Pilot Study of Recombinant Human Granulocyte-macrophage Colony-stimulating Factor in the Treatment of Chronic

Hepatitis B

JULIO MARTIN, ORENCIO BOSCH, GLORIA MORALEDA, JAVIER BARTOLOME, JUAN ANTONIO QUIROGA AND

Hepatology Unit, Fundacidn Jime'nez Diaz, E-28040 Madrid, Spain VICENTE CARRERO

Recombinant human granulocyte-macrophage colo- ny-stimulating factor is being used to improve the immunological function of patients with various dis- eases and to ameliorate hematological disorders. We investigated the tolerance and possible antiviral effect of the administration of daily doses of recombinant human granulocyte-macrophage colony-stimulating factor (3, 1 or 0.5 pg/kg body wt) to nine patients with chronic hepatitis B, alone or in combination with 5 MU interferon-%,. Recombinant human GM-CSF reduced significantly (p < 0.02) hepatitis B virus DNA levels. The three doses used were equally effective. Of the eight patients who completed the study, four became negative for HBV DNA and HBeAg; two of them seroconverted to HBe antibody. These four patients showed improvement in the histological activity of their liver disease. Ultimately, two patients regained normal AlLT values. 2',5'-Oligoadenylate synthetase activity increased significantly (p < 0.01) in cell lysates of mononuclear cells cultured in uitro, coinciding with the reductions in hepatitis B virus DNA levels. Recom- binant human granulocyte-macrophage colony-stimu- lating factor was well tolerated but produced a dose- dependent increase in white blood cell counts. It became intolerable at doses of 3 pg (and was reduced to 1.5 pg); this effect was reversible after cessation of recombinant human granulocyte-macrophage colony- stimulating factor treatment. No remarkable varia- tions occurred in other parameters. In conclusion, recombinant human granulocyte-macrophage colony- stimulating factor administration is safe and tolerable at doses of 0.6 to 1 & k g body wt and may exert an antiviral effect in chronic hepatitis B. The efficacy of granulocyte-macrophage colony-stimulating factor should be considered in the prevention of hemato- logical disorders and in the increased susceptibility to bacterial infections, during interferon therapy in

Received January 11, 1993; accepted May 23, 1993. J. Martin is a fellowship of Fundacion Conchita Mbago. G. Moraleda is a

fellow of Ministerio de Educaci6n y Ciencia. J. Bartolome and J.A. Quiroga are recipients of research grants from Fundacion para el Estudio de las Hepatitis Vudes.

Address reprint requests to: Dr. Vicente Carreno, Hepatology Unit, Fun- dacion Jimenez Dim, Avda. Reyes Catblicos, 2, E-28040 Madrid, Spain.

Copyright 0 1993 by the American Association for the Study of Liver Diseases.

0270-9139/93 $1.00 + .I0 31/1/49223

chronic hepatitis B virus infection. (HEPATOLOGY 1993; 18:775-780.)

Chronic HBV infection is associated with persistent liver injury ranging from mild disease to cirrhosis and HCC (1). Patients with these manifestations of liver damage have substantial abnormalities of cell-mediated immunity and cytokine production (2), including lower- than-normal production of interferon-a (IFN-a) (3). In this context, the biological effects of macrophage acti- vation may be relevant to the modulation of the host immune response (4).

Among the various agents tried, IFN-a is the most promising for the treatment of chronic hepatitis B because it leads to significant reduction in HBV repli- cation, with approximately 30% to 40% of treated patients experiencing disease remission (compared with 5% of untreated patients) (5,6). In attempts to increase the response rate, IFN-a has been used in combination with other antiviral or immunomodulatory agents such as corticosteroids (71, adenine arabinoside 5' monophos- phate (B), acyclovir (9) or IFN-y (10). However, the combinations do not result in response rates higher than that achieved with IFN-a alone. Hematological side effects are frequent, and leucopenia is commonly ob- served during treatment. This prevents the use of interferon in patients who are infected with HBV with hematological disorders.

