The clinical and histological features of chronic liver

Transcription

1 ORIGINAL ARTICLES Impaired Regeneration of Biliary Cells in Human Chronic Liver Allograft Rejection. Special Emphasis on the Role of the Finest Branches of the Biliary Tree Marius C. van den Heuvel, 1 Koert P. de Jong, 2 Marian L.C. van der Horst, 1 Sibrand Poppema, 1 Maarten J.H. Slooff, 2 and Annette S.H. Gouw 1 Severe loss of bile ducts is a hallmark of chronic liver rejection. We hypothesize that loss of the finest branches of the biliary tree, including the intralobular segments, contributes to an impaired regenerative response of bile ducts in chronic rejection. The number and proliferative response of bile ducts, intraportal ductules, and extraportal biliary cells were studied in graft biopsies of 12 chronic-rejection patients. Twenty-two long-term-surviving patients who experienced acute rejection without chronic rejection served as controls. Reperfusion, 1-week, 1-month, and 1-year biopsies were studied. Monochlonal antibody Ki67 was applied to assess proliferation, and cytokeratin 7 antibody to identify bile ducts, intraportal ductules, and extraportal biliary cells. Progressive loss of bile ducts, intraportal ductules, and extraportal biliary cells was observed in the chronic-rejection group. In controls, all of these structures remained well preserved. Additionally, a significant increase of intraportal ductules was present at 1 week in controls, which was not the case in the chronic-rejection group. Proliferative activity of intraportal ductules and extraportal biliary cells was significantly increased at 1 week in controls. This proliferative activity was higher in the intraportal ductules of controls, compared with the chronic-rejection group. After 1 week, proliferative activity was virtually absent in both groups. In conclusion, our results showed that a deficient proliferative response of the finer branches of the biliary tree, including its intralobular segments, might contribute to bile-duct loss in chronic rejection. This finding supports the postulation that these structures represent a regenerative compartment of the biliary unit in the liver. (Liver Transpl 2004;10:28 35.) The clinical and histological features of chronic liver allograft rejection are well known and established. 1,2 Chronic rejection occurs in three settings: as Abbreviations: CK7: cytokeratin 7 antibody; MHC, major histocompatibility complex. From the Department of 1 Pathology & Laboratory Medicine, and the 2 Department of Surgery, division of Hepatopancreatobiliary Surgery and Liver Transplantation, University Hospital Groningen, the Netherlands. Address reprint requests to Annette S.H. Gouw, M.D., Ph.D., Dept. of Pathology & Laboratory Medicine, University Hospital Groningen, P.O. Box , 9700 RB Groningen, The Netherlands. Telephone: ; FAX: ; a.s.h.gouw@path.azg.nl Copyright 2004 by the American Association for the Study of Liver Diseases Published online in Wiley InterScience ( DOI /lt the end-stage of unresolved acute cellular rejection; after multiple episodes of acute rejection; or without any preceding clinically recognized episodes of acute rejection. The two major histological characteristics of chronic rejection are obliterative arteriopathy of the medium-sized hepatic arteries, and ductopenia, or severe loss, of the interlobular bile ducts. The histological diagnosis of chronic rejection on graft biopsies relies mainly on ductopenia, because obliterative arteriopathy is usually not present in the smaller hepatic artery tributaries that are present in needle-graft biopsies. Although the early diagnosis of ductopenia is difficult, it has become more feasible by applying the criteria of bile-duct damage as recommended by an international panel. 1 Nevertheless, the pathogenesis of ductopenia is still not clarified. Ductopenia might develop following immunological attacks on bile-duct epithelial cells during acute rejection or ischemic damage due to obliterative arteriopathy. In a striking contrast to chronic biliary diseases like primary biliary cirrhosis and primary sclerosing cholangitis, ductopenia in chronic rejection is usually not associated with the profound increase in numbers of ductular structures, which is a common finding in latestage primary biliary cirrhosis and primary sclerosing cholangitis. Several studies of various types of diseased human livers showed that this ductular reaction