Long term outcome and response to therapy of primary biliary cirrhosis autoimmune hepatitis overlap syndrome *

Transcription

1 Journal of Hepatology 44 (2006) Long term outcome and response to therapy of primary biliary cirrhosis autoimmune hepatitis overlap syndrome * Olivier Chazouillères 1, *, Dominique Wendum 2, Lawrence Serfaty 1, Olivier Rosmorduc 1, Raoul Poupon 1 1 Service d Hépatologie, Centre de référence des maladies inflammatoires du foie et des voies biliaires, Hôpital Saint Antoine, Assistance Publique-Hôpitaux de Paris, Faculté de Médecine Pierre et Marie Curie, Paris, France 2 Service d Anatomie Pathologique, Centre de référence des maladies inflammatoires du foie et des voies biliaires, Hôpital Saint Antoine, Assistance Publique-Hôpitaux de Paris, Faculté de Médecine Pierre et Marie Curie, Paris, France See Editorial, pages Background/Aims: Whether primary biliary cirrhosis (PBC) autoimmune hepatitis (AIH) overlap syndrome requires immunosuppressive therapy in addition to ursodeoxycholic acid (UDCA) is a controversial issue. Methods: Seventeen patients with simultaneous form of strictly defined overlap were followed for 7.5 years. First-line treatment was UDCA alone (UDCA) in 11 and combination of immunosuppressors and UDCA (UDCACIS) in 6. Results: Characteristics at presentation were not significantly different between the 2 groups. In the UDCACIS group (f-up 7.3 years), biochemical response in terms of AIH features (ALT!2ULN and IgG!16 g/l) was achieved in 4/6 and fibrosis did not progress. In the UDCA group, biochemical response was observed in three patients together with stable or decreased fibrosis (f-up 4.5 years) whereas the eight others were non-responders with increased fibrosis in four (f-up 1.6 years). Seven of these eight patients subsequently received combined therapy for 3 years. Biochemical response was obtained in 6/7 and no further increase of fibrosis was demonstrated. Overall, fibrosis progression in non-cirrhotic patients occurred more frequently under UDCA monotherapy (4/8) than under combined therapy (0/6) (PZ0.04). Conclusions: Combination of UDCA and immunosuppressors appears to be the best therapeutic option for strictly defined PBC-AIH overlap syndrome. q 2005 Published by Elsevier B.V. on behalf of the European Association for the Study of the Liver. Keywords: Primary biliary cirrhosis; Autoimmune hepatitis; Autoimmunity; Overlap syndrome; Ursodeoxycholic acid; Corticosteroids 1. Introduction Received 27 July 2005; received in revised form 27 September 2005; accepted 10 October 2005; available online 15 November 2005 * The authors who have taken part in this study declared that they have not a relationship with the manufacturers of the drugs involved either in the past or present and did not receive funding from the manufacturers to carry out their research. * Corresponding author. Address: Service d Hépatologie, Hôpital Saint Antoine, 184 rue du Faubourg Saint Antoine, Paris Cedex 12, France. Tel.: C ; fax: address: olivier.chazouilleres@sat.ap-hop-paris.fr (O. Chazouillères). Primary biliary cirrhosis (PBC) and autoimmune hepatitis (AIH) are classically viewed as distinct liver diseases. However, patients presenting with clinical, biochemical, serological, and/or histological features reminiscent of both diseases, either simultaneously or consecutively, have been repeatedly recognised [1 5]. The term overlap syndrome is used to describe these settings [6 10]. Unfortunately, lack of universal agreement on what precisely constitutes a PBC-AIH overlap syndrome has generated considerable confusion in the literature. In 1998, /$30.00 q 2005 Published by Elsevier B.V. on behalf of the European Association for the Study of the Liver. doi: /j.jhep