The granulocyte-macrophage colony-stimulating fac- tor (GM-CSF) is a glycoprotein produced by activated T lymphocytes, endothelial cells and fibroblasts that increases production of neutrophils, monocytes and eosinophils. GM-CSF increases antibody-mediated cyto- toxicity and regulates the liberation of tumoricidal cytokines (11, 12). It has been used to improve the immunological system in patients with aplastic ane- mia, patients treated with chemotherapy and as ad- juvant after bone marrow transplantation. GM-CSF has also been used in AIDS patients with promising results (13).

The aim of this pilot study was to assess the tolerance, effective dose and possible antiviral effects of recom- binant human GM-CSF in patients with chronic HBV infection.

775

776 MARTIN ET AL. HEPATOLOGY October 1993

TABLE 1. Pretreatment characteristics of nine patients with chronic hepatitis B positive for HBeAg and HBV DNA at entry

Patient no. (yrllsex HBsAg carriage (mo) mode diagnosis (HAI) (pgkg body wt) treatment Age Time of known HBV transmission Histological rhGM-CSF dose Response to

1 9 N 5 71F 20iM 20lF 4 7 N 3 5 N 48iM 3 6 N 3 4 N

12 31 27 12 12 12 35 60 12

Familial Familial

Unknown Unknown

Homosexual Familial

Homosexual Homosexual Homosexual

CAH (4) CAH (11) CAH (9) CAH (9) 1

CAH (13) 1 CAH (12) 1

Cirrhosis (14) 0.5 CAH (9) 0.5 CAH (14) 0.5

3 reduced to 1.5 3 reduced to 1.5 3 reduced to 1.5

~ ~ _ _ _ _ ~____ ____ ____ ____ ~ ~

TABLE 2. Pretreatment values of ALT, AST and HBV DNA in the nine patients ALT/AST (IUIL) HBV DNA (ng/ml)

3 mo 2 mo 1 mo start of 2 mo 1 mo start of Patient no. before treatment before treatment before treatment treatment before treatment before treatment treatment

7613 1 82i45

109145 130165 80145

5721156 105170 103143 3811130

194165 79144

136161 77142 78143

209169 94168

107141 4231129

251177 110160 130160 106143 74/38

284191 83157 91/43

3801280

195160 11 1/64 123151 105151 152175 222176

80157 91136

532121 1

0.46 1.61 0.45 1.04 0.49 4.73 2.08 6.17 0.72

0.61 2.96 0.29 0.57 0.16 4.77 2.33 8.93 0.82

0.49 3.57 0.78 1.23 0.55 4.70 4.91

19.58 0.70

PATIENTS AND METHODS Patients. Nine patients (seven men and two women) with a

mean age of 35.1 yr (range = 19 to 57 yr) were included in this study. The pretreatment characteristics of the nine patients are summarized in Table 1. All patients had well-compensated liver disease with total bilirubin, prothrombin time and serum albumin values in the normal ranges; no signs of liver decompensation (ascites, hepatic encephalopathy or gas- trointestinal bleeding) were observed at any time during this trial. All patients were class A of the Child-Pugh system. Biopsy specimens obtained in the 6 mo before study entry proved histologically that each subject had chronic liver disease (eight had CAH and one had cirrhosis). Biopsy specimens were graded according to Knodell’s index (14). All patients had had serum HBsAg, HBeAg and HBV DNA for at least 6 mo, and had abnormal ALT and AST values (at least 1.5 times the upper limit of normal: 43 IUiL for ALT and 35 IU/L for AST) during that period. To rule out the possibility of spontaneous seroconversion, stable pretreatment values of ALT and AST and HBV DNA levels were considered entry criteria; they are summarized in Table 2. Other causes of liver disease (autoim- munity, alcohol, etc.) were ruled out, and no patient had antibodies to hepatitis delta virus, to hepatitis C virus or to human immunodeficiency virus type 1. Two patients had been previously treated with recombinant IFN-a (patient 7 received 9 MU three times weekly over 6 mo, and patient 8 received 1.5 MU three times weekly over 4 mo), but these treatments had ended more than a year before the beginning of this study and neither patient responded at all to the previous cycle (neither became HBV DNA negative and had always abnormal ALT values).