represents proliferation of putative hepatic progenitor cells The present study was performed to investigate the relation between ductopenia in chronic rejection and the regenerative potential of the biliary tree. We hypothesize that during the development of chronic rejection, loss of the finest branches of the biliary tree, including its intralobular segments, which contain the putative hepatic progenitor cells, contributes to a decreased regenerative capacity of bile ducts, which will ultimately lead to ductopenia. To test this hypothesis we studied the proliferative activity of bile ducts and the finer branches of the biliary tree, including the intralobular segments, during the development of chronic rejection, using graft biopsies of 12 chronicrejection patients taken prior to retransplantation for end-stage chronic rejection. The finer branches of the 28 Liver Transplantation, Vol 10, No 1 ( January), 2004: pp 28 35

2 van den Heuvel et al. 29 Table 1. Indication For Liver Transplantation Primary Disease biliary tree are defined according to the recommendations mentioned in the consensus paper on the nomenclature of these structures by Roskams T and Theise ND et al. (in press). Materials and Methods Patients and Biopsies In this study we included 12 chronic-rejection patients who had developed chronic rejection and in whom a retransplantation was ultimately performed. The final diagnosis of chronic rejection was made on the excised liver transplants. The control group consisted of 22 patients who were selected on the basis of biopsy-proven acute rejection at 1 week posttransplantation, the availability of a 1-month biopsy, and a minimum follow-up of at least 1 year and a biopsy at that time with normal histology. The histological diagnosis of acute rejection, chronic rejection, and ductopenia was established according to consensus criteria 11,1. The biopsies studied in both groups consisted of biopsies taken at reperfusion, at 1 week, at 1 month and 1 year posttransplantation. In the chronic rejection group, one of the 1-week biopsies was actually taken at day 10 posttransplantation. The last biopsy taken before retransplantation was considered as the 1-year biopsy (median, 160 days; range, days) of the chronic rejection group. We have chosen not to take samples of the transplantectomy specimens to avoid comparison of larger tissue samples (from the chronicrejection group) with needle biopsies (from the control group). Needle biopsies were performed with a 1.6 mm Hepafix needle (Braun, Melsungen, Germany). Clinical Data CR (n 12) Control (n 22) Primary sclerosing cholangitis 2 (17%) 4 (18%) Primary biliary cirrhosis 1 (8%) 2 (9%) Extrahepatic biliary atresia 1 (8%) 3 (14%) Cirrhosis due to tyrosinemia 1 (8%) Wilson s disease 1 (8%) 3 (14%) Autoimmune hepatitis 1 (8%) 2 (9%) Alcoholic cirrhosis 1 (5%) Acute liver failure 2 (17%) 1 (5%) Polycystic disease 1 (5%) Cryptogenic cirrhosis 2 (17%) 3 (14%) Viral hepatitis 1 (8%) 2 (9%) Abbreviation: CR, chronic-rejection group. Indications for liver transplantation are listed in Table 1. A Roux-en-Y biliary anastomosis was performed in four chronicrejection patients (33%) and in seven patients (32%) of the control group. Six chronic rejection patients (50%) and nine (41%) patients of the control group developed a cytomegalovirus infection. Ten of the 12 chronic-rejection patients developed chronic rejection following unresponsiveness to antirejection therapy. In the remaining two patients, immunosuppressive therapy was temporarily decreased because of side effects. Just prior to retransplantation all chronic-rejection patients had high bilirubin levels varying between 12.2 mg/dl and 56.5 /dl. The basic immunosuppression in the chronic-rejection group consisted of triple-drug therapy, with prednisolone, cyclosporine and azathioprine. Conversion to FK506 (tacrolimus) was performed in five patients. In the control group, 16 patients were also given the same triple-drug therapy. Conversion to FK506 was performed in one patient. In six patients, immunosuppression consisted of FK506 and prednisolone. In both groups histologically proven episodes of acute rejection were treated with boluses of methylprednisolone. Patients with steroid-resistant rejection were given antithymocyte globulin. In patients treated with cyclosporine, we aimed for trough levels of g/l. After one month, cyclosporine was tapered to levels of g/l. Furthermore, we