2 O. Chazouillères et al. / Journal of Hepatology 44 (2006) we proposed strict diagnosis criteria and found an overlap syndrome prevalence of 9% in a large series of PBC patients [8]. It is generally assumed that these criteria provide a diagnostic template that can be consistently applied [11] allowing to address the numerous questions raised by the PBC-AIH overlap syndrome. Treatment is one of the unresolved issues and whether overlap syndrome requires immunosuppressive therapy in addition to ursodeoxycholic acid (UDCA) remains controversial. On one hand, our preliminary findings [8] and those of Lohse et al. [9] have suggested that adjunction of immunosuppressive therapy was beneficial and tragic consequences of a missed opportunity of instituting immunosuppressive therapy in patients with PBC-AIH overlap syndrome have occasionally been reported [12]. On the other hand, Joshi et al. found that response to UDCA therapy in strictly defined PBC-AIH overlap syndrome patients was similar to patients with PBC without features of AIH [13]. Randomized controlled trials are the best way to address therapeutic issues. However, such trials are virtually unfeasible in a restricted subgroup of a relatively rare disease. Despite their obvious limitations, retrospective non-randomized studies may provide useful information in this context. Thus, to further clarify the optimal management of this subgroup of PBC, we report here our detailed experience of the long-term outcome (median follow-up longer than 7 years) of patients with simultaneous variant of PBC-AIH overlap syndrome treated with different regimen, i.e. UDCA alone or combination of UDCA and immunosuppressors, with a special emphasis on AIH biochemical response and fibrosis progression. 2. Patients and methods 2.2. Study design (see Fig. 1) Median follow-up was 7.5 years [ ]. Initial management has been individually decided by the physician in charge of the patient. Because the optimal type of treatment was not known, the managing physician was free to decide whether he will treat first with UDCA monotherapy (13 15 mg/kg/day in divided doses given with meals) (UDCA group) or with combination of UDCA (13 15 mg/kg/day) and immunosuppressors (ISC UDCA group). Immunosuppression regimen was corticosteroid based. Initial dose of predniso(lo)ne was 0.5 mg/kg/day and then progressively tapered when ALT levels were decreased by more than 50%; secondarily and according to the usual guidelines for treatment of AIH, azathioprine or mycophenolate mofetil was added as a corticosteroid sparing agent in most patients. Complete biochemical response of the AIH component was defined by ALT levels lower than twice the ULN [15] and IgG levels lower than 16 g/ L. A combined treatment was subsequently proposed to the patients of the UDCA group who did not respond after at least 6 months of UDCA monotherapy (Fig. 1). Liver biopsy was performed in case of incomplete biochemical response and systematically after at least 2 years of follow-up Histopathological assessment A special attention was paid to activity of chronic hepatitis. As previously done by us and others in PBC [8,16] we used the METAVIR score [17,18] which is based on lobular and interface inflammation to evaluate inflammatory activity from 0 to 3 (A0, no histological activity; A1, mild activity; A2, moderate activity; A3, severe activity). Fibrosis was also graded by using the METAVIR score on a five point scale: 0, no fibrosis; 1, portal fibrosis without septa; 2, few septa; 3, numerous septa without cirrhosis; 4, cirrhosis. Liver tissue specimen was reviewed by a single pathologist (D.W). Progression of fibrosis was defined by an increase of at least one unit of the fibrosis score Statistics Quantitative data were expressed as median-range. Comparisons were made by using non-parametric tests: the Mann Whitney rank sum test and the Wilcoxon rank sum test for paired data. Ratios were compared by using the chi-square (c 2 ) test. Differences with a P value!0.05 were considered significant Study population Charts of the 190 patients