The patients were randomly assigned to one of three groups of three patients each. They received 3, 1 or 0.5 pg recom-

binant human (rh) GM-CSF/kg body wt/day for 6 wk by subcutaneous injection followed by two resting weeks, as performed in previous trials. rhGM-CSF treatment was then restored, together with 5 MU recombinant interferon aZb (rIFN-a,,)/day for 6 wk by subcutaneous injection. Both agents were administered sequentially: first the rhGM-CSF dose and then, immediately, the 5 MU dose of rIFN-a,,. Because of increased WBC counts and possible hyperviscosity syndrome, the 3-pg dose was reduced to 1.5 pg after 4 wk of treatment; this dose was used for the combination as well. Both rhGM-CSF and rIFN-aZb were kindly provided by Schering-Plough Corp. (Kenilworth, NJ). All patients were examined every 2 wk until the end of treatment, monthly until the seventh month and every 3 mo until the end of follow-up (13 mo). At that time a second liver biopsy was performed. At each visit a clinical examination was performed and blood samples were taken for blood cell counts, liver function tests, assay of HBV markers and other assessments. The patients gave written, informed consent, and the trial was approved by the Ethics Committee of the hospital and conducted in accordance with the Declaration of Helsinki on human experimentation.

hboratory Tests. HBV markers (HBsAg, HBc antibody, HBeAg and HBe antibody) and hepatitis D virus antibody were assayed with commercial RIAs (Abbott Laboratories, North Chicago, IL). Serum HBV DNA was detected and quantitated on dot-blot hybridization (15) with a 32P-labeled HBV-DNA probe using serial dilutions of known concentrations of the whole, recircularized HBV DNA molecule as standard (detec- tion limit < 1 pg cloned HBV-DNA). All samples were tested in the same run and, levels of HBV-DNA were estimated by the counts per minute of the radioactive spots counted in the Ambis Radioanalytic Imaging System (Ambis System Inc., San

-

HEPATOLOGY Vol. 18, No. 4, 1993 MARTIN ET AL. 777

30 I OM-CSF GM-CSF4FN n I -I 1-1

-4 -2 0 2 4 6 8 10 12 14 16 18 20 A weeks

W-CSF QYCSF4FN GM-CSFNFN

2.4 - 20 /

L I OM-CSF I

I I ,--I _ i O L 1 . 8 t , 1 1 , I , , , -i I

-4 -2 0 2 4 6 8 10 12 14 16 18 20 B w e k a C w o k s -4 -2 0 2 4 6 8 10 12 14 16 18 20

FIG. 1. Variations in blood cell counts during rhGM-CSF therapy alone and in combination with rIFN-a,,. (A) Leukocyte counts, expressed as average values of the three patients receiving each dose of rhGM-CSF (O = 0.5 pg; A = 1 pg; 0 = 3 reduced to 1.5 pg). (B) Monocyte counts expressed as mean values of all nine patients. (C) Granulocyte counts expressed as mean values of all nine patients. Vertical bars represent the S.D.

Diego, CA). Serum samples from healthy individuals without HBV markers were included as negative controls. Antibody to type 1 human immunodeficiency virus was assayed on ELISA (Abbott Laboratories). Antibody to hepatitis C virus was also assayed on ELISA (Ortho Diagnostic Systems Inc., Raritan, NJ).

2',5'-Oligoadenylate Synthetase Activity. The activity of the enzyme 2'-5'-oligoadenylate (2-5A) synthetase was analyzed before, during and after therapy in peripheral blood mononu-

was measured in cell lysates with a sensitive RIA (Eiken Chemical Co., Tokyo, Japan), with an assay range between 10 and 810 pmol/dl and intraassay coefficient of variation of less than 15%.

Statistical Analysis. We analyzed data with the nonpara- metric signed-rank test to search for differences in each parameter.

RESULTS clear cells (PBMCs). PBMCs were isolated from heparinized venous blood with Ficoll-Hypaquegradient sedimentation (SEROMED; Biochrom KG, Berlin, Germany). The interphase PBMCs were collected, washed twice with PBS and resus- pended in RPMI-1640 (Serva Feinbiochemica, Heidelberg,

glutamine, 10% FCS (Imperial Laboratories, Andover, UK) and 10 p,g/ml gentamicin to yield a final concentration of 1.5 to 2 x 106 cells/ml, after viability was assessed on the basis of trypan blue exclusion. The isolated PBMCs were seeded onto 24-well cell culture plates (Costar, Cambridge, MA), at a density of 1.5 to 2 X lo6 viable cells/well and cultured in triplicate without stimulation for 24 hr at 37" C in a humidified atmosphere with 5% CO,. At the end of culture, supernatants were discarded and the cells were separated from wells with trypsidEDTA, centrifuged at low speed, washed twice with PBS and treated with a lysing buffer (0.5% NP40,25 mmoW HEPES, 5 mmol/L M~CI,, 10% glycerol, 1 mmol/L PMSF, 10 kmol/L leupeptin, 1 mmol/L benzamidine and 1 mmol/L EDTA) to release the 2-5Asynthetase activity. 2-5A synthetase

Hematological Assessment. rhGM-CSF produced a dose-dependent increase in WBC counts during the entire period of administration of the drug (Fig. la). rhGM-CSF at doses Of kg increased WBC counts to

quently, the dose was reduced to 1.5 Pi3 after wk 4 in these three patients. This side effect was reversible after cessation of rhGM-CSF therapy. We noted a slight decrease in platelet counts (about lo%), a moderate increase in lymphocyte counts (data not shown) and a marked increase in monocyte ( ~ i ~ . 1b) and granulocyte counts pig. IC) during administration of rhGM-cSF. These changes were dose dependent. During the nation of rhGM-CSF Plus rIFN-a,,, the effects on cell counts were except that we found pronounced d~creases in numbers of platelets (about 30%) and lymphocytes (data not shorn).

Antiviral Effects. Two weeks after the beginning of

Germany) supplemented with 10 mmol/L HEPES, 2 mmol/L inhderable levels (UP to 40 lo3 Ce11s/d)* Conse-

778 MARTINETAL. HEPATOLOGY October 1993

0 2 4 6 8 10 12 14 16 (e 2 0

WEEKS

FIG. 2. Percentages of reduction in serum HBV DNA concentration in the nine patients.

rhGM-CSF administration, HBV DNA levels had de- creased by about 50% (p < 0.01) compared with pre- treatment levels (Fig. 2). The reduction in serum HBV DNA concentration was observed with the three doses of treatment: 60%, 30%, and 12% with the doses of 0.5 p,g, 1 p.g and 3 p,g reduced to 1.5 p,g, respectively. This trend remained significant until completion of rhGM-CSF therapy (p < 0.02), with a reduction by about 30% in serum HBV-DNA concentration. During the 2 wk free of medication, the HBV DNA concentration increased to pretreatment levels. When rhGM-CSF treatment was restored together with rIFN-a,,, a significant reduction in HBV DNA concentration (about 70%) was again observed (p < 0.02). After cessation of therapy, a re- bound in HBV DNA concentration was noted in five cases. During the two cycles of treatment, three patients lost HBV DNA (patient 9 a t wk 6, patient 5 a t wk 10 and patient 4 at wk 14). Furthermore, patient 3 lost HBV DNA at wk 20. Serum HBV DNA levels in the pre- treatment samples of these four patients were lower, but not significantly so, than in the rest of the patients.

