130A AASLD ABSTRACTS HEPATOLOGY October 1995

93 IMMUNE MECHANISMS IN LIVER GRAFT INJURY FOLLOWING HEPATITIS B VIRUS RECURRENCE AFTER ORTHOTOPIC LIVER TRANSPLANTATION (OLT). G Marinos. S Rossol. P Carueci. PYW Wone. PT Donaldson. MJ Hassein*. D Vergani*. NV Naoumov. Roger Williams. Institute of Liver Studies and *Dept. of Immunology, King's College School of Medicine & Dentistry, London SE5 9PJ, UK.

Background: Recurrence of HBV after OLT is charaeterised by abundant expression of intrabepatic HBV antigens. In HBsAg transgenie mice, TNFa and IFN3, have been shown to be important mediators in the iysis of HBsAg containing hepatocytes. Aim: We studied the antigen specific and non specific immune mechanisms responsible for liver graft injury following HBV recurrence in patients with varying degrees of HLA mismatched graft. Methods: 12 patients with HBV recurrence following OLT, 16 patients with chronic hepatitis B, and 16 patients with OLT for HBV cirrhosis without recurrence were included. We evaluated the HBeAg-speeific, CD4+ HLA class II restricted response by lymphocyte proliferation (~H,Thymidine uptake) and in vitro IFN3, production (ELISA); serum levels of soluble TNF receptors (TNFR-55 and TNFR-75; ELISA); hepatic expression of IFN% CD4/CD8 lymphoeytes, macrophages (CD68 +), TNF receptors, HBcAg and HBsAg by immunohistochemistry; and HLA class I/II mismatch score. Results: Patients with HBV recurrence and active inflammation following OLT demonstrate a significant CD4 proliferative response (9/12) comparable to patients with chronic HBV (12/16). Liver biopsies showed a predominance of CD4+ lymphocytes in the portal tracts in 12/13 patients with HBV recurrence however, the hepatic expression of IFN7 was weak and seen only in 4/13. Serum levels of TNF receptoi's were markedly elevated in HBV recurrence (mean TNFR-55 5.94-3, TNFR-75 22.9 + 8.4 ng/ml; normal range 0.8+0.2 and 3.7+0.6) in contrast to patients transplanted for HBV but without recurrence (TNFR-55 1.4_+0.9 and TNFR-75 6.3_+3.3 ng/ml) and CLD patients (mean TNFR-55 i .6+0.4 and TNFR-75: 7.6-+ 1.2 ng/ml). Hepatic expression of both TNFRs was strongly enhanced in HBV recurrence and correlated well with the extent of CD68+ macrophage infiltration (1)<0.05) and histological activity (p < 0.01). There was no correlation between the HLA class I/II mismatch score and the degree 0f graft injury. Conclusion: Despite immunosuppression, OLT rec!pients with HBV recurrence can mount a HBcAg-speciflc CD4 response sufficient to activate the TNF/TNF receptor system. Accumulation of tissue macrophages and augmentation of the TNF/TNF receptor system in patients with HBV recurrence after OLT is an important factor in liver graft injury.

94 LIVER TRANSPLANTATION FOR PATIENTS WITH HEPATITIS DELTA CIRRHOSIS AND ACTIVE REPLICATING HEPATITIS B HAVE A POOR OUTCOME. Willem Marsman, Michael J. DeBernardi, J. Eileen Ha,/, Jeffery L. Steers, Ruud A.F. Krom, Russell H. Wiesner. Liver Transplantation, Mayo Clinic and Mayo Foundation, Rochester, Minnesota 55905.

Liver transplantation for hepatitis B virus (HBV) relate d liver disease is complicated by t-IBV recurrence and, consequently, poor patient and graft survival. Patients transplanted for hepatitis delta virus (HDV) with" cirrhosis are reported to have a diminished incidence of hepatitis B recurrence and improved graft survival. Only a few of these reported HI)V-infected patients had active HBV replicative disease before liver transplantation. In our experience, we transplanted t w o HDV-infected patients, both of which were hepatitis Be antigen and HBV-DNA positive before transplantation. In one patient HBsAg recurred four months after transplantation. Two months later HBe antigen and HBV, DNA became positive and the patient died of recurrent hepatitis B fulminant hepatitis. In the other patient, HBV persisted after transplantation, and two months thereafter the patient required retransplantation because of recurrent fulminant B hepatitis. The patient remained HBV flee for one year with the second graft. At this time the patient experienced HBV recurrence with active replication and died soon thereafter of HBV fulminant hepatitis. HDV persisted during the entire course of both patients. In the first case and in the second graft of the second case I-IBIG was administered as immunopr0phylaxis in an attempt to prevent recurrence Of HBV. The literature suggests that HDV inhibits the replication ,of I-IBV and therefore plays a role in preventing the recurrence of HBV and improving survival. On the contrary, in our experience active HBV replication despite HDV superinfection appears not to play a protective rote. Conclusion: HDV in patients with HBV active replication appear to have a poor prognosis.

95 EFFICACY OF GANCICLOVIR IN THE TREATMENT OF HEPATITIS B VIRUS INFECTION OF THE LIVER GRAFT. LONG-TERM RESULTS. M de la Mata, JL Montero, E Fraga, G Costfin, M Delgado, F L6pez, R Gonz~lez, C Pera, G Mifio. H. Reina Sofia.C6rdoba, Spain.

Hepatitis B virus (HBV) reinfection of the graft occurs in 30-50% after liver transplant (LT) for HBV liver disease, with graft failure being common cause of death. In addition, some patients may develop "de novo" HBV infection, acquired from the donor liver. Recent reports have suggested that ganciclovir may have a direct antiviral effect against HBV.

Fifteen patients with HBV graft infection were treated with IV ganciclovir (10 mg/Kg/day) for a 21 day period. In group I (9 patients)HBV recurred postransplant. All patients but one in this group had become HBsAg (-) while receiving IM hiparimmane HBV gammaglobulin until recurrence at a mean of 194 days post-LT. Group II with "de novo" HBV infection, made apparent at a mean of 482 days post-LT, included 6 cases.

Mean pretreatment HBV--DNA level was 361 ± 111 pg/ml) (group 1:414 ±164, group 1I:273±130). Mean HBV-DNA level after 21 days of treatment was 103.7±58 pg/ml (t)=0.053) (Group I:125±91, group 11:64-+31). Ganciclovir was well tolerated, with no significant adverse event.

However, 3 months post-treatment HBV-DNA concentration raised to a mean value of 336.3±151 pg/ml (Group 1:179±156, group 1I:493---257). No changes in the serum HBV marker profile was observed. Four patients in group I and 3 patients in group II received a second course of 3 weeks IV Ganciclovir with a new decrease of viral activity.

After a mean follow-up of 40 months, current survival in group I is 77.7% and in group II 83%. Two patients in group l died from severe liver failure. One patient in group II died because of pulmonary hypertension. Three patients in group I and one in group I1 became HBV-DNA (-) in serum. For those remaining HBV-DNA (+) current mean level is 147 pg/ml (range 17- 358). All patients but one in group I and every patient in group 11 maintain nomaal liver fimction. In conclusion: 1) Ganciclovir is an effective and safe inhibitor of HBV replication in the liver graft, both in the setting of recurrent disease and acquired HBV infection 2) A prospectively controlled study and a more proinngued duration of treatment is warranted.

96 ALTERED REGULATORY MECHANISMS WHICH INFLUENCE THE PROTEIN COMPOSITION OF THE HBV-ENVELOPE BEFORE AND AFTER OLT. Trautwein, C., Schrem, H., Tillmann, H., Kubicka, S., Walker, D., B6ker, K., Pichlmayr, R. and Manns, M.P. Medizinische Hochschule, Hannover, FRG.

The domain composition of the hepatitis B envelope protein (HBsAg) is regulated on the transcription level: two promoter s (pre$ and S) give rise to mRNAs (2,4 and 2,1 kb) which have up to three in frame AUGs in the env-ORF and on the translation level: Kozak-sequences, AUG and hairpin-formation. Mutations in these regions have fundamental impact on the relative epitope composition of the viral envelope. Altered immune response after liver transplantation (OLT) and reinfection with HBV changes selection pressure and could alter the assembly of the viral surface.

20 patients treated with OLT for HBV-related liver disease complicated by reinfection of the graft under immunosuppression and anfi-HBs were included into this study. HBV-DNA was extracted and the preS-region was amplified via PCR from HBsAg positive sera collected before and after OLT. In order to analyze also minor virus populations PCR-pmduots were subcloned. Sequencing was performed of at least four clones, if no difference in insert size was detected. Clones with different insert size were sequenced additionally. Computer-assisted analysis was performed on all obtained sequences for hairpin formation 5-prime of the AUG, Kozak sequences, loss of AUG and primary structure of the S-promoter.

According to serotype classification promoter mutations were found in all patients infected with ayw (n=6), but only in 43 % (6/14) with adw2 virus populations. Deletion of the S-promoter was found in two patients in 1(30 % of the virus population exclusively pre-OLT and in one case in a minor virus population (25%) only post-OLT. 9 patients show different point mutations in 100 % of the virus population pre- and Post-OLT. Interestingly patient 17 showed 15 point mutations in the promoter region and destruction of the CCAAT-motif, which is important for regulating the transcriptional ratio between the 2,4 and 2,1 kb mRNAs with implications for virus secretion. Here this mechanism is impaired and may explain the ballooning cholestatic hepatitis which lead to deterioration of liver function and death of the patient 6,5 months after OLT. Loss of AUG was only found for the preS2 domain in 35 % (7/20) of the patients. Significant hairpin-formation 3' of the cap-site of the 2,1 kb mRNA was not found; only two patients showed changes in the Kozak-sequences post-OLT in the minor virus populations (33%). Destruction of protective epitopes was found in the preS1 and preS2 regions in 8 patients pre-OLT with reemergence of the wt virus post-OLT and in two patients post-OLT.

We describe mutations which lead to altered transeript~onal and translational regulation as well as qualitative mutations in protective ep~topes of the a n y -gene of HBV. Changes are mainly found before OLT, while wt preS-DNA-sequences reappear after reinfection when selective pressure due to immunosuppression is reduced. Anti- HBs does not induce increased numbers of mutations in the preS genome.