Primary biliary cirrhosis

* Objective.-Primary biliary cirrhosis is a progressive liver disease that is believed to be autoimmune in nature. Treatment, at best, may slow the progression of the disease, although no therapy has been able to halt its progression. Preliminary data suggest a beneficial effect of methotrexate in the treatment of primary biliary cirrhosis. We evaluated the histologic effect of 2 years of treatment with methotrexate.

Design.-Liver biopsies were obtained before methotrexate was started and after 2 years of therapy. Ninety-six paired biopsies from 48 patients with primary biliary cirrhosis were reviewed by a pathologist who was blinded to all clinical history and sequence of the biopsies. Variables examined included stage of the disease, degree of portal fibrosis, portal inflammation and piecemeal necrosis, bile

duct injury or loss, bile ductular proliferation, lobular inflammation and necrosis, steatosis, granulomas, cholestasis, and nuclear pleomorphism of hepatocytes.

Results.-In most categories, pretreatment and posttreatment biopsies did not reflect a change over the 2-year period of treatment. There was a trend toward progression of the stage of the disease, portal fibrosis, bile duct loss, fat, and pleomorphism over the 2 years and toward regression in piecemeal necrosis, bile duct injury, ductular proliferation, granulomas, and lobular inflammation and necrosis.

Conclusion.-After 2 years of treatment with methotrexate, the stage of disease and fibrosis of primary biliary cirrhosis continue to progress, although overall, inflammation and bile duct injury decrease with methotrexate treatment. (Arch Pathol Lab Med 1998;122:342-345)

Although no firm data confirm an immunopathogenic cause of primary biliary cirrhosis (PBC), considerable evidence points to PBC as an autoimmune disease. Such data include the recent identification of the target antigen of the mitochondrial antibody along with the presence of various other autoantibodies often found in patients with this disease.lz Histologic examination of biopsy specimens showing cytotoxic T cells that appear to be causing bile duct destruction, as well as the similarity of graft versus host disease and allograft rejection to PBC, provides further support of the immunopathogenesis of this disease.34 Other supporting evidence of the autoimmune nature of PBC includes the high incidence of other autoimmune diseases associated with PBC, including Sjogren's syndrome, rheumatoid arthritis, scleroderma, and autoimmune thyroiditis.7 Therefore, the approach to the therapy of PBC should involve the use of immunomodulatory agents. Trials with some of the immunosuppressive agents (thus far including cyclosporine, azathioprine, chlorambucil, and prednisone) have been discouraging, however, because of toxicity of the various agents.8-13

Although the exact mechanism by which ursodeoxycholic acid might be effective is not clear, this therapy is most widely used at present. Despite clinical benefit noted with ursodeoxycholic acid, this therapy is neither a cure for the disease nor does it prevent death or the need for transplantation.l4 Thus, the search for a more effective treatment of PBC continues.

Previous reports have suggested methotrexate as a promising therapy for PBC.15-17 We began treating patients with methotrexate in 1990. Biopsies have been performed at 2-year intervals. This study is a comparison of the pretreatment and posttreatment histology of 48 patients who completed 2 years of treatment with methotrexate.

MATERIALS AND METHODS

Since 1990, more than 100 patients with PBC have begun treatment with methotrexate at our institution. Sixty-eight patients who had completed at least 2 years of treatment with methotrexate were identified. A completed pair of biopsies (taken before and 2 years after methotrexate therapy was initiated) was available for 48 of these 68 patients. Twenty patients were therefore excluded from the present study. Reasons for exclusion included lack of a followup liver biopsy because it had not been performed by the time of the initial review (n = 12), inability to locate the actual liver biopsy slides for review (n = 6), or discontinuation of methotrexate at 2 years, without a follow-up biopsy, because of adverse effects of the drug (n = 2). Discontinuation of methotrexate was due to recurrent mouth ulcers in one patient and accelerated bone loss thought to be due to methotrexate in another.

The 48 remaining patients were included in this review. All had biopsies available. The criteria used to diagnose PBC included elevation of the alkaline phosphatase and gamma glutamyl transferase activities along with detection of mitochondrial antibody (present in 98%). Histologic confirmation of the diagnosis of PBC, in accordance with criteria previously described, was also required.18 The average patient's disease duration at the time of the second biopsy was 9.3 years (range 4-22 years).