aimed for FK506 levels of g/l in the first month after transplantation and levels of 10 g/ l thereafter. Nomenclature and Scoring Method Apart from the interlobular bile ducts, the smaller tributaries of the biliary tree that were considered in the present study were defined according to the consensus paper on the nomenclature of the finer branches of the biliary tree (Roskams T, Theise ND, et al., in press). Biliary channels, lined by CK7 (anticytokeratin 7; DAKO, Glostrup, Denmark)-positive cholangiocytes and located at the portal-parenchymal junction, were addressed as intraportal ductules. Isolated CK7- positive cells located within the portal tracts were regarded as parts of these ductules and counted as such. CK7-positive cholangiocytes located within the lobule, either forming ductules or arranged as strings of cells or present as isolated cells were included as extraportal biliary cells. These cells thus represent a heterogeneous group of cells, because according to the aforementioned consensus paper, these cells could represent parts of the intralobular segments of the ductules or the cholangiocytic lining of canals of Hering. CK7-positive cells with an intermediate morphology between hepatocytes and cholangiocytes were not taken into account. In each specimen, all portal tracts present in the biopsy were included. The total numbers of bile ducts, intraportal ductules, and extraportal biliary cells were counted in each portal tract and in the liver parenchyma surrounding that particular portal tract. Subsequently, the ratio of the numbers of these structures and the numbers of portal tracts was calculated for each biopsy. To study the proliferative activity in bile ducts, intraportal ductules, and extraportal biliary cells, the number of Ki67 (MIB; DAKO, Glostrup, Denmark)- positive cells was counted as a ratio of the total number of cells in each structure. However, because the isolated cholangio-

3 30 Impaired Regeneration of Biliary Cells cytes cannot be discriminated reliably from lymphocytes on a morphological basis, these cells, both intraportal and intralobular, were not counted in the immunostaining with the proliferation marker Ki67. Immunohistology To identify bile ducts, intraportal ductules, and extraportal biliary cells, we performed immunohistology on 4 m serially cut paraffin sections using the monoclonal antibody CK7 (anticytokeratin 7; DAKO, Glostrup, Denmark) in a dilution of 1/100. Staining with CK7 was performed using automated immunoperoxidase staining (Ventana System, Illkirch, France). Sections were deparaffinized, then incubated with protease for 12 minutes and subsequently with CK7 for 30 minutes at room temperature. Biotinylated Ig was added for 8 minutes, followed by peroxidase labeled avidin for 8 minutes. The staining reaction was developed using diaminobenzidin, and sections were counterstained with hematoxylin. The proliferative activity was studied with the monoclonal antibody Ki67 (MIB; DAKO, Glostrup, Denmark) in a dilution of 1/400. After deparaffinization, sections were heated at 100 Celsius for 10 minutes. Subsequently, sections were incubated with Ki67 for 30 minutes at room temperature and peroxidase labeled rabbit antimouse antibody (dilution, 1/100) for 30 minutes, followed by incubation with peroxidase labeled goat anti rabbit antibody (dilution 1/100) for 30 minutes. The reaction was developed using diaminobenzidin, and sections were counterstained with hematoxylin. Statistics Longitudinal comparisons within groups were tested using nonparametric tests for related samples (Friedman test). Comparisons between the groups were analyzed by the Mann-Whitney test. A P value of less than 0.05 was considered statistically significant. Statistical analysis was performed using Statistical Package for Social Sciences (SPSS) 11.0 (SPSS Inc., Chicago, IL). Results Routine Histology In the chronic-rejection group, reperfusion biopsies showed normal histological features apart from mild surgical hepatitis. At 1 week, five patients suffered mild acute rejection (Banff grade I), two patients had moderate acute rejection (Banff grade II), and three patients had severe acute rejection (Banff grade III). One patient had parenchymal cholestasis, and one patient showed normal liver histology. At 1 month, three patients were diagnosed with mild acute rejection (Banff grade I), one patient had severe acute rejection (Banff grade III), three