Are preformed antibodies to biliary epithelial cells of clinical importance in liver transplantation?

Are Preformed Antibodies to Biliary Epithelial Cells of Clinical Importance in Liver Transplantation? ¨ hrstro¨m,* Xupeng Ge,* Bo-Go¨ran Ericzon,* Grzegorz Nowak,* Henrik O Ulrika Broome´,† and Suchitra Sumitran-Holgersson* During acute liver allograft rejection, most of the tissue damage to bile duct epithelium is thought to occur as a consequence of direct immunologic injury by T-cell–mediated immune effector mechanisms. However, the role of antibodies to biliary epithelial cells (BECs) in liver transplant rejection is not known. We therefore investigated cross-match sera obtained immediately before liver transplantation from 95 patients for the presence of BECreactive antibodies to determine their association with acute rejection. BECs were isolated from one normal healthy liver. Antibody binding was detected by using flow cytometric analysis. Donor lymphocyte–specific cross-matches using complement-dependent cytotoxicity (CDC) and flow cytometric assays also were performed. The 2-year patient survival rate in this study was 86.3%. Eleven patients were positive for either CDC or flow cytometric cross-matches. BEC antibodies were detected in 41 serum samples (43.2%). Patients with BEC antibodies experienced acute rejection more frequently (65.9%) compared with 42.5% without antibodies (P < .03). HLA specificity determinations indicated that in 5 of 41 cases, anti-BEC reactivity was caused by HLA antibodies. No correlations between the presence of BEC antibodies and patient survival and the occurrence of cholangitis and nonsurgical bile duct strictures were found within 2 years of follow-up. In conclusion, preformed antibodies to BECs are associated with acute rejection. Thus, the presence of these antibodies before transplantation may facilitate acute liver graft rejection. (Liver Transpl 2003;9: 1191-1198.)

shown that the presence of antibodies to epithelial cell line A549 was associated with a decrease in graft survival and with postoperative infections in lung transplantation.21 However, no similar studies have been performed in liver transplantation. It is well established that liver transplants are more resistant to hyperacute rejection caused by preformed HLA antibodies than kidney and other organ allografts, and HLA matching neither correlates with acute rejection nor influences liver transplant outcome.22 Therefore, the liver is considered to be an immune-privileged organ. Biliary tract problems constitute one of the most frequent sources of morbidity after human liver transplantation,23,24 and biliary epithelium is a major site of damage in liver allograft rejection.25,26 In autoimmune liver diseases, it has been shown that autoantibodies may have an important role for bile duct injury.27 The aim of the present study is to investigate whether preformed biliary epithelial cell (BEC)-reactive antibodies have a detrimental effect on liver allografts.

Material and Methods Patient Population

I

t generally is believed that humoral immunity is the cause of hyperacute rejection and cellular immunity is the basis of acute rejection.1,2 Donor lymphocyte and recipient serum cross-match by complement-dependent cytotoxicity (CDC) test, flow cytometry, or enzyme-linked immunosorbent assay is the routine screening method used to avoid the occurrence of hyperacute rejection.3-5 Optimal HLA matching has significantly reduced acute rejection episodes in kidney transplantation during the past decade.6,7 However, there is increasing evidence that humoral processes also may have an important role in acute allograft rejection. Efforts have been made to evaluate the clinical significance of preformed antibodies in acute rejection of both liver8-11 and other organ transplants.12-15 The role of HLA antibodies has been studied,16,17 but results are still controversial.18-20 The significance of non-HLA antibodies12,13,15,21 in heart, lung, and kidney transplants also was studied. It was

Between January 1999 and December 2000, a total of 95 consecutive primary liver transplantations performed at our center were included in this study. Mean age was 47.6 years (range, 6 months to 68 years), and 40% of patients are women. Indications for liver transplantation were mainly end-stage chronic liver diseases (Table 1). Recipient-donor

From the Departments of *Transplantation Surgery and †Gastroenterology, Huddinge University Hospital, Karolinska Institute, Stockholm, Sweden. Supported in part by grant no. K2002-06X-14004-02B from the Medical Research Council (S.S.-H.) and a grant from the Lars-Erik Gelins Foundation (S.S.-H.). Address reprint requests to Xupeng Ge, MD, Department of Transplantation Surgery, B56, Huddinge University Hospital, S-141 86 Stockholm, Sweden. Telephone: ⫹468-5858-2568; FAX: ⫹468-7743191; E-mail: [email protected] Copyright © 2003 by the American Association for the Study of Liver Diseases 1527-6465/03/0911-0010$30.00/0 doi:10.1053/jlts.2003.50236

