The pathogenesis of chronic immune (idiopathic) thrombocytopenic purpura

The Pathogenesis of Chronic Immune (idiopathic) Thrombocjtopenic Purpura Robert McMilhn Chronic immune (idiopathic) thrombocytopenic purpura (ITP) is an autoimmune disorder in which antiplatelet autoantibody induces platelet destruction. Platelet surface membrane proteins become antigenic, stimulating the immune system to produce autoantibody. The initial antigenic response probably occurs in the spleen, inducing autoantibody production followed by stimulation of other antibody-producing tissues, particularly the bone marrow. Autoantibodies against either platelet glycoprotein (GP)llb/liia or GPlb/iX are produced by about 75% of ITP patients and can be detected using antigen-specific assays. The spleen is the major site of platelet destruction in ITP because of its unique milieu. About one third of the platelet mass is present in the spleen at all times, where the local production of antiplatelet antibody leads to high autoantibody concentrations. These antibody-sensitized platelets circulate slowly through the phagocytic cell-rich spleen, resulting in their destruction. Since autoantibody binds to both platelets and megakaryocytes, both platelet destruction and inhibition of thrombopoiesis may be of importance in the pathogenesis of chronic ITP. Semin Hematol37(suppll):59. Copyright 0 2000 by W.B. Saunders Company.

C

HRONIC IMMUNE (idiopathic) thrombocytopenic purpura (ITP) is an autoimmune disorder characterized by autoantibodyinduced platelet destruction. In this disorder, platelet surface membrane proteins become antigenic, for unknown reasons, leading to stimulation of the immune system, autoantibody production, and platelet destruction primarily by phagocytosisl-3

Autoantibody

and Autoantigens

Harrington et al reported in the early 1950s that the infusion of blood from patients with chronic ITP into normal recipients resulted in thrombocytopenia .* Subsequent studies by Shulman et al showed that the thrombocytopenit effect was dose-dependent and less marked in splenectomized patients and in patients pretreated with corticosteroids or with blockade of the reticuloendothelial system induced by red blood cell membranes. Furthermore, the factor reacted with both autologous and homologous platelets, was in the IgG-rich fraction, and could be removed by adsorption with platelets.5,6 These in vivo studies suggested that the antiplatelet factor was an IgG antibody and that ITP was an autoimmune disease. In 1982, van Leeuwen et al7 showed that antibody eluates from ITP platelets would bind to normal platelets, but most would not bind to thrombasthenic platelets deficient in glycoproSeminars

in Hematology,

tein (GP)IIb/IIIa. They postulated the presence of anti-GPIIb/IIIa autoantibodies in ITP. Subsequently, several laboratories, using antigenspecific assays,have demonstrated the presence of autoantibodies to both platelet GPIIb/IIIa and GPIb/IX in ITP.* Using antigen-specific assays, autoantibody against 1 or both of these platelet surface complexes can be detected in about 75% of ITP patients.gJO Autoantibodies bind to platelets and cause their destruction. They also bind to megakaryocytesli and may affect thrombopoiesis, as suggested by studies of platelet turnover showing normal or reduced turnover in greater than SO% of patients rather than the increased turnover expected.12z13 Approximately 7 5 % of platelet autoantigens lie on either the platelet GPIIb/IIIa or GPIb/IX complexes. In the other 25%, it is believed that other membrane proteins are probably involved. Localization of the antigenic epitopes has only recently been investigated. Earlier studies showed that plasma anti-GPIIb/IIIa autoantibodies, from about 15% of ITP paFrom The Scripps Research Institute, La Jolla, CA. Address reprint requests to Robert McMillan, MD, The Scr&bs Research Institute, 10550 N Torrey Pines Rd, La Jolla, CA 92037. Copyright 0 2000 by W.B. Saunders Company 0037-1963/00/3701-1004$10.00/0