patients showed ductopenia, three patients showed parenchymal cholestasis, and two patients had reactive changes. At 1 year (the last biopsies before retransplantation), 10 patients were diagnosed with ductopenia (loss of bile ducts in more than 50% of the portal tracts), compatible with chronic rejection, and two patients showed parenchymal cholestasis, one in combination with mild acute rejection (Banff I). These two patients showed damaged bile ducts in less than 50% of the portal tracts. These features were tentatively compatible with chronic rejection but not firmly diagnostic. However, in the explants obtained during retransplantation, all characteristics of chronic rejection were present. All biopsies with parenchymal cholestasis showed damaged bile ducts but no signs of biliary tract complications. Of the control group, all reperfusion biopsies showed normal liver histology, except for a slight surgical hepatitis in a few biopsies. At 1 week, 11 patients showed mild acute rejection (Banff grade I), six patients had moderate acute rejection (Banff grade II), and five patients showed severe acute rejection (Banff grade III). The 1-month biopsies consisted of one case with mild acute rejection (Banff grade I), one case with moderate acute rejection (Banff grade II), one case with severe acute rejection (Banff grade III), and two cases with ongoing rejection. The other biopsies showed mild reactive changes. All 1-year biopsies in the control patients showed normal histological features. Immunohistology All counted structures showed a similar staining pattern for CK7. The median number of portal tracts counted per biopsy was 9 (range, 4 32). Terminal bile ducts (Fig. 1). The number of bile ducts in the reperfusion biopsies of both groups was similar, showing a median of 1 bile duct per portal tract. The number of bile ducts remained constant in the controls in the course of time. In the last biopsies before retransplantation, the chronic-rejection group showed a median value of 0.2 bile duct per portal tract, representing a significant loss, as compared with reperfusion biopsies (P 0.007) and a significant difference at this stage with the control group (P 0.002). A peak in proliferative activity of bile ducts was observed at 1 week in both groups without a significant difference between the groups. Intraportal ductules (Fig. 2). In the reperfusion biopsies, the median number of intraportal ductules per portal tract was 3.8 in the chronic rejection group and 5 in the control group, which was not significantly different. At 1 week, the number of intraportal ductules in the control group was increased, as compared with the reperfusion biopsy value (P 0.017), but also in com-

4 van den Heuvel et al. 31 Figure 1. (A) Bar graph showing median numbers and standard deviation of numbers of bile ducts per portal tract as counted in the biopsies taken at reperfusion, 1 week, 1 month, and 1 year in the chronic-rejection group and the control group. (B) Bar graph showing median numbers and standard deviation of the ratio of Ki67 positive bile duct cells as counted in the biopsies taken at reperfusion, 1 week, 1 month, and 1 year in the chronicrejection group and the control group. RPB, reperfusion; 1-W, 1 week; 1-M, 1 month; 1-Y, 1 year; CR, chronicrejection group; Ki67, Ki67 antibody. Figure 2. (A) Bar graph showing median numbers and standard deviation of intraportal ductules per portal tract as counted in the biopsies taken at reperfusion, 1 week, 1 month, and 1 year in the chronic-rejection group and the control group. (B) Bar graph showing median numbers and standard deviation of the ratio of Ki67 positive cells of intraportal ductules counted in the biopsies taken at reperfusion (RPB), 1 week (1-W), 1 month (1-M), and 1 year (1-Y) in the chronic-rejection group (CR) and in the control group. RPB, reperfusion; 1-W, 1 week; 1-M, 1 month; 1-Y, 1 year; CR, chronic-rejection group; Ki67, Ki67 antibody. parison with the 1-week value of the chronic-rejection group (P 0.003). In the controls, the number of intraportal ductules decreased to reperfusion biopsy value in the 1-year biopsies, whereas the chronic-rejection group showed a median of 1.2 intraportal ductules per portal tract in the last biopsies before retransplantation, which is lower than the reperfusion biopsy value

5 32 Impaired Regeneration of Biliary Cells (P 0.010). Hardly any proliferative activity was observed in the 1-month and 1-year reperfusion biopsies of both groups. In the control group, a very striking peak of proliferation was observed at 1 week, which was higher than the proliferative activity at 1 month and 1 year (both P 0.001). In the chronic-rejection group, only a slight increase in proliferation activity in intraportal ductules was observed at 1 week, which was significantly lower than for the controls (P 0.01). Extraportal biliary cells (Fig. 3). Numbers of extraportal biliary cells did not change significantly in the course of time in the control group. In the chronic rejection group, however, the number of extraportal biliary cells per portal tract area demonstrated a significant decrease as early as 1 week and virtually absence of extraportal biliary cells thereafter. At 1 week, lower numbers of extraportal biliary cells were observed as compared with the control group (P 0.015). In the control group, similar with the intraportal ductules, extraportal biliary cells also showed a significant increase in proliferative activity at 1 week (P 0.003), as compared with reperfusion biopsy. This increase was absent in the chronicrejection group, which showed hardly any proliferative activity in extraportal biliary cells. In Figures 4 and 5, the morphological features of the changes observed at 1 week and 1 year are depicted. Discussion Figure 3. (A) Bar graph showing median numbers and standard deviation of extraportal biliary cells per portal tract as counted in the biopsies taken at reperfusion, 1 week, 1 month, and 1 year in the chronic-rejection group and the control group. (B) Bar graph showing median numbers and standard deviation of the ratio of Ki67 positive cells of extraportal BC as counted in the biopsies taken at reperfusion, 1 week, 1 month, and 1 year in the chronic-rejection group and the control group. BC, biliary cells; RPB, reperfusion; 1-W, 1 week; 1-M, 1 month; 1-Y, 1 year; CR, chronic-rejection group; Ki67, Ki67 antibody. The present study was designed to investigate the relation between ductopenia in chronic rejection and the regenerative potential of the biliary tree and in particular its finest branches, including the intralobular segments. In the graft biopsies, acute-rejection episodes were present in the chronic-rejection group and by definition in the controls. Bile ducts are known to be prominent targets of immune attack during acute rejection, and as such, bile-duct damage represents one of the major features in the Banff grading system of acute rejection. 11 Our results showed that despite bile-duct damage during these acute rejection episodes, ductopenia is a late event during the development of chronic rejection because a significant decrease in bile-duct numbers was observed only beyond 1 month after transplantation. Starting with similar numbers of bile ducts in the reperfusion biopsies, loss of bile ducts in chronic rejection and the contrasting preservation of bile ducts in the controls as shown by the results of the present study suggest that following bile-duct damage in acute rejection, a regenerative process must have taken place in the control group. This regenerative process was apparently defective in those cases progressing to chronic rejection. Moreover, in the chronic-rejection group, the numbers of intraportal ductules and extraportal biliary cells

6 van den Heuvel et al. 33 Figure 4. Upper panel: Portal tract of a chronic rejection patient at 1 week showing mild acute rejection, patent bile duct (black arrows; š), intraportal ductules, and extraportal biliary cells (white arrows; ) and the proliferative activity in these structures. Lower panel: Portal tract of a control patient at 1 week showing mild acute rejection, patent bile duct (black arrows; š), intraportal ductules, and extraportal biliary cells (white arrows; ) and the proliferative activity in these structures. HE, hematoxylin-eosin; CK7, cytokeratin 7 antibody; Ki67, Ki67 antibody. Figure 5. Upper panel: Portal tract of a chronic rejection patient in the last biopsy taken prior to retransplantation, showing remnants of bile ducts (black arrows; š) and absence of intraportal ductules and extraportal biliary cells. Lower panel: Portal tract of a control patient at 1 year showing normal histological features with patent bile ducts (black arrows; š), intraportal ductules and extraportal biliary cells (white arrows; ). HE, hematoxylin-eosin; CK7, cytokeratin 7 antibody.

