HEPATOLOGY Vol. 18, No. 1, 1993

However, the authors failed to highlight other im- portant results in their pilot study, that should be in- corporated in the design of any future trial testing the validity of these preliminary findings. As noted by Gores and Steers in the editorial accompanying the above report (91, six of the eight HBsAg-positive pa- tients in this study died (three of fulminant hepatitis and three of recurrent HCC) with a median survival of only 12.5 mo. Also, none of the six patients with pathological evidence of tumor vascular invasion sur- vived without recurrence; the median survival of this group was just 11 mo. Patients known to have these features may best be prospectively excluded from future trials.

Other OLT/chemotherapy strategies are being tested. The University of Pittsburgh is using both pre-OLT intraarterial chemotherapy and post-OLT systemic che- motherapy in patients with stage 3 or 4 HCC. The treatment consists of intraarterial cisplatin and doxo- rubicin before “listing” for transplant, then the same chemotherapy agents given systemically after OLT. Of 11 patients who had OLT, 9 are alive and free of cancer after 1 yr, although 2 have had recurrences during their second year of follow-up. Complications include a sig- nificant worsening of underlying hepatitis in two pa- tients who did not undergo OLT (10). These same investigators question whether OLT may be preferable to hepatectomy even in patients with resectable HCC (11).

A prospective trial at the University of California-San Francisco involves tumor chemoembolization ( 12) fol- lowed by OLT in a highly selected group of patients (13). Eligibility criteria exclude patients who are HBsAg positive, have more than three tumors or have any tumor larger than 5 cm. Eleven patients with histolog- ically proved HCC have been enrolled in the study: eight proceeded to OLT and are alive without recurrence at a median follow-up of 23 mo; two developed metastases before OLT and died; and one patient’s tumor was rendered resectable. These results could, of course, reflect rigorous patient selection rather than a salutory role for chemoembolization.

The above-summarized trial and references suggest that some patients with HCC may benefit from OLT, and they offer various strategies that could be pursued toward that goal. Rigorous prospective studies need to be done for proof that OLT does have a role in patients with HCC, with or without adjunctive therapy. Such studies will require many patients, stratified for many variables, enrolled in strictly applied protocols that will only be successfully completed through a nationwide, cooper- ative effort.

If such an undertaking is to be attempted, however, it must be done in the context of the current flurry of philosophical questions. Given the relative scarcity of donor livers and the precarious status of health care finances, can we pursue such cost-intensive studies? If so, what percentage of patients must be cured (or have improved quality of life for a long-term period) to make

HEPATOLOGY Elsewhere 219

this heroic intervention cost-effective? In the end, in which if any patients with HCC is that level of success likely to be achieved?

ALAN P. VENOOK, M.D. Division of HematologylOncology University of California-San Francisco San Francisco, California 941 43

REFERENCES Smalley SR, Moertel CG, Hilton JF, Weiland LH, Weiand HS, Adson MA, Melton LJ 3d et al. Hepatoma in the non-cirrhotic liver. Cancer 1988;62:1414-1424. Malt FL4. Surgery for hepatic neoplasms. N Engl J Med 1985;313:

Zhou XD, Tang ZY, Yu YO, Yang BH, Lin ZY, Ma ZC, Tang CL. Long-term survivors after resection for primary liver cancer: clinical analysis of 19 patients surviving more than ten years. Cancer 1989;63:2201-2206. Lewis BJ, Friedman MA. Current status of chemotherapy for hepatoma. In Ogawa M, ed. Chemotherapy of hepatic tumors. Princeton, NJ: Excerpta Medica, 1984:63-74. Bismuth H, Castaing D, Ericzon BG, Otte JB, Rolles K, Ringe B, Sloof M. Hepatic transplantation in Europe. First report of the EuroDean Transplant RePistrv. Lancet 1987:2:674-676.

159 1-1596.

6. Starzi TE, Iwatsbki S, Shaw BW Jr, Nalesnik MA, Farhi DC, Van Thiel DH. Treatment of fibrolamellar hepatoma with partial or total hepatectomy and transplantation of the liver. Surg Gynecol Obstet 1986;162:145-148.

7. Jenkins RL, Pinson CW, Stone MD. Experience with transplan- tation in the treatment of liver cancer. Cancer Chemother Pharmacol 1989;23(suppl):S 1044 109.

