Original Articles
A Pilot Study of 2',3'-Dideoxyinosine for the Treatment of Chronic Hepatitis B MICHAELW. FRIED,' JULIA C. KORENMAN,'ADRIAN M. DI BISCEGLIE,~ YOON PARK,' JEANNE G. WAGGONER,l HIROAKIMITSUYA,~ NEIL R. &WIWA."I'2 ROBERTYARCHOAN,2 SAMUEL BRODER~ AND JAY H. HOOFWAGLE' 'Liver Diseases Section, Digestive Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, and 2National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892
The nucleoside analog 2',3'-dideoxyinosine, currently being used to treat patients infected with the human immunodeficiency virus, has been shown to inhibit viral replication in certain cell culture systems of hepatitis B virus and the duck model of chronic hepatitis B infection. We studied the effect of dideoxyinwine on viral replication in patients with chronic hepatitis B. In the initial dose-findingphase, patients received sequential 2-wk courses of dideoxyinosine in escalating dosea of 3,6 and 9 mg/kg/day.In the second, long-term treatment phase, patients received dideoxyinosine at a dose of 9 mg/kg/dayfor 12 wk. Dideoxyinosine was given orally in three divided doses. The effects of dideoxyinosine on hepatitis B were assessed by serial measurements of ALT, hepatitis B virus DNA and DNA polymerase activity in serum. Six patients completed the dose-finding phase, and five patients continued into the long-term treatment phase. No aigniatcantdifferences were seen in serum aminotransferaseg hepatitis B virus DNA levels or DNA polymerase activity at any time during treatment when compared with pretreatment levels. All patients remained positive for HBeAg during treatment and during 6 mo of follow-up. Thus at the doses tested, dideoxyinosine had no appreciable effect on viral replication in patients with chronic hepatitis B. (HEPATOLOGY 1992;16861-864.)
potent antiviral action against HIV in uitro by inhibiting viral reverse transcriptase and to have beneficial clinical effects in patients with HIV infection (5, 6). Although distinct differences exist between HBV and HIV,they both replicate by way of an RNA intermediate that requires reverse transcriptase activity (7). Thus it is possible that the dideoxynucleosidesmay be of benefit in the treatment of chronic hepatitis B. Indeed, in an animal model of chronic hepatitis B infection, 2',3'dideoxycytidine (DDC), 2',3'-dideoxyguanosine (DDG), 2 ',3'-dideoxyadenosine (DDA) and 2',3'-dideoxyinosine (DDI) inhibited replication of the duck HBV (8-10). In light of these findings, we initiated a pilot study to determine the efficacy of DDI for the treatment of patients with chronic hepatitis B.
PATIENTS AND METHODS Patients with chronic hepatitis B of more than 6 mo duration, abnormal aminotransferases, histological evidence of chronic hepatitis and markers of active viral replication (including HBsAg, HBeAg, HBV DNA and DNA polymerase) were selected to receive DDI. AU patients were negative for antibodies to HDV, HCV and HIV. DDI provided by the Developmental Therapeutics Program of the National Cancer Institute (Bethesda, MD)was supplied to the patient in powder form and reconstituted with normal saline to yield a concentration of 15 mg/ml of DDI. The Chronic hepatitis B remains a worldwide health reconstituted solution was mixed with either water or apple problem with significant morbidity and mortality from juice and ingested orally 2 min after a 30-cc dose of antacid to progressive liver disease and HCC (1).Efforts to treat minimize gastric acidity and enhance DDI absorption. DDI was chronic hepatitis B have met with only limited success. taken 1 hr before or 2 hr after meals on a three times daily Although interferon-a induces loss of viral replication in dosing schedule. The dose-finding phase of the study consisted of sequential 30% to 40% of patients with chronic hepatitis B (2, 31, the route of administration, side effects and limited 2-wk courses of DDI at 3, 6 and 9 m w d a y taken orally in efficacy of interferon necessitate a search for new three divided doses. Each course was separated by a drug-free interval of at least 2 wk. Thereafter, patients were treated with treatments. DDI at a dose of 9 mg/kg/day for 12 wk. Recent advances in chemotherapy for human immuSerum samples were obtained on days 1,2,3,5,8,15 and on nodeficiency virus ( H W have also provided potential day 22 (1wk after the last dose of DDI) during the dose-finding treatment options for chronic hepatitis B (4). A group of phase of the study for measurement of ALT, electrolytes,BUN, 2',3'-dideoxynucleosideanalogs has been shown to have creatinine, triglycerides,uric acid, amylase and complete blood count. Serum samples were also tested for HBV DNA as measured by a liquid-phase hybridization assay (Abbott Laboratories, North Chicago, IL) and for HBV DNA polyReceived October 16, 1991; accepted February 27, 1992. Address reprint requests to:Michael W. Fried, M.D.. Liver Diseases Section, merase activity quantitated by measurement of incorporation of tritiated thymidine into purified virus particles and exNIH, Bldg 10,Rm 9C 103B,Bethesda, MD 20892. 31/1/39660 pressed as counts per minute per 0.2 ml. Selected samples were 861
862
FRIED ET AL
HEPATOLOGY
4Ooo
HBVDNA POLYMERASE (cpml
3ooo 2000
PRE 1
15 22
15 22
1
1
15 22
DAYS
PRE 1
3
5
F’IG 1. DNA polymerase activity during DDI therapy (mean k S.E.M.).Dose-finding phase: n
tested for HBsAg and HBeAg by RIA (Abbott Laboratories). During the long-term phase of DDI treatment, the same measurements were made on samples obtained at intervals of 1to 4 wk. Pharmacokinetic studies of DDI absorption were performed on day 5 of each dosage level during the dose-findingphase of the study. Blood sampleswere drawn at 30,60and 90 min after the ingestion of DDI, and plasma DDI levels were measured by HPLC as previously described (11). The aim of treatment was to demonstrate the inhibition of viral replication defined as a decrease in the levels or loss of HBV DNA, DNA polymerase activity and HBeAg from serum. The mean values of ALT, HBV DNA and HBV DNA polymerase at various time points were compared with pretreatment values using Student’s paired t test. This study was approved by the Institutional Clinical Research Subcommittee of the National Institute for Diabetes and Digestive and Kidney Diseases, and all patients gave written informed consent.
9
7
11 12
WEEKS
=
6; long-term treatment phase: n = 5.
The mean peak plasma DDI levels measured 30 min after ingestion at a dose of 1, 2 and 3 mg/kg were 1.82 ? 0.9, 2.63 ~f:0.9and3.02 2 1.44 FmoVL, respectively. HBV DNA polymerase activity (mean r S.E.M.) during each course of DDI is shown in Figure 1. No significant changes occurred in HBV DNA polymerase activity between pretreatment levels and levels at any time point measured during any of the three dosage regimens (all p > .1). Mean serum HBV DNA and HBV DNA polymerase activity during the long-term treatment phase is shown in Figure 2. No significant differences occurred in HBV DNA levels or HBV DNA polymerase activity at any time during this phase as compared with pretreatment values (all p > .1).Mean serum HBV DNA was 76.0 ~f:27.4 pg/ml before therapy and 63.2 ~f:18.1pg/ml during the last week of the 12-wk treatment period. No patient became negative for HBeAg or HBsAg during therapy or in the 6-mo follow-up period thereafter. One patient who started the trial with the lowest level of HBV DNA became DNA polymerase negative 1 wk into the long-term treatment phase. HBV DNA decreased and became negative 1 mo after the cessation of DDI. Six months after completing DDI, HBV DNA polymerase remained negative but HBeAg was still positive. Subsequently, HBeAg became negative and anti-HBe was detectable 7 mo after the completion of DDI. DDI was well tolerated without any side effects. Specifically, no patient had peripheral neuropathy, pancreatitis, evidence of hepatotoxicity, hypertriglyceridemia or hyperuricemia develop, as has been previously described with DDI treatment in HIV-infected patients (6, 12, 13).
RESULTS Seven adult men were enrolled in the study. One patient who lost HBeAg and had an antibody to HBeAg develop after only 1wk of DDI treatment was excluded from further analysis; in retrospect, this patient had falling HBV DNA and DNA polymerase activity levels before the start of DDI, and therefore the seroconversion appeared to be spontaneous rather than drug induced. Another patient withdrew from the study for personal remons at the end of the dose-finding phase. Thus six patients completed the dose-finding phase, of whom five continued into the long-term treatment phase. The mean age of treated patients was 33 yr (range = 25 to 40 yr), and the mean known duration of hepatitis was 4.4 yr (range = 1 to 9 yr). Four patients had previously failed a trial of interferon-a, and one patient had a mild form of HBV-associated glomeruloDISCUSSION nephritis. At entry, mean ALT was 125.6 ? 74.4 U/L (mean k DDI has several characteristics that make it a novel S.D.). Liver biopsy specimens showed CAH in all six and potentially exciting treatment for patients with patients. AU patients had HBeAg in pretreatment chronic hepatitis B. DDI was chosen instead of other serum, and mean serum HBV DNA and HBV DNA dideoxynucleosides because of its availability and prepolymerase activity levels were 76 ? 61.2 pcg/ml and vious use in patients with HIV infection. Both HBV and HIV replicate through an RNA intermediate. This 2,691 2 2,708 cpdO.2 ml, respectively.
Vol. 16, No. 4, 1992
863
DDI THERAPY FOR CHRONIC HEPATITIS B
requires reverse transcriptase activity that, for HBV, produces DNA from the RNA template (7). DDI has been shown to inhibit the reverse transcriptase activity of HIV ( 5 ) and, therefore, could have a similar effect on hepatitis B. Experimental evidence in uiuo in an animal model of HBV infection has also provided support for this hypothesis. In the Pekin duck model of hepatitis B, four dideoxynucleosides (DDC, DDG, 'DDA and DDI) have been shown to rapidly decrease DNA polymerase activity and duck HBV DNA levels by 60% to 70% of pretreatment levels (8-10). In uitro studies using hepatocyte cell culture systems to test the antiviral effects of the dideoxynucleosides have yielded conflicting results. In studies using human hepatoblastoma cell lines, DDA and DDI exhibited a variable inhibitory effect on HBV replication in some cell lines but not in others (14,15).In contrast, in a primary culture system using duck HBV infected hepatocytes, DDA and DDG markedly inhibited duck HBV, whereas DDI was modestly effective (16). Despite the encouraging experimental data, the results of this study were disappointing. We treated six patients with chronic hepatitis B with DDI for up to 3 mo and found no consistent or appreciable effect on viral replication or disease activity. One patient who started the trial with the lowest levels of HBV DNA and DNA polymerase activity became negative for both these markers of viral replication, but he did not lose HBeAg in serum until 7 mo after completing DDI treatment. Even including HBV DNA and DNA polymerase from this patient, no statistically significant change occurred in mean serum HBV levels during the 3 mo of treatment with DDI at the highest dose evaluated. The reasons why DDI was ineffective in humans with chronic hepatitis B are not clear. The antiviral action of DDI against HIV appears to require successive phosphorylations to yield its 5'-triphosphate (ddA!J!P), which competes with the naturally occurring nucleotide for binding to DNA polymerase and, because ddATP is incorporated into the growing viral DNA chain, brings about termination of viral DNA synthesis and viral replication (4). Although DDI is efficiently phosphorylated in human lymphoid cells (171, DDI may be phosphorylated to a lesser extent within human hepatocytes, possibly because of lower levels of kinases. Higher levels of the naturally occurring nucleotide also may exist that can nullify the antiviral action of DDI. Finally, DDI may be rapidly catabolized within the human liver. Any of these reasons could explain the favorable viral inhibition demonstrated by DDI in duck HBV model systems (8,9,16) as compared with the limited effects in human HBV-infected cell lines (14, 15). In addition, human HBV may have different sensitivities to antiviral agents than duck HBV. Although both viruses have similar genomic structure and modes of replication, it remains to be proven whether the duck HBV model is appropriate for studying the human disease. It is also possible that the dosages of DDI used in this
O
z n 0
PRE
4
8
12
16
WEEKS
Rc.2. HBV DNA and DNA polymerase activity during long-term treatment with DDI (mean +- S.E.M.).
study were insufficient to inhibit HBV replication, especially because the mean peak plasma levels after similar dosages of DDI were lower in our patients with chronic hepatitis B than in patients with severe HIV infection (11). Patients with AIDS have been shown to have higher gastric pH and less stimulated gastric acid production than normal controls (18). Because DDI is acid labile, it is possible that DDI absorption was decreased in patients with chronic hepatitis B (and presumably normal gastric acid secretion) because of insufficient neutralization of gastric acidity by the oral antacids taken before DDI ingestion. Thus inhibition of HBV replication may require higher doses of DDI, perhaps in conjunction with other methods to suppress gastric acidity, or more stable preparations of DDI. It should be noted, however, that DDI toxicity is also dose related; side effects can be significant and potentially lethal (12, 13). In summary, despite the in uiuo and in vitro experimental evidence suggesting that DDI may be useful in treating chronic hepatitis B, we did not find a consistent, appreciable effect on viral replication in a cohort of patients with chronic hepatitis B. Additional studies would be needed to determine the reasons for lack of efficacy of DDI in these patients and to resolve the discrepancies seen with DDI in experimental models of HBV infection. REFERENCES 1. Hoofnagle JH, Alter HJ. Chronic viral hepatitis. In: Vyas GN,
Dienstag JL,Hoofnagle JH, eds. Viral hepatitis and liver disease. Orlando, FL: Grune L Stratton, 1984:97-113. 2. Hoofhagle JH, Peters M, Mullen KD, Jones DB, Rustgi V, Di Bisoeglie AM, Hallahan C, et al. Randomized controlled trial of
864
FRIED ET AL.
human alpha-interferon in patients with chronic hepatitis B. Gastroenterology 1988;95:1318-1325. 3. Perrillo RP, Schiff ER, Davis GL, Bodenheimer HC, and the Hepatitis Interventional Therapy Group. A randomized, controlled trial of interferon alfa-2b alone and after prednisone withdrawal for the treatment of chronic hepatitis B. N Engl J Med 1990;323:295-301. 4. McGowan JJ, Tomaszewski JE,Cradock J, Hoth D, Grieshaber CK, Broder S, Mitsuya H. Overview of the preclinicaldevelopment of an antiretroviral drug, 2',3'-dideoxyinosine. Rev Infect Dis 1990;12(suppl5):s513-s521. 5. Mitsuya H, Yarchoan R, Broder S. Molecular targets for AIDS therapy. Science 1990;249:1533-1544; 1986;83:1911-1915. 6. Yarchoan R, Mitsuya H, Thomas RV, Pluda JM, Hartman NR, Perno CF, Marczyk KS, et al. In vivo activity against HIV and favored toxicity profile of 2',3'-dideoxyinosine. Science 1989;245: 412-415. 7. Summers J, Mason WS. Replication of the genome of a hepatitis B-like virus by reverse transcriptase of an RNA intermediate. Cell 1982;29:403-415. 8. Kassianides C, Hoofnagle JH,Miller RH, Doo E, Ford H, Broder S, Mitsuya H. Inhibition of duck hepatitis B virus by 2',3'dideoxycytidine. Gastroenterology 1989;97:1275-1280. 9. Martin P, Kassianidw C, Korenman J, Hoofnagle JH, Ford H, Broder S, Mitsuya H. 2',3'-dideoxyinosine (DDI) and dideoxyguanosine are potent inhibitors of hepadnaviruses in viva [Abstract]. Gastroenterology 1989;96:A628. 10. Martin P, Kassianides C, Hoofnagle JH, Mitsuya H, Broder S. Effects of 2',3'-dideoxyadenosine on duck hepatitis B virus [Abstract]. HEPATOLOGY 1988;8:A1329.
HEPATOLOGY
11. Hartman NR, Yarchoan R, Pluda JM, Thomas RV, Marczyk KS, Broder S, Johns DG. Pharmacokinetica of 2',3'-dideoxyadenosine and 2',3'-dideoxyinosine in patients with severe human immunodeficiency virus infection. Clin Pharmacol Ther 1990;47:647-654. 12. Yarchoan R, Pluda JM, Thomas RV, Mitmya H, Brouwers P, WyviU KM, Hartman N, et al. Long-termtoxicity/activityprofile of 2',3'-dideoxyinosine in AIDS or AIDS-related complex. Lancet 1990;336:526-529. 13. Lai KK, Gang DL, Zawacki JK, Cooley TP. Fulminant hepatic failure associated with 2',3'-dideoxyinosine. Ann Intern Med 1991;115:283-284. 14. Ueda K, Tsurimoto T, Nagahata T, Chisaka 0, Matsubara K. An in vitro system for screening anti-hepatitis B drugs. Virology 1989;169:213-216. 15. Aoki-Sei S, O'Brien MC, Ford H, Fujii H, Gilbert DA, Cooney DA, Johns DG, et al. In vitro inhibition of hepatitis B virus replication by 2',3'-dideoxyguanosine, 2',3'-dideoxyinosine, and 3'-azido2',3'-dideomhymidine in 2.2.15(PR) cells. J Infect Dis 1991;164: 843-851. 16. Lee B, Luo W, Suzuki S, Robins MJ, Tyrrell DLJ. In vitro and in viva comparison of the abilities of purine and pyrimidine 2',3'dideoxynucleosides to inhibit duck hepadnavirus. Antimicrob Agents Chemother 1989;33:336-339. 17. Johnson MA, Fridland A. Phosphorylation of 2',3'-dideoxyinosine by cytosolic 5'-nucleotidase of human lymphoid cells. Mol Pharmacol 1989;36:291-295. 18. Lake-Bakaar G, Quadros EP, Beidas S, Elsakr M, Tom W, Wilson DE, Dinaoy HP, et al. Gastric secretory failure in patients with the acquired immunodeficiency syndrome (AIDS). Ann Intern Med 1988;109:502-504.
A Pilot Study of 2',3'-Dideoxyinosine for the Treatment of Chronic Hepatitis B MICHAELW. FRIED,' JULIA C. KORENMAN,'ADRIAN M. DI BISCEGLIE,~ YOON PARK,' JEANNE G. WAGGONER,l HIROAKIMITSUYA,~ NEIL R. &WIWA."I'2 ROBERTYARCHOAN,2 SAMUEL BRODER~ AND JAY H. HOOFWAGLE' 'Liver Diseases Section, Digestive Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, and 2National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892
The nucleoside analog 2',3'-dideoxyinosine, currently being used to treat patients infected with the human immunodeficiency virus, has been shown to inhibit viral replication in certain cell culture systems of hepatitis B virus and the duck model of chronic hepatitis B infection. We studied the effect of dideoxyinwine on viral replication in patients with chronic hepatitis B. In the initial dose-findingphase, patients received sequential 2-wk courses of dideoxyinosine in escalating dosea of 3,6 and 9 mg/kg/day.In the second, long-term treatment phase, patients received dideoxyinosine at a dose of 9 mg/kg/dayfor 12 wk. Dideoxyinosine was given orally in three divided doses. The effects of dideoxyinosine on hepatitis B were assessed by serial measurements of ALT, hepatitis B virus DNA and DNA polymerase activity in serum. Six patients completed the dose-finding phase, and five patients continued into the long-term treatment phase. No aigniatcantdifferences were seen in serum aminotransferaseg hepatitis B virus DNA levels or DNA polymerase activity at any time during treatment when compared with pretreatment levels. All patients remained positive for HBeAg during treatment and during 6 mo of follow-up. Thus at the doses tested, dideoxyinosine had no appreciable effect on viral replication in patients with chronic hepatitis B. (HEPATOLOGY 1992;16861-864.)
potent antiviral action against HIV in uitro by inhibiting viral reverse transcriptase and to have beneficial clinical effects in patients with HIV infection (5, 6). Although distinct differences exist between HBV and HIV,they both replicate by way of an RNA intermediate that requires reverse transcriptase activity (7). Thus it is possible that the dideoxynucleosidesmay be of benefit in the treatment of chronic hepatitis B. Indeed, in an animal model of chronic hepatitis B infection, 2',3'dideoxycytidine (DDC), 2',3'-dideoxyguanosine (DDG), 2 ',3'-dideoxyadenosine (DDA) and 2',3'-dideoxyinosine (DDI) inhibited replication of the duck HBV (8-10). In light of these findings, we initiated a pilot study to determine the efficacy of DDI for the treatment of patients with chronic hepatitis B.
PATIENTS AND METHODS Patients with chronic hepatitis B of more than 6 mo duration, abnormal aminotransferases, histological evidence of chronic hepatitis and markers of active viral replication (including HBsAg, HBeAg, HBV DNA and DNA polymerase) were selected to receive DDI. AU patients were negative for antibodies to HDV, HCV and HIV. DDI provided by the Developmental Therapeutics Program of the National Cancer Institute (Bethesda, MD)was supplied to the patient in powder form and reconstituted with normal saline to yield a concentration of 15 mg/ml of DDI. The Chronic hepatitis B remains a worldwide health reconstituted solution was mixed with either water or apple problem with significant morbidity and mortality from juice and ingested orally 2 min after a 30-cc dose of antacid to progressive liver disease and HCC (1).Efforts to treat minimize gastric acidity and enhance DDI absorption. DDI was chronic hepatitis B have met with only limited success. taken 1 hr before or 2 hr after meals on a three times daily Although interferon-a induces loss of viral replication in dosing schedule. The dose-finding phase of the study consisted of sequential 30% to 40% of patients with chronic hepatitis B (2, 31, the route of administration, side effects and limited 2-wk courses of DDI at 3, 6 and 9 m w d a y taken orally in efficacy of interferon necessitate a search for new three divided doses. Each course was separated by a drug-free interval of at least 2 wk. Thereafter, patients were treated with treatments. DDI at a dose of 9 mg/kg/day for 12 wk. Recent advances in chemotherapy for human immuSerum samples were obtained on days 1,2,3,5,8,15 and on nodeficiency virus ( H W have also provided potential day 22 (1wk after the last dose of DDI) during the dose-finding treatment options for chronic hepatitis B (4). A group of phase of the study for measurement of ALT, electrolytes,BUN, 2',3'-dideoxynucleosideanalogs has been shown to have creatinine, triglycerides,uric acid, amylase and complete blood count. Serum samples were also tested for HBV DNA as measured by a liquid-phase hybridization assay (Abbott Laboratories, North Chicago, IL) and for HBV DNA polyReceived October 16, 1991; accepted February 27, 1992. Address reprint requests to:Michael W. Fried, M.D.. Liver Diseases Section, merase activity quantitated by measurement of incorporation of tritiated thymidine into purified virus particles and exNIH, Bldg 10,Rm 9C 103B,Bethesda, MD 20892. 31/1/39660 pressed as counts per minute per 0.2 ml. Selected samples were 861
862
FRIED ET AL
HEPATOLOGY
4Ooo
HBVDNA POLYMERASE (cpml
3ooo 2000
PRE 1
15 22
15 22
1
1
15 22
DAYS
PRE 1
3
5
F’IG 1. DNA polymerase activity during DDI therapy (mean k S.E.M.).Dose-finding phase: n
tested for HBsAg and HBeAg by RIA (Abbott Laboratories). During the long-term phase of DDI treatment, the same measurements were made on samples obtained at intervals of 1to 4 wk. Pharmacokinetic studies of DDI absorption were performed on day 5 of each dosage level during the dose-findingphase of the study. Blood sampleswere drawn at 30,60and 90 min after the ingestion of DDI, and plasma DDI levels were measured by HPLC as previously described (11). The aim of treatment was to demonstrate the inhibition of viral replication defined as a decrease in the levels or loss of HBV DNA, DNA polymerase activity and HBeAg from serum. The mean values of ALT, HBV DNA and HBV DNA polymerase at various time points were compared with pretreatment values using Student’s paired t test. This study was approved by the Institutional Clinical Research Subcommittee of the National Institute for Diabetes and Digestive and Kidney Diseases, and all patients gave written informed consent.
9
7
11 12
WEEKS
=
6; long-term treatment phase: n = 5.
The mean peak plasma DDI levels measured 30 min after ingestion at a dose of 1, 2 and 3 mg/kg were 1.82 ? 0.9, 2.63 ~f:0.9and3.02 2 1.44 FmoVL, respectively. HBV DNA polymerase activity (mean r S.E.M.) during each course of DDI is shown in Figure 1. No significant changes occurred in HBV DNA polymerase activity between pretreatment levels and levels at any time point measured during any of the three dosage regimens (all p > .1). Mean serum HBV DNA and HBV DNA polymerase activity during the long-term treatment phase is shown in Figure 2. No significant differences occurred in HBV DNA levels or HBV DNA polymerase activity at any time during this phase as compared with pretreatment values (all p > .1).Mean serum HBV DNA was 76.0 ~f:27.4 pg/ml before therapy and 63.2 ~f:18.1pg/ml during the last week of the 12-wk treatment period. No patient became negative for HBeAg or HBsAg during therapy or in the 6-mo follow-up period thereafter. One patient who started the trial with the lowest level of HBV DNA became DNA polymerase negative 1 wk into the long-term treatment phase. HBV DNA decreased and became negative 1 mo after the cessation of DDI. Six months after completing DDI, HBV DNA polymerase remained negative but HBeAg was still positive. Subsequently, HBeAg became negative and anti-HBe was detectable 7 mo after the completion of DDI. DDI was well tolerated without any side effects. Specifically, no patient had peripheral neuropathy, pancreatitis, evidence of hepatotoxicity, hypertriglyceridemia or hyperuricemia develop, as has been previously described with DDI treatment in HIV-infected patients (6, 12, 13).
RESULTS Seven adult men were enrolled in the study. One patient who lost HBeAg and had an antibody to HBeAg develop after only 1wk of DDI treatment was excluded from further analysis; in retrospect, this patient had falling HBV DNA and DNA polymerase activity levels before the start of DDI, and therefore the seroconversion appeared to be spontaneous rather than drug induced. Another patient withdrew from the study for personal remons at the end of the dose-finding phase. Thus six patients completed the dose-finding phase, of whom five continued into the long-term treatment phase. The mean age of treated patients was 33 yr (range = 25 to 40 yr), and the mean known duration of hepatitis was 4.4 yr (range = 1 to 9 yr). Four patients had previously failed a trial of interferon-a, and one patient had a mild form of HBV-associated glomeruloDISCUSSION nephritis. At entry, mean ALT was 125.6 ? 74.4 U/L (mean k DDI has several characteristics that make it a novel S.D.). Liver biopsy specimens showed CAH in all six and potentially exciting treatment for patients with patients. AU patients had HBeAg in pretreatment chronic hepatitis B. DDI was chosen instead of other serum, and mean serum HBV DNA and HBV DNA dideoxynucleosides because of its availability and prepolymerase activity levels were 76 ? 61.2 pcg/ml and vious use in patients with HIV infection. Both HBV and HIV replicate through an RNA intermediate. This 2,691 2 2,708 cpdO.2 ml, respectively.
Vol. 16, No. 4, 1992
863
DDI THERAPY FOR CHRONIC HEPATITIS B
requires reverse transcriptase activity that, for HBV, produces DNA from the RNA template (7). DDI has been shown to inhibit the reverse transcriptase activity of HIV ( 5 ) and, therefore, could have a similar effect on hepatitis B. Experimental evidence in uiuo in an animal model of HBV infection has also provided support for this hypothesis. In the Pekin duck model of hepatitis B, four dideoxynucleosides (DDC, DDG, 'DDA and DDI) have been shown to rapidly decrease DNA polymerase activity and duck HBV DNA levels by 60% to 70% of pretreatment levels (8-10). In uitro studies using hepatocyte cell culture systems to test the antiviral effects of the dideoxynucleosides have yielded conflicting results. In studies using human hepatoblastoma cell lines, DDA and DDI exhibited a variable inhibitory effect on HBV replication in some cell lines but not in others (14,15).In contrast, in a primary culture system using duck HBV infected hepatocytes, DDA and DDG markedly inhibited duck HBV, whereas DDI was modestly effective (16). Despite the encouraging experimental data, the results of this study were disappointing. We treated six patients with chronic hepatitis B with DDI for up to 3 mo and found no consistent or appreciable effect on viral replication or disease activity. One patient who started the trial with the lowest levels of HBV DNA and DNA polymerase activity became negative for both these markers of viral replication, but he did not lose HBeAg in serum until 7 mo after completing DDI treatment. Even including HBV DNA and DNA polymerase from this patient, no statistically significant change occurred in mean serum HBV levels during the 3 mo of treatment with DDI at the highest dose evaluated. The reasons why DDI was ineffective in humans with chronic hepatitis B are not clear. The antiviral action of DDI against HIV appears to require successive phosphorylations to yield its 5'-triphosphate (ddA!J!P), which competes with the naturally occurring nucleotide for binding to DNA polymerase and, because ddATP is incorporated into the growing viral DNA chain, brings about termination of viral DNA synthesis and viral replication (4). Although DDI is efficiently phosphorylated in human lymphoid cells (171, DDI may be phosphorylated to a lesser extent within human hepatocytes, possibly because of lower levels of kinases. Higher levels of the naturally occurring nucleotide also may exist that can nullify the antiviral action of DDI. Finally, DDI may be rapidly catabolized within the human liver. Any of these reasons could explain the favorable viral inhibition demonstrated by DDI in duck HBV model systems (8,9,16) as compared with the limited effects in human HBV-infected cell lines (14, 15). In addition, human HBV may have different sensitivities to antiviral agents than duck HBV. Although both viruses have similar genomic structure and modes of replication, it remains to be proven whether the duck HBV model is appropriate for studying the human disease. It is also possible that the dosages of DDI used in this
O
z n 0
PRE
4
8
12
16
WEEKS
Rc.2. HBV DNA and DNA polymerase activity during long-term treatment with DDI (mean +- S.E.M.).
study were insufficient to inhibit HBV replication, especially because the mean peak plasma levels after similar dosages of DDI were lower in our patients with chronic hepatitis B than in patients with severe HIV infection (11). Patients with AIDS have been shown to have higher gastric pH and less stimulated gastric acid production than normal controls (18). Because DDI is acid labile, it is possible that DDI absorption was decreased in patients with chronic hepatitis B (and presumably normal gastric acid secretion) because of insufficient neutralization of gastric acidity by the oral antacids taken before DDI ingestion. Thus inhibition of HBV replication may require higher doses of DDI, perhaps in conjunction with other methods to suppress gastric acidity, or more stable preparations of DDI. It should be noted, however, that DDI toxicity is also dose related; side effects can be significant and potentially lethal (12, 13). In summary, despite the in uiuo and in vitro experimental evidence suggesting that DDI may be useful in treating chronic hepatitis B, we did not find a consistent, appreciable effect on viral replication in a cohort of patients with chronic hepatitis B. Additional studies would be needed to determine the reasons for lack of efficacy of DDI in these patients and to resolve the discrepancies seen with DDI in experimental models of HBV infection. REFERENCES 1. Hoofnagle JH, Alter HJ. Chronic viral hepatitis. In: Vyas GN,
Dienstag JL,Hoofnagle JH, eds. Viral hepatitis and liver disease. Orlando, FL: Grune L Stratton, 1984:97-113. 2. Hoofhagle JH, Peters M, Mullen KD, Jones DB, Rustgi V, Di Bisoeglie AM, Hallahan C, et al. Randomized controlled trial of
864
FRIED ET AL.
human alpha-interferon in patients with chronic hepatitis B. Gastroenterology 1988;95:1318-1325. 3. Perrillo RP, Schiff ER, Davis GL, Bodenheimer HC, and the Hepatitis Interventional Therapy Group. A randomized, controlled trial of interferon alfa-2b alone and after prednisone withdrawal for the treatment of chronic hepatitis B. N Engl J Med 1990;323:295-301. 4. McGowan JJ, Tomaszewski JE,Cradock J, Hoth D, Grieshaber CK, Broder S, Mitsuya H. Overview of the preclinicaldevelopment of an antiretroviral drug, 2',3'-dideoxyinosine. Rev Infect Dis 1990;12(suppl5):s513-s521. 5. Mitsuya H, Yarchoan R, Broder S. Molecular targets for AIDS therapy. Science 1990;249:1533-1544; 1986;83:1911-1915. 6. Yarchoan R, Mitsuya H, Thomas RV, Pluda JM, Hartman NR, Perno CF, Marczyk KS, et al. In vivo activity against HIV and favored toxicity profile of 2',3'-dideoxyinosine. Science 1989;245: 412-415. 7. Summers J, Mason WS. Replication of the genome of a hepatitis B-like virus by reverse transcriptase of an RNA intermediate. Cell 1982;29:403-415. 8. Kassianides C, Hoofnagle JH,Miller RH, Doo E, Ford H, Broder S, Mitsuya H. Inhibition of duck hepatitis B virus by 2',3'dideoxycytidine. Gastroenterology 1989;97:1275-1280. 9. Martin P, Kassianidw C, Korenman J, Hoofnagle JH, Ford H, Broder S, Mitsuya H. 2',3'-dideoxyinosine (DDI) and dideoxyguanosine are potent inhibitors of hepadnaviruses in viva [Abstract]. Gastroenterology 1989;96:A628. 10. Martin P, Kassianides C, Hoofnagle JH, Mitsuya H, Broder S. Effects of 2',3'-dideoxyadenosine on duck hepatitis B virus [Abstract]. HEPATOLOGY 1988;8:A1329.
HEPATOLOGY
11. Hartman NR, Yarchoan R, Pluda JM, Thomas RV, Marczyk KS, Broder S, Johns DG. Pharmacokinetica of 2',3'-dideoxyadenosine and 2',3'-dideoxyinosine in patients with severe human immunodeficiency virus infection. Clin Pharmacol Ther 1990;47:647-654. 12. Yarchoan R, Pluda JM, Thomas RV, Mitmya H, Brouwers P, WyviU KM, Hartman N, et al. Long-termtoxicity/activityprofile of 2',3'-dideoxyinosine in AIDS or AIDS-related complex. Lancet 1990;336:526-529. 13. Lai KK, Gang DL, Zawacki JK, Cooley TP. Fulminant hepatic failure associated with 2',3'-dideoxyinosine. Ann Intern Med 1991;115:283-284. 14. Ueda K, Tsurimoto T, Nagahata T, Chisaka 0, Matsubara K. An in vitro system for screening anti-hepatitis B drugs. Virology 1989;169:213-216. 15. Aoki-Sei S, O'Brien MC, Ford H, Fujii H, Gilbert DA, Cooney DA, Johns DG, et al. In vitro inhibition of hepatitis B virus replication by 2',3'-dideoxyguanosine, 2',3'-dideoxyinosine, and 3'-azido2',3'-dideomhymidine in 2.2.15(PR) cells. J Infect Dis 1991;164: 843-851. 16. Lee B, Luo W, Suzuki S, Robins MJ, Tyrrell DLJ. In vitro and in viva comparison of the abilities of purine and pyrimidine 2',3'dideoxynucleosides to inhibit duck hepadnavirus. Antimicrob Agents Chemother 1989;33:336-339. 17. Johnson MA, Fridland A. Phosphorylation of 2',3'-dideoxyinosine by cytosolic 5'-nucleotidase of human lymphoid cells. Mol Pharmacol 1989;36:291-295. 18. Lake-Bakaar G, Quadros EP, Beidas S, Elsakr M, Tom W, Wilson DE, Dinaoy HP, et al. Gastric secretory failure in patients with the acquired immunodeficiency syndrome (AIDS). Ann Intern Med 1988;109:502-504.
