Chinese Contribution to Immune Thrombocytopenia: the Pathogenesis-Oriented Treatment
Qin Ping,Peng Jun,Hou Ming
DOI: https://doi.org/10.3760/cma.j.issn.0366-6999.20123000
IF: 6.133
2013-01-01
Chinese Medical Journal
Abstract:Primary immune thrombocytopenia (ITP) is the most common hemorrhagic disorder characterized by isolated thrombocytopenia in the absence of conditions known to cause thrombocytopenia. The incidence rate of ITP is similar in males and females; the incidence was significantly higher among patients older than 60 years.1 The clinical manifestation of bleeding varies with patient. The severity of bleeding ranges from mild mucocutanous bleeding (petechiae and ecchymoses) to life-threatening hemorrhages (intracranial or gastrointestinal bleeding); some patients have no bleeding manifestations. Fatal and major nonfatal hemorrhages were extremely higher in patients older than 60 years old.2 The thrombocytopenia in ITP has been traditionally thought to be caused by the production of autoantibodyies targeting platelet antigens. Autoantibodies attach to the platelets and are destructed by macrophages in the reticuloendothelial system. Understanding of the mechanisms of thrombocytopenia in ITP has greatly expanded in recent years. The pathogenesis of ITP is heterogeneous and complex. The thrombocytopenia in ITP is caused not only by increased platelet destruction but also by impaired platelet production. The advances in the understanding of ITP have led to the development of a novel nonimmunologic therapeutic approach by stimulating platelet production and even revised guidelines in the diagnosis and treatment of ITP.3 In the process of exploring the pathogenesis of ITP, the researchers from all over the world have done a lot of studies. This review will provide an insight into the pathogenesis-oriented treatment of ITP, mainly focusing on the studies of Chinese researchers. PATHOGENESIS Recent evidence suggests that ITP is a complex disorder mediated by both immune destruction of platelets and suppression of platelet production.3 The underlying stimulus that triggers the platelet autoimmunity is still not known. The loss of immune tolerance to self-antigens is the critical step of the immunological dysfunction in ITP. As a result of tolerance lost, T cells and B cells of ITP patients are activated and recognize platelet self-antigens and initiate platelet destruction and/or megakaryocyte destruction/inhibition. Loss of tolerance Immune tolerance is the mechanism to ensure that B and T lymphocytes, which generally recognize and respond to foreign pathogenic invaders, do not recognize self-antigens. There is much evidence to prove the loss of immune tolerance in ITP. CD4+ T cells autoreactive to platelet glycoprotein (GP) IIb/IIIa have been identified in patients with ITP. A T helper (Th)1 polarized response is seen in active ITP patients, with increased Th1,4,5 Th17,5,6 and Th22 cells.7 Th17 cells are subset of Th cells characterized by the production of interleukin (IL)-17. The percentages of both Th1 and Th17 cells in peripheral blood of ITP patients increased significantly compared with healthy controls and there was a significantly positive correlation between Th1 and Th17 in ITP patients,6 suggesting that Th17 and Th1 cells may have a cooperative or synergetic function in the pathogenic mechanism of ITP, and Th17 cells may promote a cytokine imbalance toward a Th1-type immune response. Th22 cells, characterized by the production of IL-22, showed a positive correlation with the levels of plasma IL-22 as well as Th17 and Th1 cells in ITP patients.7 The analysis of mRNA levels in Th22 and Th17 cells demonstrated that the expression of both signal transducers and activators of transcription 3 (STAT-3) and transcription factor RAR-related organ receptor C (RORC) were significantly increased in ITP patients. Elevated STAT-3 and RORC play important roles in the crossroads of Th17 and inducible regulatory T cell differentiation; these findings indicate that the upregulated Th22 produced a high level of IL-22, thereby activating the STAT-3 pathway in the differentiation of Th17 cells. The serum cytokine profile in patients with ITP is consistent with T cell