8. Huang CC, Wu MC, Xu GW, Li DZ, Cheng H, Tu ZX, Ziang HQ, et al: Overexpression of the MDRl gene and P-glycoprotein in human hepatocellular carcinoma. J Natl Cancer Inst 1992;84:

9. Gores GJ, Steers JL. Progress in orthotopic liver transplantation for hepatocellular carcinoma. Gastroenterology 1993; 104:

262-264.

317-320. 10. Carr B. Transplant Proc, in press. 11. Iwatsuki S, Starzl TE, Sheahan DG, Yokoyama I, Demetris AJ,

Todo S, Tzakis AG, et al: Hepatic resection versus transplantation for hepatocellular carcinoma. Ann Surg 1991;214:221-229.

12. Venook AP, Stagg RJ, Lewis BJ, Chase JL, Ring EJ, Maroney TP, Hohn DC. Chemoembolization for hepatocellular carcinoma. J Clin Oncol 1990;8:1108-1114.

13. Venook AP, Lake JR, Roberts JP, Wright T, Emond J, Ferrell LD, Frye JW, et al. Orthotopic liver transplantation (OLT) in patients with hepatocellular carcinoma (HCC): results using preoperative chemoembolization (CE) [Abstract]. HEPATOLOGY 1992;16:45A.

THE THREAT OF HEPATITIS B VIRUS RECURRENCE: A SWORD OF DAMOCLES TO THE

LIVER TRANSPLANT RECIPIENT

Lucey MR, Graham DM, Martin P, Di Bisceglie A, Rosenthal S, Waggoner JG, Merrion RW, Campbell DA, Nostrant TT, Appleman HD. Recurrence of hepatitis B and delta hepatitis after orthotopic liver transplan- tation. Gut 1992;33:1390-1396.

ABSTRACT

The clinical course of 10 liver transplant recipients who had hepatitis B virus (HBV) and five recipients with HBV and D (delta) infection before transplan- tation is described. Six patients who underwent eight transplants died. The estimated one and two year-

220 HEPATOLOGY Elsewhere HEPATOLOGY July 1993

survival rates in patients with HBV only before trans- plantation were 74% and 67% respectively. The esti- mated one and two year survival in patients with HBV and HDV infection beforehand was 100%. Graft in- fection by HBV occurred in 8 of 10 patients infected with HBV only; and in 4 of 5 patients with previous HBV and HDV infection. There was a widely variable time from transplantation to the appearance of HBV markers in liver or serum, ranging from 6-331 days. Hepatitis D antigen (HDAg) appeared in three grafts very rapidly after transplantation at 4, 8, and 37 days respectively. Graft infection by HBV was accompanied by significant liver injury in six allografts in five recipients. In particular, there was a striking morpho- logical appearance in five infected livers in which the hepatocytes became progressively enlarged and dis- torted as they accumulated huge amounts of hepatitis B surface and core antigens (HBsAg, HBcAg). These features were accompanied by pericellular fibrosis and cholestasis but little associated inflammation. This syndrome carried a poor prognosis. A gradual pro- gression to cirrhosis occurred in one additional liver. Finally, recurrent HBV infection was a principal or a contributing factor in all deaths. The presence of HBcAg and inflammation in the native liver increased the risk of HBV induced tissue damage in the graft whereas HDV infection in the host liver seemed to reduce the risk of significant HBV induced tissue damage in the allograft. These data suggest that post transplant HBV infection is accompanied by a variety of changes in the liver allograft, some of which are unique to the transplanted liver and may result in impaired allograft function.

Benner KG, Lee RG, Keefe EB, Lopez RA, Sasaki AW, Pinson CW. Fibrosing cytolytic liver failure secondary to recurrent hepatitis B after liver transplantation. Gas- troenterology 1992; 103: 1307-1 3 12.

ABSTRACT

Four patients who underwent transplantation for hepatitis B virus-related liver disease developed rapidly progressive liver failure attributable to re- current hepatitis B disease typified by hyperbiliru- binemia and distinctive hepatocyte ballooning and progressive fibrosis consistent with recently reported fibrosing cholestatic hepatitis. Among these four pa- tients, the mean interval from transplantation to redocumentation of hepatitis B surface antigen (HBsAg) and hepatitis B e antigen (HBeAg) was 5 months, to development of malaise and jaundice 6 months, to histological diagnosis 7 months, and to graft failure 8 months. The only patient who underwent retransplantation had accelerated recurrence of the same syndrome with biopsy documentation 1 month later and graft failure 2 months later. Distinctive histological features included confluent hepatocel- lular ballooning and progressive periportal fibrosis followed by lobular collapse over 4-6 weeks without significant inflammation. Immunohistochemical stain- ingshowed marked HBsAg and hepatitis B core antigen (HBcAg) immunoreactivity. The rapid development of cytolytic hepatocellular necrosis and lobular collapse with prominent HBcAg immunoreactivity without sig-

nificant inflammation suggests a cytolytic rather than immune pathogenesis for this unique and devastating form of recurrent hepatitis B that might better be termed “fibrosing cytolytic hepatitis.”

COMMENTS

Recurrent HBV infection is one of the most lethal complications encountered in the liver transplant pop- ulation. This is illustrated by these two reports. The article by Lucey et al. describes the outcome of 15 patients who underwent liver transplantation for HBV infection with or without simultaneous hepatitis delta virus (HDV) infection (1). Graft reinfection occurred in 8 of 10 patients with HBV infection alone and in 4 of 5 with both HBV and HDV. Five patients in the former group died of liver failure related to recurrent infec- tion-four of these five with the histological picture known as “fibrosing cholestatic hepatitis” (FCH). In contrast, all five patients with both HBV and HDV infection were alive, despite recurrence in four. FCH did not develop in any patients in the latter group.

The article by Benner et al. focused specifically on four patients with FCH and described in detail the clinical and histological features of this syndrome (2). What do these two articles tell us about the clinical course and pathogenesis of this syndrome? First, that it is uni- formly fatal. Eight of eight patients died, despite retransplantation in two (FCH developed in the second grafts). Second, marked cholestasis with hyperbiliru- binemia and hypoprothrombinemia rapidly develop in patients (mean maximum bilirubin, 27.2 mg/dl; mean prothrombin time prolongation, 21 sec) (2). Third, FCH is associated with active pretransplant HBV replication (three of four patients in Benner’s study were HBeAg positive). Fourth, a common histological pattern is seen with ballooning degeneration of pericentral hepatocytes, followed by periportal fibrosis and strong expression of viral proteins (HBcAg and HBsAg). Most striking of the histological findings is paucity of portal or lobular inflammation despite rapid progression of hepatocyte ballooning and fibrosis. The lack of inflammatory re- sponse invokes the possibility that FCH is related to a direct cytopathic effect of HBV on liver cells. Hence Benner proposes that this syndrome be termed “fibros- ing cytolytic hepatitis.”

How common is this entity? FCH was seen in 4 of 10 patients who underwent transplantation for chronic HBV in Michigan and 3 of 6 patients in Oregon. Two other groups have reported FCH in HBV-infected patients; it occurred in 2 of 35 patients at the University of Pittsburgh and 6 of 27 patients at King’s College Hospital, London (3,4). With all four studies combined, FCH developed in 15 of 78 patients or 19% of those who underwent transplantation for either acute or chronic HBV infection. The finding of FCH is a particularly ominous sign, leading to graft failure and death in all patients (mean interval from onset of jaundice to death or retransplantation was 38 days and range was 4 to 7 wk) (2). When FCH occurs, retransplantation is futile because the syndrome recurs in the new graft (1-3).

HEPATOLOGY Vol. 18, No. 1, 1993 HEPATOLOGY Elsewhere 221

However, it must be emphasized that FCH occurs in the minority of patients with recurrent HBV infection.

What do we know about the mechanisms of liver damage in recurrent HBV infection and, more specifi- cally, about the pathogenesis of FCH? In immunocom- petent patients liver damage from chronic HBV in- fection is believed to be largely immune mediated, with the immune “target” being HBcAg expressed on the hepatocyte (5). Classic dogma has it that for immune recognition, viral peptides in association with class 1 human leukocyte ( H I 4 antigens on the target cell must be presented to cytotoxic (CD8) T cells expressing identical HLA antigens (5). This is rarely possible in liver transplantation because the organ donor is not HLA- matched to the recipient. Because the host-viral inter- actions are so radically altered in the liver transplant patient, experimental results from immunocompetent individuals likely do not apply, and other factors must be important. Why can HBV infection be so devastating in the setting after liver transplantation? Undoubtedly, part of the answer relates to immunosuppression. With immunosuppressive therapy other patients (such as kidney transplant recipients and those on cancer che- motherapy) have reactivation of quiescent HBV in- fection, which may result in an accelerated clinical course (6, 7). Steroids appear to be the main offenders. Prednisolone has long been known to have a deleterious effect on chronic HBV infection (8). The HBV genome contains a steroid-responsive promotor region that activates HBV replication (91, and prednisolone, but not cyclosporine or azathioprine, increases HBsAg ex- pression by infected hepatocytes in primary culture (10). These observations explain in part why these patients have high levels of HBV replication and may also guide the use of immunosuppressive therapy, but they do not yet tell us how the virus is destroying the liver.