Vol 37; No I, Suppl 1 (January),

2000:

pp 5-9

5

6

Robert McMillan

tients, react with the c-terminus of GPIIIa.14 Since this region is in the cytoplasm, these antibodies probably are due to exposure of cytoplasmic antigens to the immune system resulting from platelet destruction by antibodies directed to surface antigens. Subsequent studies, showing that platelet-associated autoantibodies from patients with plasma anti-cterminus autoantibody react with epitopes located on other areas of the GPIIb/IIIa complex, support this hypothesis.15 These platelet-associated antibodies could be directed to antigens located on GPIIb, GPIIIa or to antigenic sites on the GPIIb/IIIa complex. Antibodies against both extracellular and cytosolic antigens on GPIb have also been detected.16 More recent studies show that most plateletassociated anti-GPIIb/IIIa autoantibodies from chronic ITP patients bind to cation-dependent epitopes. l7 These epitopes may depend on a conformationally intact complex or be located near the calcium-binding sites on either GPIIb or GPIIIa. The latter possibility is less likely in view of studies evaluating antibody binding to a panel of large recombinant peptides spanning the GPIIIa molecule. Only 1 of 33 ITP plateletassociated autoantibodies showed binding to any region of GPIIIa. l8 These results suggest that the GPIIb/IIIa epitopes of ITP patients are rarely localized to GPIIIa alone, but probably require either a conformationally intact GPIIb/ IIIa complex or bind to antigens near the calcium binding sites on GPIIb.

The Immune

Response

The initial antigenic response in ITP should occur in either the spleen, which monitors the intravascular space, or the bone marrow, where megakaryocytes or intramedullary platelets could stimulate the immune system. The spleen is probably more important, since animal studies have shown it to be the primary organ that responds to intravascular antigens, while the marrow assumes greater importance later in the immune response.19 These animal studies are consistent with observations in chronic ITP. The spleen is a major site of antibody production in ITP. In culture, splenic cells from ITP patients produce

IgG, which binds specifically to both autologous and homologous platelets,20 as well as megakaryocytes,‘l resulting in platelet destruction and perhaps also affects platelet production. However, since antibody may persist after splenectomy, there must be additional sites of antibody production. Bone marrow is the most likely site of residual antibody production as shown by in vitro studies in which IgG produced by ITP marrow cells binds to platelets.21

Immunoregulation Normally, the immune system is carefully regulated. Two T-cell populations, CD4+ helper-inducer cells and CD8+ suppressorcytotoxic cells, modulate the response to antigens. These cells and their subpopulations are formed during the course of thymic T-cell maturation. Immature T cells (CD4CDS-) produced in the bone marrow migrate to the thymus and, during maturation, gain or lose a variety of surface proteins and synthesize a T-cell receptor capable of responding to antigen if presented with the appropriate major histocompatibility complex (MHC) molecule. Immature T cells containing self-reactive receptors are either eliminated (thymic deletion) or rendered tolerant (thymic anergy) during the CD4+CDSf intermediate thymic maturation stage. Finally, T cells lose either the CD8 or CD4 antigen and are released as either CD4+ inducer cells or CD8+ suppressor-cytotoxic cells. CD4+ inducer T cells react with processed antigen bound to MHC class II molecules on antigen-presenting cells (macrophages, B cells). Upon activation, 1 population of these cells (helper-inducer cells) stimulates the proliferation and differentiation of antigen-specific B cells into antibody-producing plasma cells and the expansion of CD8+ cytotoxic T cells, which are activated by antigen bound to MHC class I molecules. Another CD4+ population (suppressor-inducer cells) stimulates expansion of antigenMHC class I activated CD8+ suppressor cells, which suppress both humoral and cell-mediated immune responses, preventing overreaction to a stimulus. In addition, the development of antiid-

Patbogenesis

iotypic antibodies to the variable region of specific antibodies is also involved in the downregulation of the immune response. Autoimmunity results from a breakdown at some level of this immunoregulatory network.22 Several studies on immunoregulation in chronic ITP have been recently reviewed.3,*3 Interpretation of the results is difficult due to the small numbers of patients studied, the variability of patient age and disease severity, and the inclusion of treated and untreated patients in some studies; in addition, conclusions were drawn from the study of peripheral blood cells, which may not reflect the immune system as a whole. Early studies reported that T lymphocytes from most chronic ITP patients show decreased suppressor cell activity when cultured with either autologous or homologous B cells. Conversely, ITP B lymphocytes respond to suppression by normal T cells.24 More recent studies*5 show that incubation of ITP lymphocytes with platelets results in lymphocyte activation, manifested by 3H incorporation, and production of interleukin-2. The circulating lymphocyte phenotype profile in this patient group showed a reduction of CD4+ suppressor-inducer cells and an increase in CD4+ helper-inducer cells and DR+ activated T cells and B cells. The decrease in CD4+ suppressor-inducer cells is consistent with observations showing a diminished autologous mixed lymphocyte reaction (AMLR) in chronic ITP. Other studies show reduced suppression of pokeweed mitogen-induced IgG production by T cells stimulated in the AMLR reaction or by concanavalin A. Additional studies, relevant to the pathogenesis of ITP, show the following: increased circulating CD4+CD8+ lymphocytes, consistent with imperfect thymic deletion of selfreactive cells; increased numbers of both blood and splenic CD5+ B cells, which in 2 patients produced platelet-bindable IgM; and the presence of clonal B cell populations. Multiple other studies have evaluated T cells, B cells, T-/B-cell ratios, and T- or B-cell subsets with variable results.