with PBC who consecutively presented at our Liver Unit from 1982 to 2001 and who had a minimum follow-up of 1.5 years were reviewed. Patients with overlap syndrome were identified according to the criteria we proposed in 1998 [8]. PBC-AIH overlap syndrome was strictly defined by the association of PBC and AIH. For diagnosis of each disease, presence of at least 2 of the 3 accepted criteria was required. PBC criteria were the following: (1) alkaline phosphatase (AP) levels at least two times the upper limit of normal values (ULN) or g glutamyltranspeptidase (GGT) levels at least five times the ULN, (2) a positive test for antimitochondrial antibodies (AMA), (3) a liver biopsy specimen showing florid bile duct lesions. AIH criteria were the following: (1) alanine aminotransferase (ALT) levels at least five times the ULN, (2) serum immunoglobulin G (IgG) levels at least two times the ULN or a positive test for smooth muscle antibodies (ASMA), (3) a liver biopsy showing moderate or severe periportal or periseptal lymphocytic piecemeal necrosis. In addition, AIH score was determined [14]. Only simultaneous forms of overlap syndrome were considered because consecutive forms are less frequent and raise different diagnostic and therapeutic issues. According to these criteria, 17 patients (8.9%), including the 11 patients we described in 1998 [8], had a simultaneous form of PBC-AIH overlap syndrome among our PBC population. Data from laboratory or other diagnostic investigations were obtained at the initial and at follow-up visits (usually twice a year) after the initiation of therapy. Fig. 1. Treatment regimen [UDCA alone (UDCA) or UDCA and immunosuppressors (UDCACIS)] and biochemical response defined by ALT levels lower than twice the upper normal limit.

3 402 O. Chazouillères et al. / Journal of Hepatology 44 (2006) Results 3.1. Baseline characteristics Characteristics of the 17 patients at the time of referral are shown in Table 1. Three patients had jaundice but none had ascitis or variceal bleeding. A recent liver biopsy specimen was available before therapy in all patients except one who had cirrhosis demonstrated by a biopsy performed 1 year before she was referred to our liver unit and then treated. Seven patients had extensive fibrosis or cirrhosis. A moderate or severe interface hepatitis was present in all patients but one. Two patients were classified as definite AIH (score higher than 15) and the other 15 as probable AIH (score of 10 15) according to the scoring system. Eleven patients received UDCA alone (UDCA group) while six patients were given combined treatment (ISC UDCA group) as first-line treatment (Fig. 1). Initial characteristics of the UDCA group were not significantly different from those of the ISCUDCA group (Table 2). However, in the ISCUDCA group, severe interface hepatitis and ASMA positivity tended to be more frequent and IgG levels higher than in the UDCA group First-line therapy Outcome of the ISCUDCA group (nz6) Median follow-up was 7.3 [ ] years. Four patients received also azathioprine ( mg/day) as a corticosteroid sparing agent. Bilirubin, ALT, AP, GGT, IgG and IgM levels decreased significantly (Fig. 2). ALT decreased to levels persistently lower than twice the ULN in the six patients. Complete biochemical response was achieved in the four non-cirrhotic patients. A liver biopsy was Table 1 Characteristics of the patients at the diagnosis of overlap syndrome Characteristic Value Sex F/M 15/2 Age (year) 41 [21 70] Bilirubin (mmol/l) 19 [8 170] ALT (UI/L) 172 [68 800] AP (UI/L) 250 [ ] GGT (UI/L) 260 [ ] IgG (g/l) 20 [ ] IgM (g/l) 5.5 [ ] AMA nz16 ASMA nz11 ANA nz8 Liver histology (Metavir score) Activity (nz16): A1: nz1; A2: nz9; A3: nz6 Fibrosis: F1: nz4; F2: nz6; F3: nz3; F4: nz4 AIH score 13 [10 17] Abbreviations: AP, alkaline phosphatase (N!100 IU/L); GGT, g glutamyltranspeptidase (N!33 IU/L); ALT, alanine aminotransferase (N!35 IU/L); IgG, immunoglobulins G (N!14.0 g/l); IgM, immunoglobulins M (N!2.3 g/l); AMA, antimitochondrial antibody; ASMA, antismooth muscle antibody; ANA, antinuclear antibody. Positivity was defined by a titer R1/40 for AMA and R1/80 for ANA and ASMA. Table 2 Baseline characteristics of the patients treated by UDCA alone (UDCA) or by combination of UDCA and immunosuppressors (UDCACIS) UDCA (nz11) UDCACIS (nz6) Bilirubin 20 [8 62] 17 [9 170] (mmol/l) ALT (UI/L) 203 [87 800] 222 [91 560] AP (UI/L) 280 [ ] 223 [ ] GGT (UI/L) 257 [ ] 239 [ ] IgG (g/l) 17.9 [ ] 34.9 [ ] IgM (g/l) 5.5 [ ] 5.0 [ ] AMA nz10 nz6 ASMA nz6 nz5 Liver histology Activity A1: nz1; A2: nz7; A3: A2: nz2; A3: nz3 nz3 Fibrosis F1: nz3; F2: nz4; F3: nz2;f4: nz2 F1: nz1; F2: nz2; F3: nz1;f4: nz2 AIH score 13 [10 17] 14.5 [10 17] None of the parameters were statistically different between the two groups. Abbreviations: AP, alkaline phosphatase (N!100 IU/L); GGT, g glutamyltranspeptidase (N!33 IU/L); ALT, alanine aminotransferase (N!35 IU/L); IgG, immunoglobulins G (N!14.0 g/l); IgM, immunoglobulins M (N!2.3 g/l); AMA, antimitochondrial antibody; ASMA, antismooth muscle antibody. Positivity was defined by a titer R1/40 for AMA and R1/80 for ASMA. performed in five patients after 3.5 [2 7] years of therapy; activity was decreased (nz4) or stable (nz1) and fibrosis was unchanged in the four non-cirrhotic patients. One patient with cirrhosis at presentation developed hepatocellular carcinoma 10 years after overlap diagnosis and was transplanted. At the end of follow-up, four patients had a steroid-free therapy (three treated with UDCA alone and one with UDCA and azathioprine) and two patients were still receiving corticosteroids (maintenance dose of 5 mg/ day in one case and 20 mg/day in the other case because of steroid-dependence) Outcome of the UDCA group (nz11) Patients were treated with UDCA alone for 2 [ ] years. Bilirubin, ALT, AP, GGT levels decreased significantly but not IgG and IgM (Fig. 2). In terms of complete biochemical response two subgroups were identified (Fig. 1). UDCA responders. Three patients had a complete and sustained biochemical response. They remained under UDCA monotherapy (follow-up: 4.5 [ ] years) and histological evaluation performed after 3 [2 8] years of treatment showed a decreased (nz2) or stable (nz1) activity and no increase in fibrosis (decreased: nz1 or stable: nz2). UDCA non-responders. The eight other patients (followup: 1.6 [ ] years) were non-biochemical responders (ALT O2 ULN and/or elevated IgG) despite a partial improvement. Three experienced hepatitis episodes with flares of ALT levels above 10 ULN. Because of this

4 O. Chazouillères et al. / Journal of Hepatology 44 (2006) Fig. 2. Biochemical changes in patients treated by UDCA alone (UDCA) and in patients treated by combined therapy (UDCACIS). Data given as box plots representing 10, 25, 50 (, median), 75 and 90 percentiles. *: P!0.05, **: P!0.02 vs start values. incomplete biochemical response, histological reassessment (nz7) was performed after 1.6 [ ] years of treatment. Activity was increased in 3, stable in 2 and decreased in 2. Among the five patients without cirrhosis at entry, four had increased fibrosis including cirrhosis in 2 and only one had unchanged fibrosis. Characteristics of the UDCA responders at onset of therapy did not statistically differ from those of the nonresponders (Table 3). However, none of the three responders had severe interface hepatitis vs three of the eight nonresponders Comparison of responses to UDCA vs UDCAC immunosuppressors Complete biochemical response (4/6 vs 3/11) (PZ0.11), ALT response (6/6 vs 4/11) (PZ0.01) and absence of fibrosis progression (4/4 vs 4/8) (PZ0.08) tended to be more frequent in the UDCACIS group than in the UDCA group. Among the initially non-cirrhotic patients and regardless of the treatment, none of the seven biochemical responders had progression of fibrosis whereas increased fibrosis was documented in four of the five non-biochemical responders (all in the UDCA group) Second-line combined therapy Administration of corticosteroids in addition to UDCA was proposed to the eight UDCA monotherapy nonresponders. One patient declined this proposal because of fear of potential side effects. The seven others subsequently received a combined therapy and were followed-up for 3 [ ] years. In six patients, azathioprine ( mg/day) or mycophenolate mofetil (2 g/day) was added as a corticosteroid sparing agent. ALT decreased to levels persistently lower than two-fold the ULN in these seven patients including six with a complete biochemical response. One patient developed severe relapse when she was treated with 10 mg/day of prednisolone but responded well to increased dosage. At the end of follow-up, patients were still receiving corticosteroids at a maintenance dose ranging from 5 to 10 mg/day except one who had an immunosuppressor-free therapy. One patient who progressed to cirrhosis under UDCA monotherapy developed hepatocholangiocellular carcinoma and died 6 months later. A liver biopsy was performed in three patients after 4 [ ] years of combined therapy. In the two non-cirrhotic patients at the previous biopsy, fibrosis was unchanged (nz 1) or decreased (nz1). In contrast, the only patient who

5 404 O. Chazouillères et al. / Journal of Hepatology 44 (2006) Table 3 Baseline characteristics of the biochemical responders and nonresponders to UDCA alone remained under UDCA monotherapy failed to achieve biochemical response and a new biopsy demonstrated a further increase in fibrosis. Overall, immunosuppressive therapy was given in 13 patients for 4 [0.6 11] years and stopped successfully (ALT levels persistently lower than 2 ULN after) in four after a median duration of 2.7 [ ] years. These 4 patients had persistently normal ALT values 4.5 [ ] years after immunosuppression withdrawal and liver biopsy performed 3 years later in two patients did not show progression of fibrosis. Six patients experienced side effects of corticosteroids. Mild bone loss as assessed by bone densitometry was the most frequent side effect (nz5) but vertebral fractures occurred in only two patients who became asymptomatic after biphosphonate therapy. Other side effects were transient diabetes mellitus (nz2), cosmetic effects (nz2), and obesity (nz1). Finally, among the non-cirrhotic patients with serial liver biopsies available, none out of six had progression of fibrosis after initiation of the combined treatment (follow-up: 4 [ ] years) in contrast to four out of eight under UDCA monotherapy (follow-up: 2.3 [ ] years) (PZ0.04). 4. Discussion Responders (nz3) Non-responders (nz8) Bilirubin (mmol/l) 21 [12 34] 19 [8 62] ALT (UI/L) 172 [97 502] 248 [87 800] AP (UI/L) 237 [ ] 315 [ ] GGT (UI/L) 264 [ ] 229 [ ] IgG (g/l) 19.6 [ ] 17.9 [ ] IgM (g/l) 4.6 [ ] 6.1 [ ] AMA nz3 nz7 ASMA nz1 nz5 Liver histology Activity A2: nz3 A1: nz1; A2: nz4; A3: nz3 Fibrosis F1: nz1; F2: nz1; F3: nz1 F1: nz2; F2: nz3; F3: nz1; F4: nz2 AIH score 13 [11 15] 13 [10 17] None of the parameters were statistically different between the three groups. Abbreviations: AP, alkaline phosphatase (N!100 IU/L); GGT, g glutamyltranspeptidase (N!33 IU/L); ALT, alanine aminotransferase (N!35 IU/L); IgG, immunoglobulins G (N!14.0 g/l); IgM, immunoglobulins M (N!2.3 g/l); AMA, antimitochondrial antibody; ASMA, antismooth muscle antibody. Positivity was defined by a titer R1/40 for AMA and R1/80 for ASMA. Despite a great confusion in the literature with regards to the diagnosis of overlap syndromes, there is no doubt that some patients have features of both PBC and AIH whatever the term used for their description [10,13,19]. In this study, we used the strict diagnosis criteria we proposed in 1998 and that have been subsequently applied by other groups [13]. Interestingly, only 2 out of our 17 patients were classified as definite AIH according to the international AIH scoring system [14]. This is not an unexpected finding since this scoring system is tailored on the assumption that none (or very few) of the patients with cholestatic diseases also have concomitant AIH. As a consequence, biliary features have a strong negative impact and a score of definite AIH can be only observed in the rare patients with caricatural overlap syndrome [19,20]. Therapy of overlap syndrome is a much debated issue. The main finding of our study is that combination of UDCA and immunosuppressive therapy is able to induce biochemical response and to stop fibrosis progression in nearly all patients with strictly defined PBC-AIH overlap syndrome. In this setting, combined therapy appears to be superior to UDCA monotherapy. This is supported by the following: first, biochemical response occurred in only a minority of patients treated with UDCA whereas it was the rule in patients receiving combined therapy; second, non-responders to UDCA monotherapy responded to combined therapy and third, progression of liver fibrosis was observed in most of the patients who received UDCA and in none of those treated with combined therapy. These data support the view that overlap syndrome has an accelerated course under UDCA therapy when compared with classical PBC [21] since progression occurred in a relatively short period of time (medianz2 years). Interestingly, there was a strong association between complete biochemical response of the AIH component and absence of fibrosis progression. These findings extend our earlier observations [8] with more patients and a much longer follow-up (median followup: 7.5 years). It may be argued that the present study included only 17 patients and was not randomized. However, it represents the largest series of overlap patients with available histological outcome to date and randomized trials are virtually impossible to perform in rare diseases. As initial management was individually decided by the physician in charge of the patient, the trend for a more active disease at presentation in the group treated by immunosuppressors is not surprising. This does not alter our conclusions, i.e. the superiority of combined therapy over UDCA monotherapy, since only the opposite (more active disease in the UDCA group) would have been unfavourable to UDCA. A beneficial effect of combined therapy had also been suggested by others [9] but this opinion has been challenged by Joshi et al. [13]. These authors, in the context of two randomized controlled trials of UDCA for PBC patients, compared the response to treatment in subjects with or without overlap syndrome defined according our criteria. They found that, under UDCA therapy, biochemical response at 24 months and survival of overlap patients (nz12) were similar to patients with pure PBC (nz159) and, on the basis of these results, did not recommend corticosteroids for PBC patients with laboratory features of AIH [22]. Despite opposite conclusions, our results are

6 O. Chazouillères et al. / Journal of Hepatology 44 (2006) partly in keeping with theirs. Indeed, as them, we found that some biochemical parameters (including ALT levels) were significantly improved in patients treated with UDCA alone. Furthermore, 3 of the 11 patients with first-line UDCA monotherapy had a complete biochemical response and a stable or decreased fibrosis. Predictive factors of response to UDCA were not identified in our study presumably because of lack of power due to the small number of patients but it must be underscored that these three patients had only moderate interface hepatitis. Their satisfactory outcome is in keeping with the findings of Nakamura et al. who reported that UDCA treatment of mild type 1 AIH led to a decrease in ALT values and to some improvement in liver histology [23]. However, this beneficial effect of UDCA was observed in only a minority of our patients and most of them experienced progression of fibrosis. Death or liver transplantation was unrealistic judgment criteria in our study because, in PBC, it requires a great number of patients and a very long follow-up [24]. Indeed, death (or transplantation) occurred in only two cirrhotic patients (one in each group of treatment) and was related to hepatobiliary malignancies. Interestingly, one patient developed an overlap carcinoma, i.e. hepatocholangiocellular carcinoma. In contrast to Joshi et al. who did not report on changes in fibrosis score, we paid a special attention to fibrosis progression since progression to cirrhosis is considered as a valid surrogate marker to measure aggravation of PBC [24]. Suitability of liver histology as a therapeutic endpoint may be questioned because of sampling error but it must be underlined that lack of fibrosis progression was constantly associated with a marked clinical and biochemical improvement. As a consequence, we do think our findings are relevant in terms of clinical management of overlap syndrome. Care must be taken in extrapolating these results to PBC patients with some features of AIH (such as significant interface hepatitis) but who do not fulfill criteria of overlap syndrome as defined above and who represent a large subset (up to one third of PBC patients) [25]. Some arguments suggest that these patients might also benefit from combined therapy [16,26 28], but large-scale randomized trials are necessary to assess the role of immunosuppression in this setting. Side effects occurred in a minority of patients and, not unexpectedly, were mostly represented by osteoporosis. This underscores the special need for a careful and regular assessment of bone mineral density with early therapeutic intervention when indicated [29]. Lastly, the issue of the optimal duration of immunosuppression was not addressed in our study since withdrawal was not systematically attempted. However, immunosuppressors were successfully stopped in four patients (including three of the six receiving combined therapy as first-line therapy) after a median duration of 2.7 years. These data suggest that forever immunosuppressive therapy is not always required especially in case of early administration but further studies are warranted. We conclude that biochemical response is associated with absence of fibrosis progression and that liver fibrosis rapidly increases in most patients under UDCA monotherapy in contrast to patients receiving combination of UDCA and immunosuppressive therapy. Our results strongly suggest that combined therapy is the best therapeutic option in patients with strictly defined PBC- AIH overlap syndrome. References [1] Kloppel G, Seifert G, Lindner H, Dammermann R, Sack HJ, Berg PA. Histopathological features in mixed types of chronic aggressive hepatitis and primary biliary cirrhosis. Virchows Arch A Pathol Anat Histopathol 1977;373: [2] Okuno T, Seto Y, Okanoue T, Takino T. Chronic active hepatitis with histological features of primary biliary cirrhosis. Dig Dis Sci 1987;32: [3] Carrougher JG, Shaffer RT, Canales LI, Goodman ZD. A 33-year-old woman with an autoimmune syndrome. Semin Liver Dis 1991;11: [4] Ben-Ari Z, Dhillon AP, Sherlock S. Autoimmune cholangiopathy: part of the spectrum of autoimmune chronic active hepatitis. Hepatology 1993;18: [5] Colombato LA, Alvarez F, Cote J, Huet PM. Autoimmune cholangiopathy: the result of consecutive primary biliary cirrhosis and autoimmune hepatitis? Gastroenterology 1994;107: [6] Davis PA, Leung P, Manns M, Kaplan M, Munoz SJ, Gorin FA, et al. M4 and M9 antibodies in the overlap syndrome of primary biliary cirrhosis and chronic active hepatitis: epitopes or epiphenomena? Hepatology 1992;16: [7] Duclos-Vallée JC, Hadengue A, Ganne-Carrie N, Robin E, Degott C, Erlinger S. Primary biliary cirrhosis autoimmune hepatitis overlap syndrome. Corticoresistance and effective treament by cyclosporine A. Dig Dis Sci 1995;40: [8] Chazouillères O, Wendum D, Serfaty L, Montembault S, Rosmorduc O, Poupon R. Primary biliary cirrhosis autoimmune hepatitis overlap syndrome: clinical features and response to therapy. Hepatology 1998;28: [9] Lohse AW, Meyer Zum Büschenfelde KH, Franz B, Kanzler S, Gerken G. Characterization of the overlap syndrome of primary biliary cirrhosis (PBC) and autoimmune hepatitis: evidence for it being a hepatitic form of PBC in genetically susceptible individuals. Hepatology 1999;29: [10] Woodward J, Neuberger J. Autoimmune overlap syndromes. Hepatology 2001;33: [11] Ben-Ari Z, Czaja AJ. Autoimmune hepatitis and its variant syndromes. Gut 2001;49: [12] van Leeuwen DJ, Sood G, Ferrante D, Lazenby AJ, Sellers MJ. A 38- year-old African American woman with an unusually rapid progression of primary biliary cirrhosis : a missed opportunity. Semin Liver Dis 2002;22: [13] Joshi S, Cauch-Dudek K, Wanless IR, Lindor KD, Jorgensen R, Batts K, et al. Primary biliary cirrhosis with additional features of autoimmune hepatitis: response to therapy with ursodeoxycholic acid. Hepatology 2002;35: [14] Alvarez F, Berg PA, Bianchi FB, Bianchi L, Burroughs AK, Cancado EL, et al. International autoimmune hepatitis group report: review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999;31:

7 406 O. Chazouillères et al. / Journal of Hepatology 44 (2006) [15] Czaja AJ, Freese DK. Diagnosis and treatment of autoimmune hepatitis. Hepatology 2002;36: [16] Rautiainen H, Karkkainen P, Karvonen A-L, Nurmi H, Pikkarainen P, Nuutinen H, et al. Budesonide combined with UDCA to improve liver histology in primary biliary cirrhosis: a three-year randomized trial. Hepatology 2005;41: [17] The METAVIR Cooperative Group. Inter- and intra-observer variation in the assessment of liver biopsy of chronic hepatitis C. Hepatology 1994;20: [18] Bedossa P, Poynard T, for the METAVIR cooperative study group. An algorithm for the grading of activity in chronic hepatitis C. Hepatology 1996;24: [19] Poupon R. Autoimmune overlapping syndromes. Clin Liver Dis 2003; 7: [20] Chazouillères O. Diagnosis of primary sclerosing cholangitis autoimmune hepatitis overlap syndrome: to score or not to score? J Hepatol 2000;33: [21] Corpechot C, Carrat F, Bonnand A-M, Poupon RE, Poupon R. The effect of ursodeoxycholic acid therapy on liver fibrosis progression in primary biliary cirrhosis. Hepatology 2000;32: [22] Joshi S, Heathcote EJ. Ursodeoxycholic acid and primary biliary cirrhosis with features of autoimmune hepatitis. Hepatology 2002;36: (reply). [23] Nakamura K, Yoneda M, Yokohama S, Tamori K, Sato Y, Aso K, et al. Efficacy of ursodeoxycholic acid in Japanese patients with type-1 autoimmune hepatitis. J Gastroenterol Hepatol 1998;13: [24] Poupon R. Trials in primary biliary cirrhosis: need for the right drugs at the right time. Hepatology 2004;39: [25] Degott C, Zafrani ES, Callard P, Balkau B, Poupon RE, Poupon R. Histopathological study of primary biliary cirrhosis and the effect of ursodeoxycholic acid treatment on histology progression. Hepatology 1999;29: [26] Corpechot C, Carrat F, Poupon R, Poupon RE. Primary biliary cirrhosis: incidence and predictive factors of cirrhosis in ursodioltreated patients. Gastroenterology 2002;122: [27] Leuschner M, Güldütuna S, You T, Hübner K, Bhatti S, Leuschner U. Ursodeoxycholic acid and prednisolone versus ursodeoxycholic acid and placebo in the treatment of early stages of primary biliary cirrhosis. J Hepatol 1996;25: [28] Leuschner M, Maier K-P, Schlichting J, Strahl S, Herrmann G, Dahm HH, et al. Oral budesonide and ursodeoxycholic acid for treatment of primary biliary cirrhosis: results of a prospective doubleblind trial. Gastroenterology 1999;117: [29] Heathcote EJ. Management of primary biliary cirrhosis. Hepatology 2000;31: