IFN: p ¼ 0:03; IFN vs INDO: p ¼ 0:02).

P. Andreone et al. / Cytokine 26 (2004) 95e101

99

2500

pg/mL x 106 cells

2000 1500 1000 500 0

BASELINE

IFN

INDO

IFN+INDO

Fig. 4. IL-10 production by PBMCs. The bold line is the median of values (baseline vs IFN: p ¼ 0:03; baseline vs INDO: p ¼ 0:02; IFN vs INDO: p ¼ 0:004; IFN vs IFN + INDO: p ¼ 0:02).

No correlation was found between ALT or HCVRNA serum levels and Th2 cytokine synthesis by PBMCs. 3.3. Antiviral protein 2#,5#-OAS Compared to baseline (110G32), the synthesis of 2#,5#-OAS was significantly increased in all the experimental conditions (Fig. 5). Both indomethacin and the association IFN-a plus indomethacin determined a significant increase in 2#,5#-OAS synthesis compared to IFN-a alone (indomethacin: 3186G1148 pmol/dL vs IFN-a: 833G389 pmol/dL, p ¼ 0:004; IFN-a plus indomethacin: 5285G1806 pmol/dL vs IFN-a: 833G389 pmol/dL, p ¼ 0:008). No difference was found between indomethacin alone and the association IFN-a plus indomethacin (3186G1148 vs 5285G1806 pmol/dL, p ¼ not significant). No correlation was found between age, ALT levels or HCV-RNA titres and either baseline or stimulated 2#,5#-OAS.

4. Discussion The IFN-a pathway involves an IFN specific receptor on cell target surface, which is constitutively associated with tyrosine-kinase. During binding of IFN-a molecule to its receptor, the tyrosine-kinase is rapidly activated and results in the phosphorylation of several proteins acting as signal transducers and activators of transcription (STAT). The STAT proteins, in turn, form homoand heterodimer complexes which translocate to the nucleus to stimulate gene transcription via binding specific sequences, called the IFN stimulated response elements (ISREs) [13]. In 1991, Hannigan and Williams [14] demonstrated that metabolites of arachidonic acid generated by enzymes other than cyclo-oxygenase or lipoxygenase could serve as second messengers for the IFN-a pathway. In fact, they showed an enhanced production of 2#,5#-OAS, an IFN-induced protein with vigorous antiviral effects, by cultured fibroblasts incubated with inhibitors of the cyclo-oxygenase or the lipoxygenase pathway. We demonstrated that

pmol/ml x 106 cells

18000 15000 12000 9000 6000 3000 0

BASELINE

IFN

INDO

IFN+INDO

Fig. 5. 2#,5#-OAS production by PBMCs. The bold line is the median of values (baseline vs IFN: p ¼ 0:002; baseline vs INDO: p ¼ 0:002; baseline vs IFN + INDO: p ¼ 0:002; IFN vs INDO: p ¼ 0:004; IFN vs IFN + INDO: p ¼ 0:008).

100

P. Andreone et al. / Cytokine 26 (2004) 95e101

indomethacin, a potent and selective cyclo-oxygenase 1 and 2 (COX-1 and COX-2) inhibitor, is able to increase 2#,5#-OAS both in cultured liver tissue [8] and in sera of patients with chronic HCV and HBV infection [9]. Interestingly, our in vitro studies also showed a synergistic effect between IFN-a and indomethacin in increasing 2#,5#-OAS production. These effects of indomethacin are likely related to a significant increase of ISRE-dependent transcription through the stimulation of STAT 1 phosphorylation [7]. Based on this experimental background, indomethacin or other NSAIDs have been used in combination with IFN-a for the treatment of HCV chronic infection with conflicting results [11,15e18]. The reasons of such controversial results could be at least in part ascribed to the fact that clinical trials conducted so far are not homogenous in terms of enrolled patients (untreated and non-responders), treatment schedule and type of NSAID. Growing evidence suggests that the host immune response is a major determinant of both clinical outcome and response to antiviral treatment in chronic hepatitis C. In particular, an imbalance in Th1 and Th2 cytokines seems to play an important role, although this issue is not completely clarified [3]. To further investigate the interaction between IFN-a, indomethacin and the host immune response we evaluated the synthesis of Th1 and Th2 cytokines by PBMCs of naive patients with chronic hepatitis C. In this study, IFN-a was able to significantly increase IL-2 synthesis. Smaller increases were also seen after incubation with indomethacin. The new finding of the present study is that the incubation of PBMCs with a combination of IFN-a and indomethacin leads to a synergistic effect on the synthesis of this cytokine. It is well known that IL-2 activates NK cell cytotoxicity and promotes the differentiation and the proliferation of cytotoxic cells [19]. These actions, which represent the first step of host response to viral infections, seem to be needed for HCV eradication [3,19,20]. In the present study, indomethacin was also able to modify Th2 cytokine synthesis. In fact, IL-10 production by PBMCs decreased after incubation with the drug. On the contrary, IFN-a alone induced a significant increase of both IL-4 and IL-10. Interestingly, indomethacin in combination with IFN-a seems to inhibit the IFN-a-induced production of both IL-4 and IL-10. Therefore, in patients with chronic hepatitis C, indomethacin not only potentiates IFN-a-induced Th1 responses but also counteracts the IFN-a-induced Th2 response. Finally, confirming our previous findings [8,9], indomethacin promotes the antiviral activity of IFN-a, by enhancing 2#,5#-OAS production. The influence of cytokines on HCV clearance and the clinical outcome of infected patients is under debate. However, there is general agreement on the fact that for

unknown reasons in chronically infected patients a vigorous Th1 response within the liver does not lead to HCV elimination and, at the same time, promotes tissue damage via macrophage activation and cytotoxic T lymphocytes induction. On other hand, a persistent Th2 phenotype is detectable in peripheral blood [3e5]. The effect of IFN-a on the cytokine production is variable, as either a prevalent Th1 or Th2 profile can be induced: the former appears to be associated with response to treatment, while the latter is found in nonresponder patients [4]. In this study, we confirm our previous finding that IFN-a determined an increase of both Th1 and Th2 cytokine synthesis by PBMCs from chronic hepatitis C patients [12], which does not represent an optimal response to achieve the clearance of HCV infection. The enhanced expression of Th2 cytokines, a finding also reported by others [21], thus represents the limiting factor to IFN-a efficacy in this context. Interestingly, the addition of ribavirin to IFN-a seems to inhibit the IFN-a-induced production of Th2 cytokines [22], a finding we observed in the present study when PBMCs from HCV-infected patients were incubated with indomethacin plus IFN-a. However, the combination of indomethacin and IFN-a would lead to a more effective response, as indomethacin also potentiates the IFN-induced antiviral protein synthesis. In conclusion, our results not only confirm that NSAIDs can enhance the antiviral activity of IFN-a but also show that they are able to modulate the IFNinduced Th1 and Th2 cytokine responses. Thus, the synergic effect on 2#,5#-OAS production and the stimulation of Th1 cytokine response without a concomitant stimulation of Th2 could represent a further rationale for the combination treatment of IFN plus NSAIDs in chronic hepatitis C.

References [1] Alberti A, Chemello L, Benvegnu` L. Natural history of hepatitis C. J Hepatol 1999;31(Suppl 1):17e24. [2] Mosmann TR, Cherwinski H, Bond MW, Giedlin MA, Coffman RL. Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins. J Immunol 1986;136:2348e57. [3] Koziel MJ. Cytokines in viral hepatitis. Semin Liver Dis 1999; 2:157e68. [4] Tsai SL, Liaw YF, Chen MH, Huang CY, Kuo G. Detection of type 2-like helper cells in hepatitis C virus infection: implications for hepatitis C virus chronicity. Hepatology 1997;25:449e57. [5] Sarih M, Bouchrit N, Benslimane A. Different cytokine profiles of peripheral blood mononuclear cells from patients with persistent and self-limited hepatitis C virus infection. Immunol Lett 2000; 74:117e20. [6] Di Bisceglie AM, Hoofnagle JH. Optimal therapy of hepatitis C. Hepatology 2002;36(Suppl 1):s121e7. [7] Giambartolomei S, Artini M, Almerighi C, Moavero SA, Levrero M, Balsano C. Non steroidal anti-inflammatory drug metabolism