activation. An increased expression of IL-18,8,9 B cell activating factor (BAFF),10-12 and IL-2113 in active ITP patients promotes Th1 polarization. IL-18 belongs to the IL-1 cytokine superfamily. It stimulates IFN-γ production and results in the initiation and expansion of autoreactive Th1 responses. IL-18BP is a constitutively secreted protein that binds IL-18 with high affinity by neutralizing the activity of IL-18. IL-18 protein and mRNA levels were significantly increased in patients with active ITP, but IL-18BP levels were not significantly elevated, which resulted in an elevated ratio of IL-18/IL-18BP.8,9 These findings indicate that IL-18BP is not sufficient for blocking the activities of IL-18 in active ITP. IL-21 is produced by some types of CD4+ T cells (including Th17, Th1, and Th2 cells) and NK/T cells. IL-21/IL-21R system is involved in the regulation of the immune system, including promoting T cell-mediated humoral immune responses and antibody production, increasing the function of NK cells, and promoting the differentiation of naive Th cells into Th17 cells. Most studies demonstrate that IL-21 positively regulates the growth and functional activities of Th1 cells while others have opposing results. Zhu et al13 reported significantly elevated IL-21 in ITP patients and its positive correlation with Th17 cells and Th1 cells, which indicated that IL-21 might play a role in the pathogenesis of ITP partially by promoting the activity of Thl cell. BAFF belongs to the tumor necrosis factor (TNF) family; it is critical for the maintenance of normal B cell development and homeostasis, T cell costimulation, and Th1-associated inflammatory responses. Patients with active disease had higher levels of plasma BAFF and BAFF mRNA than patients in remission and controls.10-12 rhBAFF promoted the survival of CD19+ and CD8+ cells and increased apoptosis of platelets and secretion of IFN-γ in vitro.11 These results indicated that elevated BAFF was not only associated with humoral immune abnormalities by promoting the survival of B cells, but also associated with the cellular immunity abnormalities by promoting the Th1 cells polarization and survival of T cells. Yang et al14 reported that active ITP had decreased plasma and decreased mRNA expression levels of IL-27 in patients. Further studies are needed to elucidate the precise role of IL-27 in the development of ITP. In addition, an increase in Toll-like receptor 7 (TLR7) expression was found in ITP. In vitro and in vivo (ITP mouse) studies, imiquimod (TLR7 agonist) induced Th1 differentiation, while IRS 661 (TLR7 antagonist) had the exact opposite effect and skewed Th differentiation toward the Th2 subset. These results suggest that TLR7 participates in the pathogenesis of ITP by promoting Th1 polarization.15 Dendritic cells (DCs) from ITP patients are more efficient in stimulating autologous T-cell proliferation in the presence of autologous/ allogeneic platelets than DCs from healthy donors,16 due to the upregulated expression of CD86 in DCs. Stimulated by normal platelet antigen(s), CD4+ T cells from ITP patients have elevated IL-2 secretion compared with controls,17 indicated an increase of platelet-reactive T-cells in ITP. GPIIb/IIIa is one of the major target antigens recognized by platelet-reactive CD4+ T cells. The epitopes that recognize platelet glycoprotein (GP) IIIa on T helper (Th) cells have been determined and mapped by several groups.18,19 These autoreactive T-cells in active ITP patients might be polyclonal and oligoclonal.20 They lose their immune tolerance to self-antigen(s)21,22 stimulated by normal platelet antigen(s) to secrete IL-2 and promote the production of anti-GPIIb/IIIa antibodies by peripheral blood B cells in patients.16,17 The interaction between GPIIb/IIIa-reactive T and B cells occurs primarily in the spleen in vivo.23 Aberrant expression of apoptosis-associated genes (such as Fas, A20, and Bax) in autoreactive T cells24 suggested the apoptotic resistance of activated T-lymphocytes. Decreased number or dysfunction of CD4+CD25+Foxp3+ T-regulatory cells (Tregs)25,26 suggested defective suppression effects on activated T-lymphocytes, which may lead to the expansion of self-reactive T cell clones. Increased platelet destruction It is well established that both humoral immunity and cellular immunity are involved in the destruction of platelets in ITP. A new murine model of severe ITP induced by both antibody and CD8+ T cells has confirmed that both T cells and B cells participate in the immune disorder of ITP.27 Autoantibodies against platelet antigens are detectable in 50%-70% of ITP patients.28,29 Most of these autoantibodies target GPIIb/IIIa and Ib/IX on the surface of platelets. These antibodies are mainly IgG subtype, but IgM and IgA can be detected in some patients.28 Autoantibodies produced by B cells attach to the surface of platelets and bind to Fc receptors expressed on macrophages of the reticuloendothelial system and then clear platelets from circulation. Both splenic CD5+ B cells and CD5- B cells produce platelet GP-specific autoantibodies in chronic ITP,30 and play a role in the pathogenesis of ITP. Cytotoxic lymphocytes are activated in ITP.31,32 CD3+ T cells from patients with ITP show increased expression of cytotoxic genes, such as Apo-1/Fas, perforin, granzyme A, and granzyme B. Hou et al30 demonstrated the destruction of autologous platelets by CD14- and CD19- mononuclear cells with a modified 111In release assay, suggesting that T cell-induced lysis of platelets may contribute to platelet destruction in ITP. Zhang et al32 isolated the CD8+ T cells and NK cells from ITP patients and cocultured them with autologous platelets and observed platelet lysis by CD8+ T cells rather than NK cells. Apoptosis and perforin/ granzyme-mediated cytotoxicity constitute an important pathway through which CTLs destruct autologous platelets. Zhao et al33 detected the GPIIb/IIIa and GPIb autoantibodies with modified monoclonal antibody-specific immobilization of platelet antigen (MAIPA) assay and measured CTL-mediated cytotoxicity in each ITP patient. A significant difference was found in CTL-mediated platelet lysis between MAIPA positive (11/22) and MAIPA negative ITP patients (21/26), suggesting that CTL-mediated autologous platelet lysis may be the dominant mechanism of thrombocytopenia in ITP patients without autoantibodies and CTLs might be a reasonable target for therapeutic approach. Decreased platelet production Recently, more evidence proved that the thrombocytopenia in ITP was caused not only by increased platelet destruction but also by impaired platelet production.34,35 It was demonstrated by in vitro experiments that autoantibodies from patients with ITP actually inhibit megakaryocyte growth in culture, supporting the view that autoantibodies in ITP suppress megakaryocyte production and maturation and platelet release.34 Yang et al35 cultured CD34+ cells from healthy umbilical cord blood mononuclear cells in medium containing thrombopoietin (TPO), stem cell factor, IL-3, and 10% plasma from either ITP patients or healthy subjects. Similar to McMillan et al,34 plasma from some ITP patients (14/49) suppressed megakaryocyte production. However, plasma from most ITP patients (26/49) boosted megakaryocyte quantity but impaired its quality, resulting in significantly less polyploidy cells and platelet release. These megakaryocytes had a lower percentage of cell apoptosis, a lower expression of TNF-related apoptosis-inducing ligand (TRAIL), and a higher expression of Bcl-xL. These results suggested that decreased apoptosis of megakaryocytes also contributes to reduced platelet production in vitro. The abnormal expression of TRAIL and Bcl-xL in megakaryocytes may play a role in the pathogenesis of impaired megakaryocyte apoptosis in ITP. Except for antibody-mediated suppression of megakaryocyte production and maturation, CD8+ T cells might suppress megakaryopoiesis too. Li et al36 reported that in culture systems megakaryocyte count was increased in the presence of autologous CD8+ T cells from ITP patients, while platelet production was reduced. Lower percentages of polyploidy and apoptotic megakaryocytes and higher levels of soluble Fas in supernatant were observed. These studies provide evidence that activated CD8+ T cells of ITP patients might suppress megakaryocyte apoptosis and lead to impaired platelet production in vitro. A circumstantial evidence for decreased platelet production is that the serum TPO levels in patients with ITP were normal or only slightly elevated compared with other thrombocytopenic disorders such as aplastic anemia with TPO levels up to 10 times higher.37 Relative endogenous TPO deficiency may play a role in the pathophysiology of thrombocytopenia in ITP. Supplemental therapy with TPO may be successful in some ITP patients. PATHOGENESIS-ORIENTED APPROACHES FOR THE MANAGEMENT OF ITP Restoration of tolerance As we know that the loss of tolerance to self-antigens by T cells is the critical step of the immune dysregulation in ITP, restoration of tolerance and deletion of GP-reactive T cells may be a novel therapeutic approach. Peng et al21 found that platelet-specific anergic T cells could be induced by blocking the B7-CD28 interaction with cytotoxic T-lymphocyte-associated antigen 4 immunoglobulin (CTLA4-Ig) and/or cyclosporin A (CsA) in vitro in patients with chronic ITP. Further studies22 demonstrated that the anergized GP-specific T cells have suppressive effects on both GP-specific T cell proliferation and antibody production by inducing tolerogenic DCs. CTLA4-Ig may become a promising new therapeutic agent for the treatment of chronic ITP, and the combination of CTLA4-Ig and CsA would be a more powerful strategy for the management of refractory ITP. In addition to the induction of anergic T cells, agents that increase the apoptosis of GP-specific T cells, restore the balance of Th1/Th2, increase the number or improve the activity of Tregs could also restore the tolerance and exert their therapeutic effects. High-dose dexamethasone could correct the aberrant Th1/Th25 and IL-18/IL-18BP ratios9 by inhibiting the expression of BAFF11 and increasing the number of Tregs25 in ITP patients. IL-18BPa/Fc, a disulfide-linked homodimeric protein transcribed from a DNA sequence encoding amino acid residues 1-192 of human IL-18BP and fused to the Fc region of human IgG1 via a polypeptide linker, could correct the Th1/Th2 ratio of ITP patients in vitro, suggesting its therapeutic potential in the treatment of ITP.38 In addition, BR3-Fc, a novel BAFF blocking reagent, restores the apoptosis of both B and T cells.12 Rituximab39 and indirubin40 could increase the number or improve the function of Tregs in ITP. Zhang et al41 reported that GP-specific-induced Tregs (GP-iTregs) could be successfully generated de novo from nonregulatory CD4+CD25-CD45RA+ cells in vitro, and iTreg could further induce antigen-specific suppression of other GP-reactive T cells by modulating the T cell stimulatory capacity of DCs, thus providing a clue to the potential use of inducing GP-iTregs from the patients in vitro for the antigen-targeted cellular immunotherapy. In conclusion, the induction of T cell tolerance may be a useful strategy for the treatment of ITP. Decreased platelet destruction Corticosteroids, intravenous immunoglobulin (IVIG), and splenectomy are the main treatments for ITP patients. For those patients unresponsive to corticosteroids and/or splenectomy, the treatment is difficult. Hou et al42 treated refractory ITP with mycophenolate mofetil (MMF); the overall response rate was 62% (13 of 21), including 24% (5 of 21) of complete response (CR). The CD3+ cells, CD19+ cells, and antibody levels were significantly decreased in responders 12 weeks after MMF administration. These findings suggested that MMF might correct the immunologic abnormalities underlying the destruction of circulating platelets in ITP. Several studies in ITP have shown high response rates both prior to and after splenectomy with the use of rituximab. In a systematic review of 19 observational studies enrolling 313 patients, the rates of CRs and overall responses with rituximab were 43.6% and 62.5%, respectively, with a median follow-up of 9.5 months.43 In a recent follow-up study enrolling 72 adult ITP patients treated with rituximab who achieved CRs and PRs that lasted at least 1 year, 21% of adults with chronic ITP treated with standard dose rituximab maintained a treatment-free response for at least 5 years without major toxicity. Initial CR and prolonged B cell depletion predicted sustained responses.44 The appropriate dose of rituximab in ITP patients is not yet established; the most used dose is 375 mg·m-2·w-1 for 4 weeks. It seems that lower doses (100 mg·m-2·w-1 for 4 weeks) could be useful in achieving responses and minimizing adverse effects, but time to response was longer than standard dose.45 Liu et al46 demonstrated decreased expression of FcγRIIb (inhibitory FcγR) and increased expression of FcγRIIa and FcγRIII (activating FcγRs) on monocytes in active ITP patients, and the ratio of FcγRIIa/IIb expression returned to normal after dexamethasone treatment. FcγRIIb could decrease antibody production and inhibit activating FcγR functions such as phagocytosis and proinflammatory cytokines release of monocytes and DCs. An increase in the expression of inhibitory FcγRIIb relative to the activating FcγRs could correct the dysfunction of macrophages in ITP, resulting in decreased destruction of platelets. Increased platelet production Several types of thrombopoietin receptor agonists (TPORAs), including recombinant human TPO (in China); peptide TPO mimetics such as romiplostim; and nonpeptide TPO mimetics such as eltrombopag, have been developed and approved for clinical use in patients with chronic ITP in recent years (second-line treatment). The reported response rates to TPO-RAs range from 59% to 88% and time to response is 1-4 weeks.47-49 Cessation of these drugs results in a rapid decline of platelet counts back to baseline in most patients; therefore, a maintenance therapy is required. Most reported adverse effects with TPO-RAs were mild. Patients may experience mild-to-moderate headache, fatigue, myalgias, and gastrointestinal disturbances. Deep venous thrombosis owing to secondary thrombocytosis has been occasionally reported. Eltrombopag has hepatotoxicity and it is advised that ALT and bilirubin levels are measured before and during treatment. Although the toxicity of these drugs seems acceptable, long-term safety data beyond 5 years are limited. The potential toxicity of these agents should be monitored in subsequent long-term clinical trials. Combined therapy for ITP Rituximab provides long-term CR comparable with any therapy other than splenectomy, but the time to response is too long. Adding other agents to rituximab may improve the efficacy. In a randomized trial in newly diagnosed ITP patients with rituximab combined with one cycle of high-dose dexamethasone, sustained response (i.e., platelet count > 50×109/L for 6 months) was achieved by 63% of patients in the rituximab plus dexamethasone group compared with 36% of patients receiving dexamethasone alone.50 The ongoing multicenter clinical trial of recombinant human thrombopoietin combined with rituximab or IL-11 combined with rituximab is currently underway in Chinese group. PERSPECTIVE The mechanisms of thrombocytopenia in ITP involve both increased platelet destruction and impaired platelet production. As we learn more about the pathogenesis of ITP, new therapeutic approaches will emerge. TPO-RAs appear to be very effective and tolerable in refractory patients. The potential long-term side effects of these agents are currently a major limitation. Variability in natural history and response to therapy suggests that primary ITP is heterogeneous. Future research will require understanding of the pathophysiology in each ITP patient. For example, in the use of rituximab in ITP, certain patients will get a great response, while others get no response at all. There may be differences between these two groups of patients. Zeng et al51 found that ITP patients with anti-GPIb antibodies are less responsive to steroid therapy. Our recent research showed that the presence of autoantibodies to GPIb/IX is an independent predictive factor for poor response to IVIG treatment. Identifying these nonresponders is important to avoid this ineffective treatment and decrease unnecessary IVIG consumption. More progress in ITP requires a consortium to assemble a large international multicenter cohort for prospective study. Controlled clinical trials to compare treatments of ITP will help us determine the pros and cons of common approaches, investigate which is better, and select treatments for individual patients.