Liver Transplantation, Vol 9, No 11 (November), 2003: pp 1191-1198

1191

1192

Ge et al

Donor Lymphocyte-Specific Cross-Matches Using the Microcytotoxic and Flow Cytometric Assays

Table 1. Diagnosis

Indications Hepatitis C cirrhosis PSC* Familial amyloidotic polyneuropathy PBC* AIH* Hepatitis B cirrhosis Alcoholic cirrhosis Others Total

No. of Patients

BEC Antibody– Positive Patients

22 18

7 (31.8) 10 (55.6)

14 8 7 7 7 12 95

3 (21.4) 6 (75.0) 4 (57.1) 4 (57.1) 5 (71.4) 2 (16.7) 41 (43.2)

NOTE. Values expressed as number (percent). *Autoimmune liver diseases.

selection was based on blood group A, group B, group O (ABO) compatibility and size matching. No attempts were made for HLA matching. Follow-up time for every patient was 2 years. Occurrence of acute rejection episode, cholangitis, and nonsurgical bile duct stricture was investigated. Blood total bilirubin and ␥-glutamyl transpeptidase (GGT) values at such different times as 1 month, 3 months, 1 year, and 2 years also were collected and compared. Serum samples obtained just before liver transplantation from all patients were stored at ⫺20°C for the detection of BEC antibodies. The immunosuppressive protocol for most patients (n ⫽ 71) was based on tacrolimus/FK506 (Prograf; Fujisawa, Munich, Germany) and steroids, with or without mycophenolate mofetil (Cellcept; Roche, Stockholm, Sweden). For the other 24 patients, cyclosporine A– based (Neoral; Novartis, Basel, Switzerland) triple immunosuppression with steroid was used. Acute rejection episodes were diagnosed by clinical signs and histopathologic confirmation. In some rejection episodes, biopsy was contraindicated or not available for logistic reasons, but all these patients received standard rejection treatment based on clinical and biochemical data and responded to the treatment. Standard treatment for acute rejection episodes in this study was a 1,000-mg bolus dose of intravenous methylprednisolone (Solu-Medrol; Pharmacia & Upjohn Co, Stockholm, Sweden), followed by a full recycling of prednisolone (i.e., tapering daily dose from 200 to 20 mg). In the case of steroid-resistant rejection, treatment was switched to tacrolimus/FK506 (Fugisawa) for cyclosporinetreated patients or treated for 7 to 14 days with OKT3 (Ortho-Biotech Inc, Raritan, NJ). The diagnosis of cholangitis required the presence of fever, right upper-quadrant pain, pathological liver test results, and histological signs of cholangitis on biopsy.28 All nonsurgical bile duct strictures were confirmed by percutaneous transhepatic cholangiography or endoscopic retrograde cholangiopancreatography.

Donor T and B lymphocytes were isolated from peripheral blood. Pretransplantation microcytotoxic T- and B-lymphocyte cross-match was performed routinely using CDC assay, as described elsewhere.29 Flow cytometric cross-match using donor spleen lymphocytes and recipient sera also was performed, as described earlier.30

Isolation and Phenotyping of BECs Human BECs were isolated from the liver of a healthy donor by using a method similar to one already described.31 In short, liver tissue was mechanically disrupted and enzymatically digested. The isolated cells were seeded on gelatin-coated tissue-culture flasks until they reached confluence. BECs were purified by immunomagnetic isolation using Dynabeads conjugated with the epithelial cell–specific antibody BerEp4 (Dynal, Oslo, Norway). Approximately 2 ⫻ 107 immunomagnetic beads were washed and incubated with cells trypsinized from flasks. The mixture was incubated at 4°C for 30 to 40 minutes with gentle agitation. Cells coupled to the beads were isolated and washed by applying a magnetic particle concentrator (Dynal). Immunoisolated BECs were cultured as described previously.32 Routine characterization indicated that cultures were more than 95% pure for BECs. Cells were positive for cytokeratin 19 and 7 and a fluorescein isothiocyanate (FITC)-conjugated antibody directed to a common cytokeratin epitope. Cells stained negative for antibodies against von Willebrand factor (endothelial cell marker), ␣-actin (smooth muscle cell marker), and fibroblast. Permission for isolation of BECs was granted from the local ethical committee.

Flow Cytometric Analysis of BEC Antibodies A pool of sera from patients who had formed alloantibodies as a result of multiple blood transfusions or organ transplantations was used as a positive control. Sera from healthy nontransfused blood group AB males served as negative controls. For the detection of BEC antibodies, 5 ⫻ 105 BECs were incubated with 50 ␮L of patient serum for 1 hour at 22°C, then washed three times with phosphate-buffered saline. Ten microliters of 1:4 diluted fluoresceinated (FITC) F(ab⬘)2 fragments of goat anti-human immunoglobulin G (IgG; Fc specific) and IgM antibodies (Jackson Immunoreasearch, West Grove, PA) were added and incubated at 4°C on ice in the dark for 25 minutes. Cells were washed and analyzed on a Becton Dickinson flow cytometer (FACSorter; Becton Dickinson, San Jose, CA). Fluorescence signals from 10,000 cells were counted and the percentage of FITC-positive cells was recorded. A shift in the mean fluorescence of 20 channels in the test sample compared with a negative control was considered positive.33

Biliary Epithelial Cell Antibodies and Liver Transplantation

All sera giving a positive reaction were diluted further (1:5, 1:100, 1:500, and 1:1,000) in phosphate-buffered saline to determine the titer of antibodies. In all experiments, cell samples were stimulated with recombinant tumor necrosis factor-␣ and interferon-␥ (20 and 200 ng/mL respectively; R&D Systems, Abingdon, England), which was added to the culture medium overnight before harvesting of cells for analysis.

Specificity Determination of BEC Antibodies We previously described a new and quick method for the isolation of HLA class I and class II antigens using paramagnetic microbeads.34 Microbeads coated with pooled HLA class I or II antigens (in some cases, donor-specific HLA class I– or class II– coated microbeads) were used for the removal of HLA antibodies from patient sera.34 Absorbed sera were retested for binding to BECs. We also checked whether BEC antibodies were directed against red blood cells. For this purpose, all sera giving positive reactions with BECs were first absorbed with packed red blood cells obtained from 10 different donors. Sera then were retested for binding to BECs. To test for tissue specificity, we used human lung (bronchial) epithelial cells (LECs) as control target cells (Clonetics, BioWhittaker, San Diego, CA).

Statistical Analysis Chi-square test was used to compare categorical parameters. Yates correction and Fisher’s exact test were applied, when appropriate. Total bilirubin and GGT values at different times between BEC antibody–positive and BEC antibody– negative groups were compared by means of repeated analysis of variance. Log-rank test was used to compare patient survival rates between different groups. Comparison of mean values in different groups was performed using Student’s t-test. Differences are considered significant for P less than .05.

Results Patient Results The 2-year patient survival rate in our study was 86.3%. Fifty patients (52.6%) developed acute rejection episodes after liver transplantation. Seventy-six percent (38 of 50 episodes) of rejection episodes were confirmed by biopsy, and seven episodes were steroid-resistant rejection. There was no difference in acute rejection episodes between FK506-based (50.7%) and cyclosporine A– based (58.3%) immunosuppression regimens (P ⫽ not significant [NS]). Twelve patients developed cholangitis and seven patients developed nonsurgical bile duct strictures within 2 years after transplantation.

1193

Donor Lymphocyte-Specific Cross-Matches and Their Correlation With Acute Rejection Routine microcytotoxic T- and B-lymphocyte crossmatch showed that only 5 of 95 cross-match sera samples were positive, and three of these five patients experienced rejection episodes. Four of five sera samples also were positive for flow cytometric cross-match. All pretransplantation sera also were tested for reactivity with their respective donor spleen lymphocytes by means of flow cytometric assay. Ten of 95 sera samples (10.5%) gave positive reactions. Six of 10 of these patients (60%) experienced rejection episodes. However, the presence of donor lymphocyte–specific antibodies detected by flow cytometry did not correlate with either rejection episodes or patient survival (P ⫽ NS). In addition, of the 10 sera samples positive in flow cytometric donor lymphocyte cross-match, 7 sera also reacted with BECs. Presence of Antibodies to BECs in Sera of Liver Transplant Recipients and Its Correlation to Acute Rejection In this study, 41 of 95 pretransplantation sera smaples (43.2%) were positive for BEC antibodies. Anti-BEC reactivity could be detected in all patient groups, but a greater frequency of patients with autoimmune liver diseases (primary sclerosing cholangitis [PSC], primary biliary cirrhosis [PBC], and autoimmune hepatitis [AIH]) had BEC antibodies (20 of 33 patients; 60.6%) than other patients (21 of 62 patients; 33.9%; P ⬍ .02). There was no significant age difference between BEC antibody–positive and BEC antibody–negative patients. However, there were more women in BEC antibody– positive (53.7%) than BEC antibody–negative patients (29.6%; P ⬍ .02; Tables 1 and 2). Significantly higher numbers of patients with BEC antibodies developed acute rejection (65.9%) compared with 42.5% without antibodies (P ⬍ .03). There was no significant difference in number of steroid-resistant rejection episodes, nonsurgical bile duct strictures, and 2-year patient survival rates between patient groups with and without BEC antibodies. Of patients with BEC antibodies, 19.5% developed cholangitis compared with 7.4% without antibodies (P ⫽ NS; Table 2). Both groups of patients showed a similar decreasing pattern of total bilirubin level changes within 2 years after transplantation. Total bilirubin values in patients with BEC antibodies were greater than those in patients without antibodies at all checked times, but the difference was not statistically significant (Fig. 1). GGT levels in BEC antibody–negative patients showed a similar