Each patient completed 2 years of therapy with 15 mg of methotrexate per week for a cumulative dose of greater than 1.5 g. Other therapy for PBC, including ursodeoxycholic acid and colchicine, was discontinued 6 months before patients began methotrexate. Twenty-six patients did not receive therapy other than methotrexate during the 2-year period. Ursodeoxycholic acid was started in 18 patients after 1 year of methotrexate alone. Four patients began ursodeoxycholic acid at the same time methotrexate was started.

The 96 biopsy specimens were examined in pairs by a pathologist who was blinded to the clinical history and sequence of the biopsies (ie, whether the biopsy was performed pretreatment or after 2 years of therapy). Several variables were analyzed and graded, including determination of the stage"8 and extent of fibrosis, grading of the degree of portal inflammation, piecemeal necrosis, macrovesicular steatosis, and lobular inflammation and necrosis (Table 1). Assessment of the presence of bile ductular proliferation, granulomas, and hepatic pleomorphism was performed along with estimation of the percent of bile duct loss in each biopsy specimen.

After all data were recorded, the grading of each paired set of biopsies was classified as showing progression, regression, or no change over the 2-year period. A X2 test was performed for statistical analysis.

RESULTS

The majority of variables examined showed no difference between the biopsy taken at entry and that obtained after 2 years of treatment with methotrexate. Although no variable assessed reached statistical significance, showing either improvement or deterioration over the 2-year period, several trends were noted.

Progression of the stage of the disease was seen in 21% of paired biopsies, compared with only 6% showing a lower stage after 2 years. The stage of the disease prior to treatment did not appear to have an effect on the results posttreatment; however, the number of patients with stage I disease was too small for proper analysis (Fig 1). Twenty-three percent of patients had increased fibrosis, compared with 8% having decreased fibrosis after 2 years of methotrexate. Progression of bile duct loss was seen in 27%, compared with 19% having less bile duct loss following treatment (Table 2).

A greater proportion of paired slides showed a decrease rather than an increase in piecemeal necrosis (29% vs 23%), bile duct injury (40% vs 10%) and ductular proliferation (4% vs 2%), and lobular inflammation (27% vs So/a) and necrosis (23% vs 8%) (Table 2).

Two years of treatment with methotrexate decreased the presence of granulomas in 20 paired biopsies, compared with 2 paired biopsies that showed an increase in the presence of granulomas. Nuclear pleomorphism of hepatocytes was increased in 17% of biopsies, compared with a decrease in 2%. Macrovesicular fat droplets increased in 29% and decreased in 8% of biopsies, but they were numerous only in obese patients (Table 2).

The majority of patients had not received therapy other than methotrexate (ie, addition of ursodeoxycholic acid) during the 2-year period. Because it was only a small subset of patients who had received either 1 or 2 years of ursodeoxycholic acid along with methotrexate, inferences regarding the effect of ursodeoxycholic acid on the histology were limited. Nonetheless, a comparison of those who had received ursodeoxycholic acid along with methotrexate did not reveal any major differences in change of stage, fibrosis, or inflammation compared with those who had taken methotrexate alone.

COMMENT

The findings presented suggest that a 2-year course of methotrexate does not stop the progression of PBC histologically. The trend was toward progression of disease, as evidenced by a greater number of patients displaying an increase in fibrosis, stage of the disease, and bile duct loss despite therapy (Fig 2).

Improvement in Knodell scoring of PBC patients treated with methotrexate has been reported by one group.19 The use of a cumulative score, such as Knodell scoring, however, may not accurately assess disease progression in PBC. Improvement in inflammation may overshadow increased fibrosis, resulting in a better Knodell score despite evidence of disease progression. Results of three previous studies with methotrexate did not support a histologic benefit from this therapy.zo-zz The number of patients included in these reports and the extent of histologic review in these studies was limited, although the results are in agreement with our findings. This is in contrast to another report of biochemical and histologic improvement in 5 of 19 prearrhotic patients treated with methotrexate.23 The trend we observed toward regression of bile duct injury, piecemeal necrosis, and lobular inflammation could be considered evidence of either disease progression or regression. As PBC progresses histologically, inflammation and bile duct injury become less prominent histologic features of the disease.ls By contrast, regression of inflammation and bile duct injury could signify lessening of the immune-mediated attack on the bile ducts, perhaps resulting from the action of methotrexate. Because a control group was not included, we cannot say whether these findings were due to treatment with methotrexate or were evidence of the natural progression of the disease.

The exact pathogenesis of PBC has yet to be elucidated. Speculation exists that early on an unknown, perhaps nonspecific, trigger causes an immune-mediated response in a susceptible individual. Through a series of events, possibly including leakage of hydrophobic bile salts from damaged bile ducts, bile duct destruction and cirrhosis may ensue. With speculation that the early injury in PBC is immune-mediated, any immunomodulatory therapy may need to be initiated early on in the course of the disease. Therefore, the lack of patients with early disease in our study may have negatively influenced our observations and our conclusions about short-term therapy with methotrexate.

The most important standard for evaluating any therapy is its impact on survival. For inclusion in this study, all patients needed to be able to undergo liver biopsy at the completion of the 2 years and were therefore alive and doing well. A 2-year follow-up period, however, is an inadequate amount of time to evaluate survival in a slowly progressive disease such as PBC. These patients are the subject of an ongoing study looking at survival and outcome with methotrexate therapy. Morphologic changes play some role in prognosis, with death or need for transplantation generally occurring in those with cirrhosis (stage IV). Various stages of PBC may be present concomitantly in individual livers. Therefore, sampling error in liver biopsy specimens may result in difficulty utilizing morphology alone to assess effectiveness of therapy. Differences in stages I, II, and III, which rely on assessment of various features in individual portal tracts, may therefore be subject to such sampling error. Progression to stage IV or the cirrhotic phase of the disease, however, suggests disease progression and a worsening prognosis. Increasing fibrosis as observed in this study, therefore, suggests ineffectiveness of treatment.

A positive finding of this study was a lack of evidence of methotrexate hepatotoxicity at least at the 2-year mark or after a cumulative dose of methotrexate ranging between 1.5 to 2 g. Morphologic features of methotrexate hepatotoxicity, including macrovesicular steatosis and hepatocyte pleomorphism, were not noted. Review of the fibrotic pattern noted also did not reveal the intraacinar fibrosis previously described with methotrexate toxicity.24 The administration of methotrexate at weekly rather than daily intervals, the low cumulative dose, and the low alcoholic intake of our patients may have reduced the risk of hepatotoxicity previously described.25

Thus, our preliminary data do not support histologic benefit from methotrexate in the short-term treatment of PBC. A longer follow-up period and assessment of the drug's influence on the natural history of PBC are necessary-and are in progress-before a judgment regarding the efficacy or methotrexate can be made.

References

1. Gershwin ME, Mackay IR, Sturgess A, et al. Identification and specificity of a cDNA encoding the 70kD mitochondrial antigen recognized in primary biliary cirrhosis. J Immunol. 1987;138:3525-3531.

2. Van de Later J, Cooper A, Surh CD, et al. Detection of autoantibodies to recombinant mitochondrial proteins in primary biliary cirrhosis. N Engl J Med. 1989;320:1377-1380.

3. Colucci G, Schaffner F Paronetto F. In situ characterization of the cell surface antigens of the mononuclear cell infiltrate and bile duct epithelium in primary biliary cirrhosis. Clin Immunol Immunopathol. 1986;41:35-42. 4. Krams SM, Van de Water J, Coppel RL, et al. Analysis of hepatic T lymphocyte and immunoglobulin deposits in patients with primary biliary cirrhosis. Hepatology. 1990;12:306-313.

5. Bernuau D, Feldmann G, Degott C, et al. Ultrastructural lesions of bile ducts

in primary biliary cirrhosis: a comparison with the lesions observed in graft versus host disease. Hum Pathol.1981;12:782-793.

6. Portman B, Neuberger JM, William R. Intrahepatic bile duct lesions. In: Calne RY, ed. Liver Transplantation: The Cambridge-Kings College Hospital Experience. New York, NY: Grune and Stratton; 1983:279-287. 7. Kaplan MM. Primary biliary cirrhosis. N Engl J Med.1996;335:1570-1580. 8. Weisner RH, Ludwig J, Lindor K, et al. A controlled trial of cyclosporine in the treatment of primary biliary cirrhosis. N Engl J Med. 1990;322:1419-1422.

9. Minuk GY, Boehme CE, Burgess E, et al. Pilot study of cyclosporine A in patients with symptomatic primary biliary cirrhosis. Gastroenterology. 1988;95: 1536-1563.

10. Christensen E, Neuberger J, Crowe J, et al. Beneficial effect of azathioprine and prediction of prognosis of primary biliary cirrhosis. Gastroenterology.1985; 89:1084-1091.

11. Heathcote J, Ross A, Sherlock S. A prospective controlled trial of azathioprine in primary biliary cirrhosis Gastroenterology. 1976;70:656-660.

12. Hoofnagle J, Davis GL, Schafer DE, et al. Randomized trial of chlorambucil for primary biliary cirrhosis. Gastroenterology. 1986;91:1327-1334. 13. Mitchinson HC, Palmer IM, Bassendine MF, et al. A controlled trial of prednisolone treatment in primary biliary cirrhosis. J Hepatol.1992;15:336-344.

14. Heathcote EJ, Cauch-Dudek K, Walker V, et al. The Canadian multicenter double blind randomized controlled trial of ursodeoxycholic acid in primary biliary cirrhosis. Hepatology.1994;19:1149-1156.

15. Kaplan MM, Knox TA, Arora S. Primary biliary cirrhosis treated with low dose pulse methotrexate. Ann Intern Med. 1989;107:429-431.

16. Kaplan MM, Knox TA. Treatment of primary biliary cirrhosis with low dose weekly methotrexate. Gastroenterology. 1991;101:1332-1338.

17. Hoofnagle JH, Bergasa NV. Methotrexate therapy of primary biliary cirrhosis: promising therapy but worrisome. Gastroenterology. 1991;101: 1440-1442.

18. Popper H, Schaffner F. Non suppurative destructive chronic cholangitis and chronic hepatitis. In: Popper H, Schaffner F, eds. Progress in Liver Diseases, III. New York, NY: Grune and Stratton; 1970:336-354.

19. Kaplan MM, Dickstein G, Schmid C. Methotrexate improves histology in primary biliary cirrhosis. Hepatology 1994;20:A221. Abstract. 20. Bergasa NV, Jones A, Kleiner DE, et al. Pilot study of low dose oral methotrexate treatment for primary biliary cirrhosis. Am I Gastroenterol.1996;91:295299.

21. Lindor KD, Dickson ER, Jorgenson RA, et al. The combination of ursodeoxycholic acid and methotrexate for patients with primary biliary cirrhosis: the results of a pilot study. Hepatology 1995;22(4 pt 1):1158-1162. 22. Gonzalez-Koch A, Brahm J, Antezana C, Smok G, Cumsille MA. The combination of ursodeoxycholic acid and methotrexate for primary biliary cirrhosis is not better than ursodeoxycholic acid alone. J Hepatol.1997;27:143-149. 23. Kaplan MM, DeLellis RA, Wolfe HJ. Sustained biochemical and histologic remission of primary biliary cirrhosis in response to medical treatment. Ann Intern Med. 1997;126:682-688.

24. Zimmerman HI, Ishak KG. Hepatic injury due to drugs and toxins. In: Mac Sween RNM, Anthony PP, Scheuer PP, Burt AD, Portmann BC, eds. Pathology of the Liver. 3rd ed. New York, NY: Churchill Livingstone Inc; 1994:584. 25. Dahl MGC, Gregory MM, Scheuer P. Methotrexate hepatotoxicity in psoriasis-a comparison of different dose regimens. BMJ. 1972;1:654-655.

Accepted for publication December 19, 1997. From the Department of Medicine, Division of Liver Diseases (Drs Bach and Schaffner), and the Lillian and Henry M. Stratton-Hans Popper Department of Pathology (Drs Thung and Schaffner), The Mount Sinai Medical Center of the City University of New York, NY.

Reprint requests to 5 E 98th St, 6th Floor, New York, NY 10029 (Dr Bach).

Copyright College of American Pathologists Apr 1998
Provided by ProQuest Information and Learning Company. All rights Reserved