Immune thrombocytopenic purpura in patients with chronic hepatitis C virus infection

THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2002 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.

Vol. 97, No. 8, 2002 ISSN 0002-9270/02/$22.00 PII S0002-9270(02)04201-6

Immune Thrombocytopenic Purpura in Patients With Chronic Hepatitis C Virus Infection Paul J. Pockros, M.D., Andrea Duchini, M.D., Robert McMillan, M.D., Lisa M. Nyberg, M.D., John McHutchison, M.D., and Eric Viernes, M.D. Division of Gastroenterology and Hepatology and the Division of Hematology/Oncology, Scripps Clinic, La Jolla, California

OBJECTIVE: Hepatitis C virus (HCV) infection has been associated with the production of autoantibodies and the development of several autoimmune disorders. Immune thrombocytopenic purpura (ITP) is an immune-mediated syndrome of unknown etiology characterized by the presence of autoantibodies against platelet membrane proteins. METHODS: Retrospective chart review. RESULTS: Seven patients with chronic HCV infection (five with cirrhosis and two with chronic active hepatitis) developed thrombocytopenia, out of proportion to their liver disease, and were diagnosed with ITP based on the presence of anti-platelet antibodies and their response to treatment. The number of patients with ITP which occurred in a population of 3440 HCV patients seen over this time interval is much greater than would be expected by chance (p ⬍ 0.00001). Six patients required treatment and four required hospitalization. Four of the six responded to corticosteroids alone. Both of the patients who failed to respond to corticosteroids responded to cyclophosphamide. No mortality occurred from complications of thrombocytopenia. CONCLUSIONS: ITP occurs more commonly in patients with chronic HCV infection than would be expected by chance. This should be considered in patients with liver disease and unexplained thrombocytopenia, as well as in patients with newly diagnosed ITP. Evaluation of antiplatelet antibodies, using an antigen-specific assay, was useful in supporting this diagnosis. Therapy with either corticosteroids or cyclophosphamide was successful in the six patients who required treatment. (Am J Gastroenterol 2002;97:2040 –2045. © 2002 by Am. Coll. of Gastroenterology)

INTRODUCTION Immune thrombocytopenic purpura (ITP) is a common hematological syndrome characterized by antibody-induced platelet destruction by the reticuloendothelial system. The syndrome can be primary or associated with a variety of disorders, including collagen-vascular disease, lymphoproliferative disorders, or viral infection (1). Autoantibodies against platelet glycoprotein (GP) IIb/IIIa or GPIb/IX can be demonstrated in the majority of patients (2, 3).

Hepatitis C virus (HCV) infection may trigger immunological mechanisms and the development of autoimmunity (4, 5). This has been implicated in the pathogenesis of numerous extrahepatic manifestations of HCV including mixed cryoglobulinemia (6, 7), lichen planus (8), polyarteritis nodosa (9), Sjo¨gren’s syndrome (10), diabetes mellitus (11) and leukocytoclastic vasculitis (12). Furthermore, numerous autoantibodies might be produced during HCV infection, including antinuclear (ANA), anti-smooth muscle (ASMA), anti-liver-kidney microsomal (LKM), and antithyroid antibodies (13, 14). We report seven cases of ITP in patients with chronic HCV infection (HCV-ITP) who were seen at our institution between 1995 and 1999. The clinical and laboratory features are summarized together with the response to treatment and subsequent clinical course.

PATIENTS AND METHODS Diagnosis of HCV-ITP The diagnosis of HCV-ITP required the following: ● ● ● ●

The diagnosis of HCV either before or concurrent with documentation of thrombocytopenia. Thrombocytopenia out of proportion to the severity of the liver disease. A positive antiplatelet antibody. A response to agents known to be effective in the treatment of ITP (e.g., corticosteroids, cyclophosphamide) in patients who required therapy (six of the seven patients).