This brings us to another hypothesis discussed by both Lucey and Benner; namely, that the virus itself or virus-associated proteins are toxic to the cell. Although attractive as an explanation, supportive experimental evidence is circumstantial. FCH is at the extreme end of the spectrum of HBsAg-associated liver damage, but it might serve as a model for all virus-associated liver damage in this population. Both Lucey and Benner demonstrated excess hepatic production of viral proteins (with HBsAg in the cytoplasm and HBcAg in both cytoplasm and nucleus) (1,2). Intrahepatic expression of hepatitis B viral proteins (HBsAg, HBcAg and pre- surface proteins) in patients with FCH has previously been shown to be higher than that in transplant patients without FCH or nontransplant patients with chronic HBV infection (11). In contrast, no relationship has been found between titers of viral proteins (HBsAg and pre-S1 protein) in serum and the degree of liver damage. Thus it has been proposed that a defect occurs in excretion of surface proteins from the cell and that the accumulation of viral proteins within the endoplasmic reticulum is directly cytotoxic (11). Much interest has focused on the inhibitory effect of pre-S1 proteins of the secretion of HBsAg (12). Pre-S1 proteins consist of the

surface domain plus an additional 163 amino acids on the N-terminus end. Transfection of mammalian cell lines with the surface gene either alone or in combi- nation with pre-S1 sequences has shown that the pre-S1 protein markedly inhibits secretion of HBsAg from the endoplasmic reticulum (12). Because the degree of inhibition appears to be proportional to the amount of pre-S1 protein present, this effect may only be physio- logically important in a state of massive overproduction of these proteins (such as in the patient with recurrent HBV infection after liver transplantation). A direct cytopathic role for viral proteins has also been suggested by experiments with transgenic mice, genetically manip- ulated to overproduce “large” surface antigens (which include both surface and pre-S proteins) (13). Although production of HBsAg, even in the absence of intact virus, can result in hepatocyte necrosis and even HCC, the possibility of T-cell-mediated hepatic damage in this intact animal model cannot be excluded (13).

Given the almost universal demise in patients in whom FCH developed, it is incumbant on those involved in selection of patients to identify pretransplant char- acteristics predictive of this lethal outcome. Although HBV recurs in almost 100% of patients, progressive liver disease will not develop in all. In some, a more indolent course is found, with little in the way of graft damage. The number of patients studied by Lucey and Benner is small, but they both found that active pretransplant replication (documented by HBeAg, HBV DNA or HBcAg staining of the native liver) and “active cir- rhosis” of the native liver predict more aggressive posttransplant liver disease (1, 2). Because previous studies have found posttransplant recurrence is partic- ularly devastating in patients with active pretransplant replication, most centers have abandoned liver trans- plantation in HBeAg-positive patients (4, 14). The converse is also true; that is, patients with little pretransplant HBV replication have a better posttrans- plant outcome. Delta virus is a naturally occurring inhibitor of HBV replication. We know from Lucey’s report and from reports of others that patients with simultaneous HBV and HDV infection do better with liver transplantation (4, 14, 15). In addition, one of the goals of peritransplant immunoprophylaxis with hepa- titis B immunoglobulin is to reduce circulating levels of virus (141, and with this therapy reinfection is delayed even in those who are HBeAgpositive (14). Why does the pretransplant replicative state influence posttransplant outcome? Because circulating virions are the likely source of allograft infection, the size of the inoculum may determine the rapidity with which the virus spreads horizontally from cell to cell in the new liver. Thus HBeAg-positive patients with high pretransplant levels of circulating Dane particles may be predisposed to more efficient replication in the early posttransplant period, the time of maximal immunosuppression. Alternatively, patients with active pretransplant replication may have an inherent host-immune defect resulting in failure of control of viral replication both before and after trans- plantation.

222 HEPATOLOGY Elsewhere HEPATOLOGY July 1993

Although the threat of recurrent infection and death from liver failure is hanging “like the sword of Damocles” over the head of the HBV-infected liver transplant recipient, identification of pretransplant pre- dictors should aid in selection of patients who are at reduced risk of having the devastating syndrome of FCH develop. This information, in conjunction with more powerful antiviral agents and improved understanding of the effects of specific immunosuppressives on HBV replication, will allow us to overcome the greatest challenge in liver transplantation.

TERESA WRIGHT, M.D. Department of Veterans Affairs Medical Center

San Francisco, California 94143 and the University of California-San Francisco

REFERENCES 1. Lucey MR, Graham DM, Martin P, Di Bisceglie A, Rosenthal S,

Waggoner JG, Merion RM, et al. Recurrence of hepatitis B and delta hepatitis after orthotopic liver transplantation. Gut 1992;33:

2. Benner K, Lee RG, Keeffe EB, Lopez RR, Sasaki AW, Pinson CW. Fibrosing cytolytic liver failure secondary to recurrent hepatitis B after liver transplantation. Gastroenterology 1992;103: 1307- 1312.

3. Davies SE, Portmann BC, O’Grady JG, Aldis PM, Chaggar K, Alexander GJM, Williams R. Hepatic histological findings after transplantation for chronic hepatitis B virus infection, including a unique pattern of fibrosing cholestatic hepatitis. HEPATOLOGY

4. Todo S, Demetris AJ, Van Thiel DH, Teperman L, Fung JJ , Starzl TE. Orthotopic liver transplantation for patients with hepatitis B virus-related liver disease. HEPATOLOGY 1991; 13:619-626.

1390-1396.

1991;13:150-157.

5. Peters M, Vierling J , Gershwin ME, Milich D, Chisari FV, Hoofnagle JH. Immunology and the liver. HEPATOLOGY 1991;13: 977-994.

6. Parfrey PS, Forbes RDC, Hutchinson TA, Kenick S, Farge D, Dauphinee WD, Seely JF, et al. The impact of renal transplan- tation on the course of hepatitis B liver disease. Transplantation

7. Lok ASF, Liang RHS, Chiu EKW, Wong K-L, Chan T-K, Todd D. Reactivation of hepatitis B virus replication in patients receiving cytotoxic therapy. Gastroenterology 1991;100:182-188.

8. Lam KC, Lai CL, Ng RP, Trepo C, Wu PC. Deleterious effect of prednisolone in HBsAg-positive chronic active hepatitis. N Engl J Med 1981;304:380-386.

9. Tur-Kapsa R, Burk RD, Shaul Y, Shafritz DA. Hepatitis B virus contains a glucocorticoid-responsive element. Proc Natl Acad Sci

1985;39:610-615.

USA 1986;83:1627-1631. 10. Lau JYN. Bain VG. Smith HM. Alexander GJM. Williams R.

Modulation of hepatitis B viral antigen expression by immuno- suppressive drugs in primary hepatocyte culture. Transplantation

11. Lau JYN, Bain VG, Davies SE, O’Grady JG, Alberti A, Alexander GJM, Williams R. High-level expression of hepatitis B viral antigens in fibrosing cholestatic hepatitis. Gastroenterology 1992;

12. Persing DH, Varmus HE, Ganem D. Inhibition of secretion of hepatitis B surface antigen by a related presurface polypeptide. Science 1986;234:1388-1391.

13. Chisari FV, Klopchin K, Moriyama T, Pasquinelli C, Dunsford HA, Sell S, Pinkert CA, et al. Molecular pathogenesis of hepatocellular carcinoma in hepatitis B virus transgenic mice. Cell 1989;59: 1145- 1156.

14. Samuel D, Bismuth A, Mathieu D, Arulnaden J-L, Reynes M, Benhamou J-P, Brechot C, et al. Passive immunoprophylaxis after liver transplantation in HBsAg-positive patients. Lancet 1991;

15. Ottobrelli A, Marzano A, Smedile A, Recchia S, Salizzoni M, Cornu C, Lamy ME, et al. Patterns of hepatitis delta virus reinfection and disease in liver transplantation. Gastroenterology 1991;lOl: 1649- 1655.

1992;53:894-898.

102:956-962.

337~813-815.