1194

Ge et al

Table 2. Correlation Between BEC Antibodies and Clinical Results

BEC* Reactivity BEC antibody positive BEC antibody negative P

No. of Patients

Mean Age (yr)

Women

Acute Rejections

Steroid-Resistant Rejections

Cholangitis

Nonsurgical Bile Duct Stricture

Two-Year Survival

41 54

48.9 46.6 NS

22 (53.7) 16 (29.6) ⬍.02

27 (65.9) 23 (42.5) ⬍.03

4 (9.8) 3 (5.6) NS

8 (19.5) 4 (7.4) NS

4 (9.8) 3 (5.6) NS

35 (85.4) 47 (87.8) NS

NOTE. Values expressed as number (percent). *All antibody reactivity was detected using cytokine-stimulated BECs. BECs were stimulated with recombinant human tumor necrosis factor-␣ and interferon-␥ (20 and 200 ng/mL, respectively) overnight.

type of change as bilirubin levels, whereas BEC antibody–positive patients showed a different pattern in which GGT levels increased for the first 3 months and remained constant at this level for 1 year, then decreased and reached a value even lower than that of the antibody-negative group at 2 years posttransplantation. The difference was not statistically significant (Fig. 2). Specificities of BEC Antibodies and Their Correlation With Rejection The majority of immunoglobulins in BEC antibody– positive sera were a mixture of IgG and IgM. A greater BEC antibody titer was detected in patients with rejection episodes (1:500 to 1:1,000) compared with those without rejection episodes (1:5 to 1:100); however, the difference was not statistically significant (data not shown). We tested whether the antibodies were HLA specific. Our absorption assays with HLA class I and class

Figure 1. Total bilirubin level changes within 2 years after transplantation. Total bilirubin values in patients with BEC antibodies were greater than those in patients without antibodies at all times, but the difference was not statistically significant (P ⴝ NS).

II antigen-coated magnetic beads showed that in 5 of 41 sera samples (12.2%), reactivity against BECs was caused by HLA antibodies, whereas in 36 of 41 sera samples (87.8%), no demonstrable HLA antibodies were detected, indicating the presence of non-HLA antibodies. It is important to remember that the HLA reactivity detected in the five patients was not donor specific, but against the HLA type expressed by BECs. We also compared correlations between these antibodies and clinical results. A higher number of patients with HLA antibodies had cholangitis (80.0%) compared with 11.1% with non-HLA antibodies (P ⫽ .03). However, there were no significant differences between patients with HLA antibodies and non-HLA antibodies with regard to acute rejection episodes, steroid-resistant rejection episodes, nonsurgical bile duct stricture, and

Figure 2. GGT level changes within 2 years after transplantation. In BEC antibody–negative patients, GGT levels showed similar type of changes with bilirubin levels. However, BEC antibody–positive patients showed a different pattern in which GGT levels increased for the first 3 months and remained constant at this level for 1 year, then decreased and reached a value even lower than that of the antibody-negative group at 2 years posttransplantation. The difference was not statistically significant (P ⴝ NS).

1195

Biliary Epithelial Cell Antibodies and Liver Transplantation

Table 3. HLA Specificity of BEC Antibodies and Their Correlation With Clinical Results HLA Specificity of BEC Antibodies HLA antibodies Non-HLA antibodies P

No. of Patients

Acute Rejections

Steroid-Resistant Rejections

Cholangitis

Nonsurgical Bile Duct Stricture

Two-Year Survival

5 36

5 (100.0) 22 (61.1) NS

1 (20.0) 3 (8.3) NS

4 (80.0) 4 (11.1) .03

2 (40.0) 2 (5.6) NS

5 (100.0) 30 (83.3) NS

NOTE. Values expressed as number (percent).