Antiplatelet Antibody Assay This assay has been published previously (15). Briefly, polystyrene beads were coated for 2 h at room temperature with murine monoclonal antibody (either 2A9 for antiGPIIb/IIIa studies [provided by Dr. Virgil Woods, UCSD] or P3 for anti-GPIb/IX studies [provided by Dr. Zaverio Ruggeri, Scripps]) and then blocked with 2% bovine serum albumin. After washing the beads four times with 0.05% Tween 20 in PBS (PBS-Tween), a bead was incubated for 2 h with 1.0 ml of either Triton X-100 solubilized patient or control platelets (108 per ml) to allow binding of immune complexes. After washing, bound IgG antibody was de-

AJG – August, 2002

ITP in Hepatitis C

2041

Table 1. HCV-ITP: Patients Demographics Patient

Age/Sex

Risk

Liver Histology

Viral Genotype

Viral Replication

Antiviral Therapy

1

44/M

i.v. DU

Cirrhosis

1a

5,000,000

None

2 3 4 5 6 7

61/M 47/F 60/F 50/F 70/M 48/F

Transfusion Transfusion Unknown Unknown Transfusion Needle stick

CAH Cirrhosis CAH Cirrhosis Cirrhosis NA

1b 1b

7,000,000 3,100,000

Interferon None None None None None

1,600,000 1,700,000

Antiplatelet Antibodies* Anti-GPIIb/IIIa: 75.6 Anti-GPIb/IX: 50.4 Anti-GPIIb/IIIa: 147.6 Anti-GPIIb/IIIa: 9.8 Anti-GPIIb/IIIa: 39.7 Anti-GPIIb/IIIa: 3.3† PAIgG: Positive Anti-GPIIb/IIIa: 28.6 Anti-GPIb/IX: 13.8

i.v. DU ⫽ intravenous drug use; CAH ⫽ chronic active hepatitis. * Anti-glycoprotein antibody results are expressed as the ratio of cpm 125I-anti-IgG bound to patient sample/mean cpm bound to three normal control samples (normal ⬍2.0); PAIgG-platelet-associated IgG. † Assay done after a treatment response to cyclophosphamide which is known to reduce antibody positivity (15).

tected with 125I-monoclonal antibody specific for human IgG (HB-43, American Type Culture Collection, Rockville, MD). Results are expressed as a ratio of counts per minute patient sample/mean counts per minute bound to three normal control samples. A normal ratio is ⬍2.0. Statistics The probability of ITP and HCV occurring in the same patients by chance was evaluated using a binomial calculation to estimate statistical significance.

RESULTS Patient Demographics During a 54-month interval, seven patients with HCV-ITP (Table 1) were diagnosed among 3,440 new HCV cases seen at our institution. Four patients were female and three were male; ages ranged between 44 and 70 yr. Comorbid conditions included hypothyroidism (n ⫽ 2) and diabetes mellitus (n ⫽ 2). All patients had positive hepatitis C serology. Five patients had chronic hepatitis and cirrhosis at presentation (four diagnosed at liver biopsy and one by radiology and clinical examination) and two had chronic active hepatitis without cirrhosis. Other etiologies of chronic liver disease were excluded. Transmission of HCV occurred through blood transfusion (n ⫽ 3), needle stick (n ⫽ 1) or i.v. drug use (n ⫽ 1); the origin of HCV was unknown in two patients. Of the three patients who were transfused, all transfusions were given for unrelated causes and no patient was thrombocytopenic at the time blood was given. The average time interval between transfusion and diagnosis of ITP was 22 yr (range 11– 40 yr). Five patients had evidence of active viral replication with viral loads between 1,600,000 and 7,000,000 copies/ml at the onset of ITP; data were not available on the other two patients. In the three patients studied, the viral genotypes were: Ia (patient 1) and Ib (patients 2 and 3). Clinical Features All patients presented with easy bruisability; two patients had epistaxis, which was severe in one. Six patients had