7 34 Impaired Regeneration of Biliary Cells showed a progressive decrease in time, reaching values significantly below the values of the reperfusion biopsies. Contrastingly, in the controls, both the intraportal ductules and the extraportal biliary cells remained well preserved. Apart from the significant differences in numbers, there was also a significant difference of proliferative activity of intraportal ductules and extraportal biliary cells between the chronic-rejection group and the controls at 1 week. Strikingly, despite an increase of proliferative activity of bile ducts in the chronic-rejection group at 1 week, as compared with reperfusion biopsies, which is also the case in the control group, the chronic-rejection group showed a loss of bile ducts. At this stage, the numbers and proliferative activity of the intraportal ductules and extraportal biliary cells of the controls were higher than those of the chronic-rejection group. These findings suggest that intraportal ductules and extraportal biliary cells and the proliferative responses of these structures are indispensable for maintaining the integrity of the terminal bile ducts after bile-duct damage during acute rejection. Therefore, the intraportal ductules and extraportal biliary cells seem to represent a regenerative compartment for the bile ducts. These results are much in line with the findings of others in various liver diseases in which the intraportal ductules and extraportal biliary cells were identified as an intrahepatic progenitor cell compartment. 3,5 14 There is also evidence that bile ducts contain cells with progenitor-cell features. 12 Although ductopenia was apparent at a relatively late stage of chronic rejection, our results showed that the regenerative response was at its zenith at 1 week, taking place simultaneously with the acute-rejection episode. At 1 month and later, the proliferative response of bile ducts, intraportal ductules, and extraportal biliary cells of both groups has practically disappeared. Therefore, an early and vigorous regenerative response of the whole biliary system seemed to be a prerequisite for maintaining bile-duct integrity. In chronic rejection, we have shown that not only bile ducts but also intraportal ductules and extraportal biliary cells were lost. These findings imply that the acute-rejection episode of the chronic-rejection group, unlike that of the controls, included concomitant damage to intraportal ductules and extraportal biliary cells. It has long been established that bile-duct epithelial cells are highly immunogenic targets to alloreactive lymphocytes, partly due to the expression of class II MHC antigens. Canals of Hering have been reported to express HLA-DR. 15 Therefore canals of Hering in the liver allograft, which are included in the present study in the extraportal biliary cells, could also be the targets of alloresponse. We and others have demonstrated the importance of several parenchymal features, such as centrizonal hepatocyte dropout, central venulitis, perivenular fibrosis, and lobular inflammation during early acute rejection, as predictive factors for the development of chronic rejection. 16,17 This lobular inflammation could thus be targeted against class II MHC antigens positive extraportal biliary cells in a similar way, as the portal inflammation is targeted against bile ducts and most probably also against intraportal ductules. Centrizonal hepatocyte dropout and hemorrhage might cause disruption of the normal architecture of the liver cell plates and lead to damage of extraportal biliary cells, which are located along the liver cords. The vulnerability of extraportal biliary cells might also partly result from lower numbers of extraportal biliary cells than intraportal ductules in the reperfusion biopsy. The role of extraportal biliary cells, which include the canals of Hering, is of special interest since during repair (e.g., following acute rejection), bone-marrowderived hepatic progenitor cells are recruited into the liver, and they might enter the liver through the canals of Hering. 18 Loss of extraportal biliary cells could lead to a loss of a compartment for hepatic progenitor cells, resulting in the usual absence of a ductular reaction in chronic rejection. The contrast with primary biliary cirrhosis and primary sclerosing cholangitis, in which the ductular reaction is a regular feature, might be explained by the vigor of the immune attack in a relatively vulnerable organ still recovering from several major injuries (e.g., agonal donor phase, organ harvesting, procurement, transplantation, and reperfusion). This might contribute to the rapid and progressive loss of intraportal ductules and extraportal biliary cells during the early stage of chronic rejection. The failure of the chronic-rejection cases to establish an adequate repair process in contrast with the controls may also be explained by the occurrence of several acute-rejection episodes in the chronic-rejection cases. Another possible factor in the disturbance of the regenerative potential is the denervation of the liver transplant. A study has shown that bile ducts and canals of Hering are under trophic influence of the vagal innervation. 