A similar pattern was observed for serum HBeAg. Four patients lost HBeAg (patient 5 at wk 14, patient 9 at wk 20 and patients 3 and 4 at wk 241, and two seroconverted to HBe antibody positivity (patient 5 at wk 24 and patient 3 at wk 28) after the combination of rhGM-CSF plus rIFN-a,,.

In Vitro Assessment of rhGM-CSF Effects. We noted an increase in 2-5A synthetase activity in PBMCs obtained during rhGM-CSF therapy compared with pretreatment levels. This increase became significant a t wk 4 of treatment (p < 0.01). Individual patient re- sponses are shown in Figure 3. The effects of rhGM-CSF doses of 3 and 1 p,g were more pronounced than those of the 0.5 p,g dose, with peak activity by wk 2 or 4 in most of the patients. After wk 4, 2-5A synthetase activity tended to decline to near-pretreatment levels. The increase in 2-5A synthetase activity coincided with the reduction in HBV DNA concentration. During the combination of rhGM-CSF plus rIFN-a,,, we noted again an increase in 2-5A synthetase activity (Fig. 3). These increases were significant (p < 0.05) compared

with values in the resting period. After cessation of treatment, 2-5A synthetase activity tended to decline to near-pretreatment levels.

Histological Evaluation. A second liver biopsy was performed at mo 13 in eight patients (one dropped out of the study). The total histological activity index (HAI) decreased in all patients (patients 3 , 4 , 5 and 9) who lost serumHBVDNAandHBeAg(ll.2 '-t 2 . 6 ~ s . 8.5 2 1.7). Amelioration was observed in piecemeal necrosis and portal inflammation, but without changes in liver fibrosis. In contrast, patients who remained serum HBV DNA positive showed different outcomes: patient 1 had an improved HAI score to 3, two cases remained unchanged (patients 2 and 8 ) and patient 7 had a worsened total HAI score of 15; this patient had active cirrhosis. No change in the histological diagnosis was observed in any patient.

Biochemical Assessment. ALT values decreased within 2 wk of therapy (mean = 124 IU/L; range = 79 to 192 IU/L) with respect to pretreatment levels (mean = 179 IU/L; range = 80 to 532 IU/L). This trend remained constant until completion of rhGM-CSF therapy (mean = 93 IU/L; range = 46 to 140 IU/L) (Fig. 4). A further decrease in ALT was observed within 2 wk of the start of therapy with combined rhGM- CSF/rIFN-a,, therapy (mean = 79 IUIL; range = 42 to 119 IU/L). One patient dropped out of the trial a t that time. After cessation of therapy, only one patient (patient 4) had a normalized ALT value; flares in ALT values were observed in the remaining seven patients (mean = 148 IU/L; range = 27 to 382 IU/L). ALT levels were higher in patients in whom HBV DNA was absent at the end of the treatment than in the rest of the patients, although not significantly so. At the end of follow-up, two of the patients without serum HBV DNA had normal ALT values (patients 3 and 4). No oth- er biochemical parameters (albumin, bilirubin, pro- thrombin time, y-globulin, y-glutamyl transpeptidase, alkaline phosphatase) showed remarkable variations.

Side Effects. During rhGM-CSF therapy, all but one patient had a local reaction at the injection site con- sisting of an erythematous, pruritic papule, but this reaction disappeared within 24 to 36 hr of the injection. A mild flulike syndrome, with myalgia and fever of less than 38" C, was observed in three patients. No patient had anorexia, asthenia or hair or weight loss attrib- utable to GM-CSF. None reported bone pain or respi- ratory symptoms. We found no renal, hepatic, neurologic or cardiac symptoms during this cycle of treatment.

During the combination of rhGM-CSF plus IFN, all patients experienced a flulike syndrome, with fever greater than 38" C, arthralgias and myalgias in seven cases, mild headache in six, asthenia in four, anorexia in three and mild hair loss in three; none had weight loss. These symptoms are usually associated with IFN ad- ministration and disappeared with acetaminophen treatment. During the combination of rhGM-CSF with rIFN-a,,, neutropenia did not appear in any case, and no other complications such as bacterial infection de- veloped.

HEPATOLOGY Vol. 18, No. 4, 1993 MAFtTINETAL. 779

2 N

WEEKS A

I I

N $ 1 0 ' 4 * I m U Y I

B WEEKS

a

1

D O l . e a m u Y *

n C WEEKS

FIG. 3. Individual responses of 2-5A synthetase activity in cell lysates of PBMCs. (A) Patients treated with 3 reduced to 1.5 pg. (B) Patients treated with the 1-pg dose. (C) Patients treated with the 0.5-pg dose. Data expressed as the average of three independent experiments; vertical bars represent the S.D. Patient 6 dropped out by wk 10.

Only one patient (patient 6) dropped out of the trial; leukocytoclastic vasculitis had developed in his legs in the first week of the combined therapy with rhGM-CSF and rIFN-a,,. Therapy was discontinued, and the vasculitis subsequently disappeared.

DISCUSSION In this study we administered rhGM-CSF to patients

with HBeAg-positive , HBV DNA-positive chronic hep- atitis B to study tolerance and possible antiviral effects. GM-CSF was generally well tolerated and led to signif- icant increases in the numbers of WBCs in all patients. This result was expected because GM-CSF has been implicated in the maturation and proliferation of pro- genitor cells (11,12). rhGM-CSF has been administered to patients with low WBC counts (13). Because our patients had WBC counts in the normal range at entry, the increase of these cells (up to 40 x lo3 cells/Fl) was predictable. We saw no important differences in the antiviral effects between the doses used in our study; however, we do not recommend the 3-kg dose because the increase of WBCs may be harmful. It should be noted that, when rhGM-CSF administration was carefully monitored, no irreversible toxicity was noted, and its use was safe in patients with chronic hepatitis B. The other secondary effects observed, such as the flulike syn- drome, were much milder in our experience than those found with IFN administration.

A significant reduction in HBV DNA concentration was observed during therapy with rhGM-CSF. Four of the eight patients who completed the study lost HBV DNA and HBeAg; two of them seroconverted to HBe

QM-CSF QM-CSF*IFN T - . 2 225 3

135

90

45

0 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 ;L c t , * . . weeks / / I

FIG. 4. Variations in the mean ALT values of the nine patients during rhGM-CSF therapy alone and in combination with rIFN-a,,; vertical bars represent the S.D.

antibody positivity. Two patients ultimately had normal ALT values. In addition, a significant increase was observed in 2-5A synthetase activity. It has been demonstrated that the activation of this enzyme results in the degradation of viral messenger RNAs in vivo and i n vztro (16). The effect of rhGM-CSF administration on 2-5A synthetase activity has not been reported. The increase in this activity may represent a direct effect of rhGM-CSF. From these results, it can be suggested that rhGM-CSF has an antiviral effect in chronic HBV infections, as it does on human immunodeficiency virus infection (17). However, the antiviral mechanism of GM-CSF should be further investigated. When rhGM-

780 MARTIN ET A L HEPATOLOGY October 1993

CSF was given in combination with rIFN-a,,, a more pronounced reduction was observed in HBV DNA levels, together with a higher increase in 2-5A synthetase levels, compared with rhGM-CSF therapy alone. In future studies, it should be determined whether the antiviral effect is synergistic or additive.

In summary, the administration of rhGM-CSF is safe and tolerable a t low doses (i.e., 0.5 to 1 pgkg body wt) and seems to exert an antiviral effect on chronic HBV infection. In the future, one application of rhGM-CSF in chronic viral hepatitis could be to increase leukocyte counts in patients with low baseline levels to prepare them for a cycle of IFN. The efficacy of GM-CSF should be considered in the prevention of hematological dis- orders and in the increased susceptibility to bacterial infections during IFN therapy in chronic HBV infection. Also, rhGM-CSF may be given in combination with IFN to patients who do not respond to a cycle of IFN, although the possible benefit of this therapy should be proved in controlled trials.

1.

2.

3.

4.

5.

REFERENCES Gerber MA, Thung SN. Biology of disease: molecular and cellular pathology of hepatitis B. Lab Invest 1986;52:572-590. Anastassakos C, Alexander GJM, Wolstencroft RA, Avery JA, Portmann BC, Panayi GS, Dumonde DC, et al. Interleukin-1 and interleukin-2 activity in chronic hepatitis B virus infection. Gastroenterology 1988;94:999- 1005. Ikeda T, Lever AML, Thomas HC. Evidence for a deficiency of interferon production in patients with chronic hepatitis B virus infection acquired in adult life. HEPATOLOGY 1986;6:962-965. Kleinerman ES, Knowles RD, Lachman LB, Gutterman JU. Effect of recombinant granulocyteimacrophage colony-stimulating factor on human monocyte activity in vitro and following intravenous administration. Cancer Res 1988;48:2604-2609. Hoofnagle JH, Peters M, Mullen KD, Jones DB, Rustgi V, DiBisceglie A, Hallahan C, et al. Randomized, controlled trial of

recombinant human alpha-interferon in patients with chronic hepatitis B. Gastroenterology 1988;95:1318-1325.

6. Alexander GJM, Brahm J, Fagan EA. Loss of HBsAg with interferon therapy in chronic hepatitis B virus infection. Lancet

7. Perrillo RP, Regenstein FG, Peters MG, DeSchryver-Kecskemeti K, Bodicky CJ, Campbell CR, Kuhns MC. Prednisone withdrawal followed by recombinant alpha interferon in the treatment of chronic type B hepatitis. Ann Intern Med 1988;109:95-100.

8. Hoofnagle JH. Therapy of chronic type B hepatitis with adenine arabinoside and adenine arabinoside monophosphate. J Hepatol

9. Weller IV, Carreiio V, Fowler MJF, Monjardino J, Makinen D, Varghese Z, Sweny P, et al. Acyclovir in hepatitis B antigen positive chronic liver disease: inhibition of viral replication and transient renal impairment with iv bolus administration. J Antimicrob Chemother 1983;11:223-231.

10. Carreiio V, Mora I. Combination of recombinant interferons alpha and gamma in treatment of chronic hepatitis B. Lancet 1987;2:

11. Andreeff M, Welte K. Hematopoietic colony-stimulating factors. Semin Oncol 1989;16:2 11-229.

12. Lieschke GJ, Burgess AW. Granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor (part I). N Engl J Med 1992;327:28-35.

13. Lieschke GJ, Burgess AW. Granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor (part 11). N Engl J Med 1992;327:99-106.

14. Bianchi L, De Groote J , Desmet VJ, Gedigk P, Korb G, Popper H, Poulsen H, et al. Acute and chronic hepatitis revisited. Lancet

15. Berninger M, Hammer M, Moyer B, Gerin JL. An assay for the detection of the DNA genome of hepatitis B virus in serum. J Med Virol 1982;9:57-68.

16. Houglum JE. Interferon: mechanisms of action and clinical value. Clin Pharmacol 1983;2:20-28.

17. Perno CF, Yarchoan R, Cooney DA, Hartman NR, Webb DSA, Hao Z, Mitsuya H, et al. Replication of human immunodeficiency virus in monocytes: granulocyte/macrophage colony-stimulating factor (GM-CSF) potentiates viral production yet enhances the antiviral effect mediated by 3’-azido-2’,3’-dideoxythymidine (AZT) and other dideoxynucleoside congeners of thymidine. J Exp Med

1987;2:66-68.

1986;3 (~~ppl 211573-580.

1086-1087.

1977;2:914-919.

1989; 169:933-951.