P. Andreone et al. / Cytokine 26 (2004) 95e101

[8]

[9]

[10]

[11]

[12]

[13] [14] [15]

potentiates interferon alpha signaling by increasing STAT1 phosphorylation. Hepatology 1999;30:510e6. Andreone P, Cursaro C, Gramenzi A, Buzzi A, Miniero R, Bernardi M, et al. Indomethacin increases 2#,5#-oligoadenylate synthetase release by cultured liver tissue of patients with HCV chronic active hepatitis. Int Hepatol Commun 1994;2:289e94. Andreone P, Cursaro C, Gramenzi A, Buzzi A, Miniero R, Sprovieri G, et al. Indomethacin enhances 2#,5#-oligoadenylate synthetase in patients hepatitis B and C virus chronic active hepatitis. J Hepatol 1994;21:984e8. Kapicioglu S, Sari M, Kaynar K, Baki A, Ozoran Y. The effect of indomethacin on hepatitis B virus replication in chronic healthy carriers. Scand J Gastroenterol 2000;35:957e9. Andreone P, Cursaro C, Gramenzi A, Fiorino S, Di Giammarino L, Miniero R, et al. Interferon alpha plus ketoprofen or interferon alpha plus ribavirin in chronic hepatitis C non-responders to interferon alpha alone: results of a pilot study. Ital J Gastroenterol Hepatol 1999;31:688e94. Andreone P, Cursaro C, Gramenzi A, Margotti M, Ferri E, Talarico S, et al. In vitro effect of thymosin-alpha1 and interferonalpha on Th1 and Th2 cytokine synthesis in patients with chronic hepatitis C. J Viral Hepat 2001;8:194e201. Platanias LC, Fish EN. Signaling pathways activated by interferons. Exp Hematol 1999;27:1583e92. Hannigan GE, Williams BRG. Signal transduction by interferond a through arachidonic acid metabolism. Science 1991;251:204e7. Zarski JP, Maynard-Muet M, Chousterman S, Baud M, Barnoud R, Abergel A, et al. Tenoxicam, a non-steroidal anti-inflammatory drug, is unable to increase the response rate in patients with

[16]

[17]

[18]

[19] [20]

[21]

[22]

101

chronic hepatitis C treated by alpha interferon. Hepatology 1998;27:862e7. Fabris P, Tositti G, Negro F, Marranconi F, Infantolino D, Rassu M, et al. Interferon alfa-2b alone or in combination with ketoprofen as treatment for interferon-naive chronic hepatitis C patients. Aliment Pharmacol Ther 1999;13:1329e34. Mun˜oz AE, Levi D, Podesta` A, Gorin JM, Gonzalez J, Bartellini MA, et al. Interferon-a2b combined with daily ketoprofen administration improves virological response in chronic hepatitis C: a prospective and rondomized trial. Gut 2000;46:427e31. Andreone P, Gramenzi A, Cursaro C, Biselli M, Lorenzini S, Loggi E, et al. Interferon-a combined with ketoprofen as treatment of naive patients with chronic hepatitis C: a randomized controlled trial. J Viral Hepat 2003;10:306e9. Biron CA. Cytokines in the generation of immune response to and resolution of virus infection. Curr Opin Immunol 1994;6:530e8. Kamal SM, Rasenack JW, Bianchi L, Al Tawil A, El Sayed Khalifa K, Peter T, et al. Acute hepatitis C without and with schistosomiasis: correlation with hepatitis C-specific CD4(C) T-cell and cytokine response. Gastroenterology 2001;121: 646e56. 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 treatment. Clin Exp Immunol 1998;114:55e60. Cramp ME, Rossol S, Chokshi S, Carucci P, Williams R, Naoumov NV. Hepatitis C virus-specific T-cell reactivity during interferon and ribavirin treatment in chronic hepatitis C. Gastroenterology 2000;118:346e55.