of Chronic

7

ITP

Platelet

Destruction

Following antibody binding, platelets are removed from the circulation in chronic ITP by either phagocytosis or complement-induced lysis. Platelet phagocytosis has been demonstrated by electron microscopy and by in vitro studies,26 and it may be triggered by either the Fc portion of IgG or by C3b fixation to the cell surface. The Fc mechanism is clearly important in ITP, as shown by the effect of anti-Fc blockade using either red blood cell stroma or monoclonal anti-PC receptor antibody27 and the recently published case of a patient with hightiter platelet-associated IgG4 autoantibody (IgG4 is poorly recognized by phagocytic cells) and a bleeding diathesis, but with normal platelet counts.28 Complement activation probably plays a role in some patients with ITP. Increased plateletassociated C3, C4, and C9 have been demonstrated on ITP platelets29a30 and in vitro studies show binding of C4 and C3 followed by platelet lysis after incubation of platelets with patient autoantibody.3l The spleen is the primary site of platelet destruction in ITP, since about one third of the platelet mass is present in the spleen at all times and the local production of antiplatelet antibody subjects the intrasplenic platelets to high antibody concentrations. These antibody-sensitized platelets then circulate slowly through the spleen, which is rich in phagocytic cells. The liver, although a more efficient phagocytic organ, contains no intraorgan platelet pool, produces no local antibody, and has a rapid circulation. The liver assumes importance in severe ITP when the heavily sensitized platelets can be removed by the liver’s efficient reticuloendothelial system. The bone marrow, with a resident megakaryocyte and platelet population, produces antibody and contains a reticuloendothelial system. Since the antibody binds to platelets and megakaryocytes, both inhibition of thrombopoiesis and intramedullary platelet destruction may occur. In summary, there is now ample evidence that ITP is an autoimmune disease. Despite tantalizing hypotheses supported by limited

Robert McMillan

data, there is no clear explanation for the development of this autoimmunity. The inhibition of thrombopoiesis while there is an increase of megakaryocytes in the marrow remains a mystery that requires further study.

14.

15.

References 1. McMillan purpura.

R: Chronic idiopathic thrombocytopenic N Engl J Med 304:1135-1147, 1981

2. Karpatkin S: Autoimmune (idiopathic) thrombocytopenic purpura. Lancet 349:1531-1536, 1997 R, Imbach P: Immune thrombocytopenia, 3 McMillan in Schafer AI, Loscalzo J (eds): Thrombosis and

4

5

6

7

8

9

10.

11.

12.

13.

Hemorrhage. Cambridge, 1998, p 643-664 Harrington WJ, Sprague Immunologic mechanisms thrombocytopenic purpura. 469, 1953 Shulman NR,

Marder

MA,

Blackwell

van Leeuwen EF, van der Ven JTH, Engelfriet CP, et al: Specificity of autoantibodies in autoimmune thrombocytopenia. Blood 59:23-26, 1982 McMillan R: Antigen-specific assays in immune thrombocytopenic. Transfus Med Rev 4:136-143, R,

Tani

P,

Millard

18.

RS: Similarities

between known antiplatelet antibodies and the factor responsible for thrombocytopenia in idiopathic purpura: Physiologic, serologic, and isotopic studies. Ann NY Acad Sci 124:499-542, 1965 Shulman NR, Weinrach RS, Libre EP, et al: The role of the reticuloendothelial system in the pathogenesis of idiopathic thrombocytopenic purpura. Trans Assoc Am Physicians 78:374-390, 1965

1990 McMillan

17.

Scientific,

CC, Minnich V, et al: in idiopathic and neonatal Ann Intern Med 38:433-

VJ, Weinrach

16.

F, et

al:

Platelet-

associated and plasma anti-glycoprotein autoantibodies in chronic ITP. Blood 70:1040-1045, 1987 Kiefel V, Santoso S, Kaufmann E, et al: Autoantibodies against platelet glycoprotein Ib/Ix: A frequent finding in autoimmune thrombocytopenic purpura. Br J Haematol 79:256-262, 1991 McMillan R, Luiken GA, Levy R, et al: Antibody against megakaryocytes in idiopathic thrombocytopenic purpura. JAMA 239:2460-2462, 1978 Heyns A du P, Badenhorst PN, Lotter MG, et al: Platelet turnover and kinetics in immune thrombocytopenic purpura: Results with autologous 11 l-Inlabeled and homologous 5 l-Cr-labeled platelets differ. Blood 67:86-92, 1986 Ballem PA, Segal GM, Stratton JR, et al: Mechanisms of thrombocytopenia in chronic autoimmune thrombocytopenic purpura: Evidence for both impaired platelet production and increased platelet clearance. J Clin Invest 80:33-40, 1987

19.

20.

Fujisawa K, O’Toole TE, Tani P, et al: Autoantibodies to the presumptive cytoplasmic domain of platelet glycoprotein IIIa in patients with chronic immune thrombocytopenic purpura. Blood 77:22072213, 1991 Fujisawa K, Tani P, O’Toole TE, et al: Different specificities of platelet-associated and plasma autoantibodies to platelet GPIIb-IIIa in patients with chronic immune thrombocytopenic purpura. Blood 79:1441-1446, 1992 He R, Reid DM, Jones CE, et al: Spectrum of Ig classes, specificities, and titers of serum antiglycoproteins in chronic idiopathic thrombocytopenic purpura. Blood 83:1024-1032, 1994 Fujisawa K, Tani P, McMillan R: Platelet-associated antibody to glycoprotein IIbiIIIa from chronic immune thrombocytopenic purpura patients often binds to divalent cation-dependent antigens. Blood 81:1284-1289, 1993 Bowditch RD, Tani P, McMillan R: Reactivity of autoantibodies from chronic ITP patients with recombinant glycoprotein IIIa peptides. Br J Haematol 91:178-184, 1995 Askonas BA, Humphrey JH: Formation of specific antibodies and g-globulin in vitro: A study of the synthetic ability of various tissues from rabbits immunized by different methods. Biochem J 68:252261, 1958 McMillan R, Longmire RL, Yelenosky R, et al: Quantitation of platelet-binding IgG produced in vitro by spleens from patients with idiopathic thrombocytopenic purpura. N Engl J Med 291:812-817,

1974 McMillan R, Yelenosky RJ, Longmire RL: Antiplatelet antibody production by the spleen and bone marrow in immune thrombocytopenic purpura, in Battisto JR, Streinlein JW (eds): Immunoaspects of the Spleen. Amsterdam, The Netherlands, North Holland Biomedical, 1976 22. Sinha AA, Lopez MT, McDevitt HO: Autoimmune diseases: The failure of self tolerance. Science 248: 1380-1388, 1990 21. Semple JW, Freedman J: Abnormal cellular immune mechanisms associated with autoimmune thrombocytopenia. Transfus Med Rev 9:327-338, 1995 24. Trent R, Clancy RL, Danis V, et al: Disordered immune homeostasis in chronic idiopathic thrombocytopenic purpura. Clin Exp Immunol 45:9-17, 1981 JW, Freedman J: Increased antiplatelet T 25. Semple helper lymphocyte reactivity in patients with autoimmune thrombocytopenia. Blood 78:2619-2625, 1991 26. McMillan R, Longmire RL, Tavassoli M, et al: In vitro platelet phagocytosis by ITP splenic leukocytes in idiopathic thrombocytopenic purpura. N Engl J Med 290:249-251, 1974 21.

9

Patbogenesis of Chronic ITP

27.

28.

29.

Clarkson

SB, Bussel

JB,

Kimberly

RP,

et al: Treat-

ment of refractory immune thrombocytopenia purpura with an anti-Fc gamma-receptor antibody. N Engl J Med 314:1236-1239, 1986 McMillan R, Bowditch R, Tani P, et al: A nonthrombocytopenic bleeding disorder due to an IgG4kappa anti-GPIIbiIIIa autoantibody. Br J Haematol 95:747-749, 1996 Hauch TW, Rosse WF: Platelet-bound complement

(C3) 30.

31.

in immune

thrombocytopenia.

Blood

5O:l l29-

1136, 1977 Kurata Y, Curd JG, Tamerius JD, et al: Plateletassociated complement in chronic ITP. Br J Haemato1 60~723-733, 1985 Tsubakio T, Tani P, Curd JG, et al: Complement activation in vitro by antiplatelet antibodies in chronic immune thrombocytopenic purpura. Br J Haematol 63:293-300, 1986