2-year patient survival rates (Tables 3 and 4). Comparison of acute rejection episodes in patients with HLA BEC antibodies with that in patients without BEC antibodies showed no significant difference (5 of 5 versus 23 of 54 patients; P ⫽ NS). This was true even in patients with non-HLA BEC antibodies (22 of 36 versus 23 of 54 patients; P ⫽ NS). No decrease in reactivity of sera with BECs was observed after absorption with packed red blood cells. Thus, BEC antibodies were not directed against blood group antigens. Patient sera were tested further for tissue specificity using LECs. We found that a lower number of patients (10 of 95 patients; 10.5%) had antibodies against LECs compared with BECs. The presence of anti-LEC antibodies did not correlate with acute rejection. None of the anti-BEC antibodies cross-reacted with LECs.

Discussion In our group of patients, the 2-year patient survival rate was 86.3%, and the occurrence of acute rejection episodes was 52.6%. Clinical results are similar to those of

other studies.35,36 Only 11.6% of patients in our study had either a positive CDC or flow cytometric donor lymphocyte-specific cross-match. This is consistent with the observation that the significance of the traditional cross-match test in liver transplantation is still uncertain.37 In our study, BEC-reactive antibodies were found in all patient categories, but the presence of these antibodies was more common in patients with autoimmune liver diseases (PSC, PBC, and AIH) and women (Tables 1 and 2). Alloantibodies are formed as a result of wholeblood transfusions, multiple pregnancies, or earlier lost transplants.38-40 Few patients in our study group were alloimmunized; thus, the origin of BEC antibodies is not clear. Most likely these antibodies are autoreactive.33 It is well known that autoimmune liver diseases are featured by autoreactive antibodies.41 Patients with alcoholic cirrhosis showed a high incidence of BEC antibodies (five of seven patients; 71.4%) in our study. It is reported that alcoholic liver diseases are associated with autoimmunity.42,43 Thus, it is likely that patients with alcoholic cirrhosis may have autoantibodies reac-

Table 4. HLA Specificity of Preformed Antibodies Detected Using Different Methods and Target Cells and Its Correlation With Acute Rejection in Sera of Patients Awaiting a Liver Allograft Method/Target Cell Used for Antibody Detection

No. of Patients

Presence of HLA Antibodies

Rejection in Patients With HLA Antibodies

CDC⫹/FCXM⫺/BECs⫺ CDC⫺/FCXM⫹/BECs⫺ CDC⫺/FCXM⫺/BECs⫹ CDC⫹/FCXM⫹/BECs⫺ CDC⫺/FCXM⫹/BECs⫹ CDC⫹/FCXM⫺/BECs⫹ CDC⫹/FCXM⫹/BECs⫹

0 1 33 2 5 1 2

0 0 1* 0 2* 0 2*

0 0 1 0 2 0 2

NOTE. CDC and FCXM are donor-lymphocyte specific. All antibody reactivity was detected using cytokine-stimulated BECs. BECs were stimulated with recombinant human tumor necrosis factor-␣ and interferon-␥ (20 and 200 ng/mL, respectively) overnight. Abbreviation: FCXM, flow cytometric cross-match. *Reactivity against BEC HLA type, but not donor HLA.

1196

Ge et al

tive with BECs. However, because of the small number of patients in this study, a definitive conclusion cannot be drawn. Our study shows that the presence of preformed antibodies to BECs was associated with acute rejection, but not patient survival within 2 years. One possible explanation could be that the occurrence of acute rejection does not necessarily lead to increased mortality.44,45 The exact mechanism of liver allograft rejection is not yet elucidated. In liver allograft rejection, most tissue damage is thought to occur as a consequence of direct cellular immunologic injury to bile duct epithelium. Humoral mechanisms for liver allograft destruction have been reported to be responsible for hyperacute rejection, which seldom occurs in liver transplantation.46,47 It also has been shown that antibodies to blood group antigens may have a deleterious effect on liver allograft outcome.48-50 In a previous study, we showed that BEC autoantibodies on binding induced increased CD44 expression (not expressed on normal bile ducts) on BECs and increased production of high levels of the proinflammatory cytokine interleukin-6 (IL-6) by BECs.33 Lymphocyte activation and infiltration into a transplanted organ is an integral component of the rejection process. Graft infiltration of lymphocytes requires adhesion of leukocytes to the endothelium, diapedesis, and transmigration. One of several proteins involved in this process is CD44, known to interact with endothelial hyaluronan. It is shown that blocking the CD44 and hyaluronic acid interaction prolongs rat allograft survival.51 Induced expression of CD44 by BEC antibodies thus may facilitate other cellular mechanisms leading to the destruction of bile ducts.33 It has been reported that under conditions of inflammation or stress in vitro, BECs produce, secrete, and respond to IL-6 through upregulation and activation of the IL-6 receptor (gp80)/gp130 signaling system in an autocrine/paracrine manner.52 Acute rejection is the recipient’s inflammatory response to the grafted organ. Elevated IL-6 levels in urine and bile have been associated with acute renal and liver allograft rejection.53,54 Thus, BEC antibodies may contribute to acute rejection by these mechanisms. No difference was observed with regard to the development of cholangitis in patients with compared to without BEC antibodies (19.5% versus 7.4%; P ⫽ NS). Even if there is still not enough evidence showing that BEC antibodies are capable of causing graft destruction either directly or in association with host effector cells, such as in an antibody-dependent cell-

mediated cytotoxicity type response, these antibodies might potentiate an indirect pathway of T-cell activation.30 However, the possible functional capacity of BEC antibodies in liver allograft rejection still needs to be confirmed by additional in vivo studies. Specificity determinations of BEC antibodies showed they were not directed against blood group antigens and were HLA specific in only few cases (12%). In the present study, a higher number of patients with HLA antibodies had cholangitis (80.0%) compared with 11.1% with non-HLA antibodies (P ⫽ .03). This finding needs to be confirmed in a larger study. In addition, no clinical correlation of non-HLA antibodies to BECs could be drawn in the present study. This may reflect either that the predominantly non-HLA antibodies observed in the present study are somehow weaker or may be caused by an improvement in the current management of acute liver allograft rejection. To study polymorphisms in the antigenic system, we currently are setting up a panel of BECs isolated from different healthy livers. Our study provides a feasible method to predict the occurrence of acute rejection in liver transplants. Using a panel of BECs isolated from three to four donors, the presence of these antibodies may be determined by a simple screening procedure for all liver transplant candidates already when the patient is accepted on the waiting list. Although a positive reaction with BECs may not hinder or deny the patient a liver transplant, it may help clinicians in deciding an alternative or stronger immunosuppressive protocol. In conclusion, we show that preformed antibodies to BECs detected in patients before liver transplantation are associated with acute rejection. No correlation of the specificity of the antibodies with rejection was found. However, results of the present study warrant additional studies that include a larger number of patients to evaluate the role of anti-BEC non-HLA antibodies in liver transplantation.

Acknowledgment ¨ stman and Maria The authors thank research nurses Aina O Svenaeus Lundgren for their help in collecting some of the clinical data and Dr Henrik Gjertsen for valuable suggestions in the preparation of the manuscript.

References 1. Carpenter CB, Winn HJ. Hyperacute rejection. N Engl J Med 1969;280:47-48.

Biliary Epithelial Cell Antibodies and Liver Transplantation

2. Tilney NL, Kupiec-Weglinski JW. The biology of acute transplant rejection. Ann Surg 1991;214:98-106. 3. Patel R, Terasaki PI. Significance of the positive crossmatch test in kidney transplantation. N Engl J Med 1969;280:735-739. 4. Talbot D. The flow cytometric crossmatch in perspective. Transpl Immunol 1993;1:155-162. 5. Buelow R, Chiang TR, Monteiro F, Cornejo MC, Ellingson L, Class F, et al. Soluble HLA antigens and ELISA—A new technology for crossmatch testing. Transplantation 1995;60:15941599. 6. Thompson JS, Thacker LR II, Krishnan G. Human leukocyte antigens DR and AB and kidney retransplantation. Transplantation 2003;75:718-723. 7. Taylor CJ, Dyer PA. Maximizing the benefits of HLA matching for renal transplantation: Alleles, specificities, cregs, epitopes, or residues. Transplantation 1999;68:1093-1094. 8. Takaya S, Bronsther O, Iwaki Y, Nakamura K, Abu-Elmagd K, Yagihashi A, et al. The adverse impact on liver transplantation of using positive cytotoxic crossmatch donors. Transplantation 1992;53:400-406. 9. Nikaein A, Backman L, Jennings L, Levy MF, Goldstein R, Gonwa T, et al. HLA compatibility and liver transplant outcome. Improved patient survival by HLA and cross-matching. Transplantation 1994;58:786-792. 10. Ogura K, Terasaki PI, Koyama H, Chia J, Imagawa DK, Busuttil RW. High one-month liver graft failure rates in flow cytometry crossmatch-positive recipients. Clin Transplant 1994;8:111115. 11. Talbot D, Bell A, Shenton BK, Hussein KA, Manas D, Gibbs P, Thick M. The flow cytometric crossmatch in liver transplantation. Transplantation 1995;59:737-740. 12. Dunn MJ, Rose ML, Latif N, Bradd S, Lovegrove C, Seymour C, et al. Demonstration by Western blotting of antiheart antibodies before and after cardiac transplantation. Transplantation 1991; 51:806-812. 13. Brasile L, Zerbe T, Rabin B, Clarke J, Abrams A, Cerilli J. Identification of the antibody to vascular endothelial cells in patients undergoing cardiac transplantation. Transplantation 1985;40:672-675. 14. Martin S, Brenchley PE, Postlethwaite RJ, Johnson RW, Dyer PA. Detection of anti-epithelial cell antibodies in association with pediatric renal transplant failure using a novel microcytotoxicity assay. Tissue Antigens 1991;37:152-155. 15. Schuurman HJ, Jambroes G, Borleffs JC, Slootweg PJ, Meyling FH, deGast GC. Acute humoral rejection after heart transplantation. Transplantation 1988;46:603-605. 16. Karuppan S, Ericzon BG, Moller E. Relevance of a positive crossmatch in liver transplantation. Transpl Int 1991;4:18-25. 17. Charco R, Vargas V, Balsells J, Lazaro JL, Murio E, Jaurrieta E, et al. Influence of anti-HLA antibodies and positive T-lymphocytotoxic crossmatch on survival and graft rejection in human liver transplantation. J Hepatol 1996;24:452-459. 18. Lobo PI, Spencer C, Douglas MT, Stevenson WC, Pruett TL. The lack of long-term detrimental effects on liver allografts caused by donor-specific anti-HLA antibodies. Transplantation 1993;55:1063-1066. 19. Fujita S, Rosen C, Reed A, Langham MR Jr, Howard RJ, et al. Significance of preformed anti-donor antibodies in liver transplantation. Transplantation 1997;63:84-88. 20. Donaldson PT, Thomson LJ, Heads A, Underhill JA, Vaughan RW, Rolando N, Williams R. IgG donor-specific crossmatches are not associated with graft rejection or poor graft survival after

21.

22.

23. 24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

1197

liver transplantation. An assessment by cytotoxicity and flow cytometry. Transplantation 1995;60:1016-1023. Smith JD, Crisp SJ, Dunn MJ, Pomerance A, Yacoub MH, Rose ML. Pre-transplant anti-epithelial cell antibodies and graft failure after single lung transplantation. Transpl Immunol 1995;3: 68-73. Opelz G, Wujciak T, Dohler B, Scherer S, Mytilineos J. HLA compatibility and organ transplant survival. Collaborative Transplant Study. Rev Immunogenet 1999;1:334-342. Testa G, Malago M, Broelseh CE. Complications of biliary tract in liver transplantation. World J Surg 2001;25:1296-1299. Inomata Y, Tanaka K. Pathogenesis and treatment of bile duct loss after liver transplantation. J Hepatobiliary Pancreat Surg 2001;8:316-322. Vierling JM, Fennell RH Jr. Histopathology of early and late human hepatic allograft rejection: Evidence of progressive destruction of interlobular bile ducts. Hepatology 1985;5:10761082. Donaldson PT, Alexander GJ, O’Grady J, Neuberger J, Portmann B, Thick M, et al. Evidence for an immune response to HLA class I antigens in the vanishing-bile duct syndrome after liver transplantation. Lancet 1987;1:945-951. Malmborg AC, Shultz DB, Luton F, Mostkov KE, Richly E, Leung PS, et al. Penetration and co-localization in MDCK cell mitochondria of IgA derived from patients with primary biliary cirrhosis. J Autoimmun 1998;11:573-580. Barkholt L, Ericzon BG, Tollemar J, Malmborg AS, Ehrnst A, Wilezek H, Andersson J. Infections in human liver recipients: Different patterns early and late after transplantation. Transpl Int 1993;6:77-84. Karuppan SLA, Moller E. Pre- and post-transplant sensitization in cyclosporin A-treated patients. Clin Transplant 1988;2:245251. Karuppan SS, Lindholm A, Moller E. Fewer acute rejection episodes and improved outcome in kidney-transplanted patients with selection criteria based on crossmatching. Transplantation 1992;53:666-673. Joplin R, Strain AJ, Neuberger JM. Immuno-isolation and culture of biliary epithelial cells from normal human liver. In Vitro Cell Dev Biol 1989;25:1189-1192. Bo X, Broome U, Remberger M, Sumitran-Holgersson S. Tumour necrosis factor alpha impairs function of liver derived T lymphocytes and natural killer cells in patients with primary sclerosing cholangitis. Gut 2001;49:131-141. Xu B, Broome U, Ericzon BG, Sumitran-Holgersson S. High frequency of autoantibodies in patients with primary sclerosing cholangitis that bind biliary epithelial cells and induce expression of CD44 and production of interleukin 6. Gut 2002;51:120127. Sumitran-Karuppan S, Moller E. The use of magnetic beads coated with soluble HLA class I or class II proteins in antibody screening and for specificity determinations of donor-reactive antibodies. Transplantation 1996;61:1539-1545. The US Multicenter FK506 Liver Study Group. A comparison of tacrolimus (FK506) and cyclosporine for immunosuppression in liver transplantation. N Engl J Med 1994;331:1110-1115. Williams R, Neuhaus P, Bismuth H, McMaster P, Pichlmayr R, Calne R, et al. Two-year data from the European multicentre tacrolimus (FK506) liver study. Transpl Int 1996;9(suppl): S144-S150. Donaldson PT, Williams R. Cross-matching in liver transplantation. Transplantation 1997;63:789-794.

1198

Ge et al

38. Dausset J. Leuco-agglutinins and blood transfusion. Vox Scag 1954:190-198. 39. Payne R, Rolfs MR. Fetomaternal leukocyte incompatibility. J Clin Invest 1958;37:1756-1763. 40. Walford RL, Carter PK, Anderson RE. Leukocyte antibodies following skin homografting in the human. Transplant Bull 1962;29:16-18. 41. Krawitt EL, Wiesner RH, Nishioka M. Autoimmune liver diseases (ed 2). Amsterdam: Elsevier, 1998:21-35. 42. Vidali M, Stewart SF, Rolla R, Daly AK, Chen Y, Mottaran E, et al. Genetic and epigenetic factors in autoimmune reactions toward cytochrome P4502E1 in alcoholic liver disease. Hepatology 2003;37:410-419. 43. Zetterman RK. Autoimmunity and alcoholic liver disease. Am J Med 1990;89:127-128. 44. Klintmalm GB, Nery JR, Husberg BS, Gonwa TA, Tillery GW. Rejection in liver transplantation. Hepatology 1989;10: 978-985. 45. Seiler CA, Renner EL, Czerniak A, Didonna D, Buchler MW, Reichen J. Early acute cellular rejection: No effect on late hepatic allograft function in man. Transpl Int 1999;12:195-201. 46. Imagawa DK, Noguchi K, Iwaki Y, Busuttil RW. Hyperacute rejection following ABO-compatible orthotopic liver transplantation—A case report. Transplantation 1992;54:1114-1117. 47. Bird G, Friend P, Donaldson P, O’Grady J, Portmann B, Calne

48.

49.

50.

51.

52.

53. 54.

R, Williams R. Hyperacute rejection in liver transplantation: A case report. Transplant Proc 1989;21:3742-3744. Rego J, Prevost F, Rumeau JL, Modesto A, Fourtainier G, Durand D, et al. Hyperacute rejection after ABO-incompatible orthotopic liver transplantation. Transplant Proc 1987;19:45894590. Demetris AJ, Jaffe R, Tzakis A, Ramsey G, Todo S, Belle S, et al. Antibody-mediated rejection of human orthotopic liver allografts. A study of liver transplantation across ABO blood group barriers. Am J Pathol 1988;132:489-502. Angstadt J, Jarrell B, Maddrey W, Munoz S, Yang SL, Moritz M, Carabasi A. Hemolysis in ABO-incompatible liver transplantation. Transplant Proc 1987;19:4595-4597. Zhang W, Gao L, Qi S, Liu D, Xu D, Peng J, et al. Blocking of CD44-hyaluronic acid interaction prolongs rat allograft survival. Transplantation 2000;69:665-667. Yokomuro S, Lunz JG III, Sakamoto T, Ezure T, Murase N, Demetris AJ. The effect of interleukin-6 (IL-6)/gp130 signalling on biliary epithelial cell growth, in vitro. Cytokine 2000;12:727730. Kaden J, Priesterjahn R. Increasing urinary IL-6 levels announce kidney graft rejection. Transpl Int 2000;13(suppl):S34-S41. Umeshita K, Monden M, Tono T, Hasuike Y, Kanai T, Gotoh M, et al. Determination of the presence of interleukin-6 in bile after orthotopic liver transplantation. Its role in the diagnosis of acute rejection. Ann Surg 1996;223:204-211.