splenomegaly; one (patient 2) was asplenic following surgery for trauma in the past. No patient was on treatment for HCV at the time of ITP diagnosis. Patient 2 had received long-acting interferon alpha for 4 months although this had been stopped 4 wk before the diagnosis of ITP. Two patients (patients 1 and 5) required hospitalization at our center for severe thrombocytopenia, initially in the ICU, and two other patients (patients 4 and 6) were hospitalized in outside facilities at the onset of ITP. One patient (patient 5) developed severe gastrointestinal bleeding from hemorrhagic gastritis and required red blood cell and platelet transfusions. Another patient (patient 1) also received platelet transfusions for severe thrombocytopenia. As shown in Table 1, six of the seven patients had antibodies against platelet GPIIb/IIIa with ratios ranging from 3.3–147.6 (normal ⬍2.0); two of these (patients 1 and 7) also had antibodies against GPIb/IX with ratios of 50.4 and 13.8, respectively (normal ⬍2.0). The antibody assay on patient 5, whose ratio of 3.3 was near normal, was obtained when her platelet count was normal just before the fourth dose of i.v. cyclophosphamide. Cyclophosphamide responses are associated with a marked decrease in the antibody level, (15) which may explain why this patient’s result was only modestly positive. Antiglycoprotein antibodies were not assayed on patient 6 but he had elevated levels of platelet-associated IgG. The autoantibody results (mean ⫾ SD) of patients with congestive splenomegaly from other forms of liver disease (n ⫽ 12: alcoholic cirrhosis ⫽ 5; cryptogenic cirrhosis ⫽ 4; portal vein thrombosis ⫽ 1, nonalcoholic steatohepatitis ⫽ 1 and hepatitis B ⫽ 1) were: anti-GPIIb/IIIa 1.2 ⫾ 0.3 (range 0.6 –1.7) and anti-GPIb/IX 1.3 ⫾ 0.3 (range 0.7–1.7), both within the normal limits. In addition, we studied seven HCV patients with either a normal platelet count or a platelet count compatible with their degree of congestive splenomegaly (62,000 –140,000 per ␮l) with the following autoantibody results (mean ⫾ SD): anti-GPIIb/IIIa 1.6 ⫾ 0.5 (range 1.0 –2.3) and anti-GPIb/IX 1.5 ⫾ 0.3 (range 1.2–1.9). All patients had evidence of other autoimmune markers: ANA (1:40 to 1:160) was present in five patients (patients

2042

Pockros et al.

AJG – Vol. 97, No. 8, 2002

Table 2. HCV-ITP: Response to Therapy Platelet Count (per ␮l) Patient 1 2 3 4 5 6 7

Treatments Steroids Steroids Steroids Steroids, splx, i.v. IgG, danazol Steroids, i.v. IgG, cyclophosphamide Steroids, cyclophosphamide None

Response* Complete (off Rx) Complete (off Rx) Partial (off Rx) Requires periodic steroid Rx Partial (off Rx) Partial (on Rx) NA

Platelet Transfusion

Outcome

131,000 298,000 38,000

Yes No No No

Alive, OLTx Alive Alive Alive

2,000

49,000

Yes

2,000

56,000

No

Deceased: Liver failure Alive

29,000

29,000

No

Alive

Diagnosis

Last

3,000 9,000 5,000 13,000

i.v. IgG ⫽ intravenous gammaglobulin; NA ⫽ not applicable; OLTx ⫽ orthotopic liver transplantation; Rx ⫽ therapy; splx ⫽ splenectomy. * Response: Complete⫽normal platelet count; Partial⫽platelet count ⬎30,000 per ␮l.

1–5); ASMA (1:80 titer) was positive in two patients (patients 5 and 6) and two patients had trace positive cryoglobulins (patients 1 and 7). Platelet counts, on diagnosis, ranged from 1,000 to 29,000 per ␮l (Table 2). Three patients (patients 1, 5, and 6) were anemic and three patients (patients 5, 6, and 7) were mildly leukopenic. Four patients had elevated alfa-fetoprotein values without radiological evidence of hepatocellular carcinoma. Four patients had elevated alfa-fetoprotein values without radiological evidence of hepatocellular carcinoma. A disseminated intravascular coagulation (DIC) panel was tested in the two hospitalized patients and was normal.

The Probability of ITP and HCV Occurring by Chance The seven ITP cases were observed among 3440 new HCV patients seen at our institution over a 56-month period. The binomial probability was calculated assuming that: a) the incidence of ITP in the population is 66 cases/million/year (16) and b) the occurrence of ITP is unaffected by the presence of HCV. The null hypothesis was tested by calculating the Z statistic which gave a p value of ⬍ 0.00001. This provides strong support that there is an increased risk of ITP in patients with HCV.

DISCUSSION

Therapy and Clinical Course As shown in Table 2, six patients needed therapy and one patient (patient 7) did not. Three patients (patients 1, 2, 3) required only corticosteroid treatment. Two patients (patients 1 and 2) achieved normal platelet counts which persist off therapy and another patient (patient 3) maintains safe platelet counts (⬎25,000 per ␮l) off treatment. Patient 4 had a splenectomy which resulted in a normal platelet count for 1 yr. Upon relapse, she received i.v. gammaglobulin (IgG) and danazol without significant improvement and requires chronic corticosteroid therapy to maintain safe platelet counts. Two patients (patients 5 and 6) did not respond to either corticosteroids or i.v. IgG but were successfully treated with cyclophosphamide either by i.v. bolus (500 mg/sqM i.v., patient 5) or p.o. (50 –100 mg p.o./q.i.d., patient 6). Both patients achieved normal platelet counts after treatment and their counts remain greater than 50,000 per ␮l after stopping treatment for 30 months and 16 months, respectively. Figure 1 shows the clinical responses of patient 1 (A: response to corticosteroids) and patient 5 (B: response to cyclophosphamide) who presented with severe, lifethreatening thrombocytopenia. No patient died of complications from thrombocytopenia. One patient (patient 1) has undergone a successful orthotopic liver transplantation and two patients are currently on a waiting list for transplantation. Patient 5 died of liver failure.

Infection with HCV is known to cause profound changes in the host immune response, including the development of autoimmunity. These autoimmune mechanisms may play a role in the pathogenesis of hepatic damage as well as in the development of the extrahepatic complications of HCV infection (16). The virus has been shown to infect lymphocytes and to induce a non-specific B-cell proliferation which could lead to the production of autoantibodies and lymphoproliferative disorders (17, 18). Autoantibodies are found in up to 40% of patients with hepatitis C (14). Their presence does not seem to change the course of the liver disease although immune complex formation might explain the development of complications such as vasculitis (19) or glomerulonephritis (20). In view of the propensity for autoimmune phenomena in HCV infection, the occurrence of ITP in some patients is not surprising. There are three lines of evidence which support the diagnosis of autoimmune thrombocytopenia in the seven patients with hepatitis C described in this report and suggest an association between the two diagnoses. First, each patient developed thrombocytopenia out of proportion to the severity of their liver disease, either concurrent with or following the diagnosis of HCV infection. Second, all patients had demonstrable antiplatelet antibodies, four against platelet GPIIb/IIIa alone, two against both GPIIb/IIIa and GPIb/IX, and one patient had elevated PAIgG platelet antibody IgG. Conversely, 16 patients with congestive splenomegaly

AJG – August, 2002

ITP in Hepatitis C

2043

Figure 1. Clinical response of HCV-ITP to therapy. Patient 1 (A) was treated on diagnosis with high-dose corticosteroids and responded rapidly. The response persisted after the corticosteroids were tapered slowly. Patient 5 (B) was treated initially with high-dose corticosteroids and i.v. IgG on diagnosis with no response. On day 21, cyclophosphamide (500 mg/sq M) was given i.v. resulting in improvement of his platelet count; this was repeated on six occasions and the platelet count remained in safe levels for the period of follow-up.

caused by other types of liver disease had normal results. The autoantibody assay used here measures antibodies which bind to either the platelet glycoprotein complex, which is being tested, or to materials which might bind to these complexes, such as drugs (e.g., quinine or quinidine) or, conceivably, viral proteins. Because patients with hep-

atitis C are prone to produce autoantibodies, it seems most likely that the positive tests were due to autoantibodies against the platelet complexes. Finally, of the six patients who required treatment for their thrombocytopenia, all responded to some form of immunosuppressive therapy. Alternative explanations for these findings must be con-

2044

Pockros et al.

sidered. The association of HCV and ITP may be coincidental because both ITP and HCV infection are relatively common in the general population. However, the occurrence, by chance, of seven patients with ITP in a population of 3440 HCV patients was calculated to be extremely unlikely p ⬍ 0.00001. In addition, the thrombocytopenia may be due, in part, to the direct effect of the virus on platelets. HCV-RNA has been isolated from washed platelets of HCV patients (21) and the virus can replicate in megakaryocytes (17). However, in view of the presence of antiplatelet antibodies and the response of these patients to corticosteroids and cyclophosphamide, an immunological cause seems almost certain. Because three out of seven patients had transfusions as their risk factor for HCV, one could argue selection bias is the entire cause of the reported association between ITP and HCV. This seems highly unlikely because of the documented long time interval between the transfusions and the development of thrombocytopenia. To our knowledge, a prior transfusion does not increase the risk for development of ITP. Interferon, which has been shown to trigger the production of autoantibodies in patients with chronic HCV, may precipitate ITP (22). One of the patients developed ITP 1 month after the completion of treatment with interferon alpha, so that interferon as a causative factor in this case cannot be excluded. Because antiplatelet antibodies are not routinely obtained before interferon therapy, we do not know if the production of antibodies was induced by, exacerbated by, or unrelated to the drug. Others have also noted an association between HCV infection and immune thrombocytopenia. Silva et al. (23) and Pawlotsky et al. (24) reported a higher frequency of HCV positivity in ITP patients than in the general population. In addition, the HCV infection predated the ITP in a small number of their patients (five of 112 and five of 139 patients, respectively), suggesting it might have a causative role. Nagamine et al. (21) noted thrombocytopenia in 151 of 368 patients with HCV infection and elevated PAIgG in 88% of these patients. The PAIgG values varied inversely with the platelet count. PAIgG levels are known to be nonspecific (25, 26), so the importance of these findings is difficult to interpret, although they are certainly compatible with an immune cause. More recently, two groups have reported thrombocytopenia in HCV which is responsive to corticosteroids or i.v. IgG. Hernandez et al. reported 19 patients with chronic thrombocytopenia and HCV infection without hypersplenism. They noted increased PAIgG levels were present in 81% of these patients. Of the eight patients requiring therapy (platelet counts ⬍ 20,000 per ␮l), six of seven responded to prednisone (response duration 2–32 months) and one had transient responses to i.v. IgG (27). Baudeur et al. reported one response to corticosteroids and a transient response to i.v. IgG in two patients with HCV infection and ITP requiring therapy (28). In conclusion we describe seven patients with chronic

AJG – Vol. 97, No. 8, 2002

HCV infection who developed ITP as shown by thrombocytopenia out of proportion to their liver disease and the presence of antiplatelet autoantibodies. Six patients required treatment and all responded to immunosuppressive therapy. Immune thrombocytopenia should be considered in patients with chronic HCV infection and otherwise unexplained severe thrombocytopenia. In addition, patients with newly diagnosed ITP should be screened for HCV infection at the time of diagnosis because treatment may differ. Five of the six patients with ITP associated with HCV responded to either corticosteroids or cyclophosphamide and splenectomy was not required.

ACKNOWLEDGMENT Supported by grant HL61809 from the National Institutes of Health. Reprint requests and correspondence: Paul J. Pockros, M.D., F.A.C.P., Division of Gastroenterology and Hepatology, Scripps Clinic, La Jolla, CA 92037. Received June 13, 2001; accepted Dec. 14, 2001.

REFERENCES 1. McMillan R, Imbach P. Immune thrombocytopenia. In: Schafer AI, Loscalzo J, eds. Thrombosis and hemorrhage. Cambridge, MA: Blackwell Scientific Publications, 1998:643– 64. 2. Tani P, Berchtold P, McMillan R. Autoantibodies in chronic ITP. Blood 1989;59:44 – 6. 3. Kiefel V, Santoso S, Kaufmann E, Mueller-Eckhardt C. Autoantibodies against platelet glycoprotein Ib/Ix: A frequent finding in autoimmune thrombocytopenic purpura. Br J Haematol 1991;79:256 – 62. 4. Dickson RC. Clinical manifestations of hepatitis C. Clin Liver Dis 1997;1:569 – 85. 5. Pivetti S, Novarino A, Merico F, et al. High prevalence of autoimmune phenomena in hepatitis C virus antibody positive patients with lymphoproliferative and connective tissue disorders. Br J Haematol 1996;95:204 –11. 6. Casato M, Taliani G, Pucillo LP, et al. Cryoglobulinemia and hepatitis C virus. Lancet 1991;337:1047– 8. 7. Agnello V, Chung RT, Kaplan LM. A role for hepatitis C infection in type II cryoglobulinemia. N Engl J Med 1992; 327:1490 –5. 8. Agner T, Fogh H, Weismann K. The relation between lichen planus and hepatitis C: A case report. Acta Derm Venereol 1992;72:380. 9. Carson CW, Conn DL, Czaja AJ, et al. Frequency and significance of antibodies to hepatitis C virus in polyarteritis nodosa. J Rheumatol 1992;20:304 –9. 10. Haddad J, Deny P, Munz-Gotheil C, et al. Lymphocytic sialoadenitis of Sjo¨ gren’s syndrome associated with chronic hepatitis C virus liver disease. Lancet 1992;339:321–3. 11. Alison MED, Wreightt T, Palmer CR, Alexander GJM. Evidence for a link between hepatitis C infection and diabetes mellitus in a cirrhotic population. J Hepatol 1994;21:1135–9. 12. Daoud MS, Gibson LE, Daoud S, el-Azhary RA. Chronic hepatitis C and skin disease. A review. Mayo Clin Proc 1995; 70:559 – 64. 13. Gregorio GV, Pensati P, Iorio R, et al. Autoantibody preva-

AJG – August, 2002

14. 15. 16. 17. 18.

19. 20. 21.

lence in children with liver disease due to chronic hepatitis C. Clin Exp Immunol 1998;112:471– 6. Pawlotsky JM, Yahia MB, Andre C, et al. Immunological disorders in C virus chronic active hepatitis: A prospective case-control study. Hepatology 1994;19:841– 8. Fujisawa K, Tani P, Piro L, McMillan R. The effect of therapy on platelet-associated autoantibody in chronic immune thrombocytopenic purpura. Blood 1993;81:2872–7. McMillan R. Therapy for adults with refractory chronic immune thrombocytopenic purpura. Ann Int Med 1997;126:307– 14. Hadziyannis SJ. Nonhepatic manifestations and combined diseases in HCV infection. Dig Dis Sci 1996;41:63S–74S. Ferri C, Zignego AL, Bombardieri S, et al. Etiopathogenetic role of hepatitis C virus in mixed cryoglobulinemia, chronic liver disease and lymphomas. Clin Exp Rheumatol 1995; 13(Suppl 13):S135–140. Nitynand S, Holm G, Lefvert AK. Immune complex mediated vasculitis in hepatitis B and C infections and the effect of antiviral therapy. Clin Immunol Immunopath 1997;82:250 –7. Roth D. Hepatitis C virus: The nephrologist’s view. Am J Kid Dis 1995;25:3–16. Nagamine T, Ohtuka S, Takehara K, et al. Thrombocytopenia associated with hepatitis C viral infection. J Hepatol 1996;24: 135– 40.

ITP in Hepatitis C

2045

22. Shrestha R, McKinley C, Bilir BM, Everson GT. Possible idiopathic thrombocytopenic purpura associated with natural alpha interferon therapy for chronic hepatitis C infection. Am J Gastroenterology 1995;90:1146 –7. 23. Silva M, Li X, Cheinquer LH, et al. HCV-associated idiopathic thrombocytopenic purpura (ITP). Gastroenterology 1992;102:A889 (abstract). 24. Pawlotsky JM, Bouvier M, Fromont P, et al. Hepatitis C virus infection and autoimmune thrombocytopenic purpura. J Hepatol 1995;23:635–9. 25. Kelton JG, Powers PJ, Carter CJ. A prospective study of the usefulness of the measurement of platelet-associated IgG for the diagnosis of idiopathic thrombocytopenic purpura. Blood 1982;60:1050 –3. 26. Mueller-Eckhardt C, Mueller-Eckhardt G, Kayser W, et al. Platelet-associated IgG, platelet survival, and platelet sequestration in thrombocytopenic states. Br J Haematol 1982;52: 49 –58. 27. Hernandez F, Blanquer A, Linares M, et al. Autoimmune thrombocytopenia associated with hepatitis C virus infection. Acta Haematol 1998;99:217–20. 28. Baudeur F, Marty F, Larrouy M, Ducout L. Immunologic thrombocytopenic purpura as presenting symptom of hepatitis C infection. Am J Hematol 1998;57:338 – 40.