19 In summary, the results of the present study indicate that during early acute rejection, a prominent proliferative response is present in the bile ducts, intraportal ductules, and extraportal biliary cells of patients who do not develop chronic rejection. In contrast, in patients with chronic rejection, loss of terminal bile ducts is consistently observed, despite a proliferative response of bile ducts. However, there was significant loss of intraportal ductules and extraportal biliary cells and a con-

8 van den Heuvel et al. 35 comitant low proliferative response of these structures. We conclude that the patency of the finest branches of the biliary tree, including its intralobular segments, could play an important role in maintaining the integrity of the terminal bile ducts. These findings support the concept that these structures represent a regenerative compartment for the biliary unit within the liver. REFERENCES 1. Demetris A, Adams D, Bellamy C, Blakolmer K, Clouston A, Dhillon AP, et al. Update of the international Banff schema for liver allograft rejection: working recommendations for the histopathologic staging and reporting of chronic rejection. An international panel. Hepatology 2000;31: Wiesner RH, Batts KP, Krom RA. Evolving concepts in the diagnosis, pathogenesis, and treatment of chronic hepatic allograft rejection. Liver Transpl Surg 1999;5: Theise ND, Saxena R, Portmann BC, Thung SN, Yee H, Chiriboga L, et al. The canals of Hering and hepatic stem cells in humans. Hepatology 1999;30: van den Heuvel MC, Slooff MJH, Visser L, Muller M, De Jong KP, Poppema S, Gouw ASH. Expression of anti-ov6 antibody and anti-ncam antibody along the biliary line of normal and diseased human livers. Hepatology 2001;33: Crosby HA, Hubscher S, Fabris L, Joplin R, Sell S, Kelly DA, Strain A. Immunolocalization of putative human liver progenitor cells in livers from patients with end stage primary biliary cirrhosis and sclerosing cholangitis using the monoclonal antibody OV-6. Am J Pathol 1998;152: Demetris AJ, Seaberg EC, Wennerberg A, Ionellie J, Michalopoullos G. Ductular reaction after submassive necrosis in humans. Special emphasis on analysis of ductular hepatocytes. Am J Pathol 1996;149: Roskams T, De Vos R, Desmet V. Undifferentiated progenitor cells in focal nodular hyperplasia of the liver. Histopathology 1996;28: Roskams T, De Vos R, van den Oord JJ, Desmet V. Cells with neuroendocrine features in regenerating human livers. APMIS 1991;23(suppl): Roskams T, De Vos R, Van Eyken P, Miyazaki H, Van Damme B, Desmet V. Hepatic OV-6 expression in human liver disease and rat experiments: Evidence for hepatic progenitor cells in man. J Hepatol 1998;29: Lowes K, Brennan BA, Yeoh GC, Olynyx JK. Oval cell numbers in human chronic liver diseases are directly related to disease severity. Am J Pathol 1999;154: Demetris AJ, Batts KP, Dhillon AP, Ferrel L, Fung J, Geller SA, Hart J, et al. Banff schema for grading liver allograft rejection: An international consensus document. Hepatology 1997;25: Baumann U, Crosby HA, Ramani P, Kelly DA, Strain AJ. Expression of the stem cell factor receptor c-kit in normal and diseased pediatric liver: identification of a human hepatic progenitor cell? Hepatology 1999; Alison MR, Poulson R, Forbes SJ. Update on hepatic stem cells. Liver 2001;21: Sell S. Comparison of liver progenitor cells in human atypical ductular reactions with those seen in experimental models of liver injury. Hepatology 1998;27: Saxena R, Hytiroglou P, Thung SN, Theise ND. Destruction of canals of Hering in primary biliary cirrhosis. Hum Pathol 2002; 33: Gouw ASH, van den Heuvel MC, van den Berg AP, Slooff MJH, de Jong KP, Poppema S. The significance of parenchymal changes of acute cellular rejection in predicting chronic liver graft rejection. Transplantation 2002;73: Neil DA, Hubscher SG. Histological and biochemical changes during the evolution of chronic rejection of liver allografts. Hepatology 2002;35: Theise ND, Nimmakayalu M, Gardner R, Illei PB, Morgan G, Teperman L, Henegariu O, Krause DS. Liver from bone marrow in humans. Hepatology 2000;32: Cassiman D, Libbrecht L, Sinelli N, Desmet V, Denef C, Roskams T. The vagal nerve stimulates activation of the hepatic progenitor cell compartment via muscarinic acetylcholine receptor type 3. Am J Pathol 2002;161: