Immune Thrombocytopenia
Adam Cuker and Douglas B. Cines
Departments of Medicine and of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA
Wang Qian Zuo Wenli Translator Wei Xudong Reviewer
Xudong Wei, Department of Hematology, Cancer Hospital of Henan Province, Affiliated Cancer Hospital of Zhengzhou University, Henan Province, China
Immune thrombocytopenia (ITP) is a heterogeneous group of diseases characterized by autoimmune-mediated platelet destruction and impaired platelet production.ITP may occur without an obvious trigger (primary ITP) or secondary to a series of associated factors (secondary ITP), and its diagnosis must be differentiated from other causes of thrombocytopenia. This review focuses on primary ITP in adults, a group for whom the traditional treatment goal is to achieve a platelet count at or above a safe level of ≥30 × 109/L in the majority of patients and to minimize the occurrence of treatment-related complications. With the advent of better tolerated and more effective drugs, the traditional treatment approach has been questioned. These include: shock doses of dexamethasone, rituximab, and prothrombopoietin receptor activators. Recent studies suggest that aggressive treatment given at the time of diagnosis may alter the natural course of ITP and point to the importance of taking quality of survival into account when developing a treatment plan.
Introduction
Immune thrombocytopenia (ITP) is a group of autoimmune bleeding disorders caused by impaired platelet destruction and production mediated by humoral and cellular immunity. Recently, an international working group recommended the use of ITP to name all immune-mediated thrombocytopenia, whether these cases are clinical manifestations of another disease, or are drug-induced (secondary ITP), or have no apparent predisposing factors (primary ITP). This international working group also recommended the inclusion of a platelet count below 100 × 109/L in the diagnosis, rather than 150 × 109/L. This threshold was based on observational evidence that less than 10% of some healthy individuals with platelets stable at 100-150 × 109/L progressed to severe, unexplained ITP over the next 10 years. this review focuses on adults with This review focuses on primary ITP in adults, but will also include some aspects of secondary and pediatric ITP in the relevant sections. The treatment of ITP in pregnancy is discussed separately in this article (see “Thrombocytopenia in Pregnancy”).
Incidence and Demographics
The estimated incidence of ITP in adults is approximately 1.6-3.9 per 100,000 per year, or 9.5-23.6 per 100,000 based on diagnostic codes from the United Kingdom Health Registry; these figures are somewhat lower for hospital discharges in the United States based on the International Classification of Diseases, Ninth Revision (ICD-9). However, the actual incidence of ITP and the number of individuals requiring treatment are uncertain, given the variability of diagnostic criteria and diagnosis codes and the possibility that some patients go untreated due to masking of the condition.
Adult ITP was once thought to be a disorder that caused great distress in young women. Indeed, some recent studies have affirmed a higher incidence in women of childbearing age than in men of the same age group, but surprisingly, studies have also highlighted a higher incidence in older age and no significant difference in the incidence between men and women. It is uncertain whether these findings represent an evolution in the epidemiology of the disease or a change in the pathway used by clinicians to confirm the diagnosis of ITP.
Etiology
The underlying cause of autoantibody production is not known. Although susceptibility polymorphisms in cytokines and Fcg receptors have been described, the inheritance of the disease is less common. increased Th1/Th0 cytokines, decreased suppressor T regulatory cells, and increased B cell activating factors make autoantibodies more likely to be produced in response to stimulation by foreign antigens. Molecular mimetic mechanisms play an important role in the formation of platelet autoantibodies induced by vaccines and certain viral infections. For example, autoantibodies have been found in patients with ITP associated with HIV, hepatitis C virus, and Helicobacter pylori infection, and these antibodies are usually antigenic determinants clusters within glycoprotein IIIa that can cross-react with platelets. Although it is still questioned how anti-platelet autoantibodies are produced and why there are such significant differences in response rates to treatment in patients from different parts of the world in cases of H. pylori infection, this molecular mechanism supports the idea that reduction or eradication of microorganisms leads to remission in most patients with ITP and that the production of autoantibodies due to this mechanism may be different from immunosuppression or immune disorders associated with secondary ITP is different.
Pathogenesis
The reduction in platelet lifespan is the result of autoantibody-mediated clearance of platelets by tissue macrophages, which occurs in essentially all patients. Multiple lines of evidence from platelet kinetic studies confirm the role of immune-mediated inhibition of megakaryocyte and platelet production in the development of ITP in many patients. These studies include in vitro assays of plasma/immunoglobulin G or T cells in patients with ITP leading to apoptosis and suppression of megakaryocyte production, and response to thrombopoietin (TRAs) in patients with ITP. Direct autoantibody-mediated platelet lysis by peroxis has been reported in patients with ITP associated with HIV infection. Anti-platelet antibodies are not detectable in all patients with ITP, and inhibition of antibody-mediated platelet clearance and B-cell function by pharmacologic or surgical means is ineffective in some patients, suggesting that other pathogenic mechanisms may be present, such as autoantibody-mediated apoptosis, antigen escape, T-cell-mediated platelet destruction, or bone marrow suppression.
Table 1. common causes of non-immune thrombocytopenia in adults
Pseudo-thrombocytopenia
Decreased platelet production
l Certain viral infections
l Myelosuppressive therapy (e.g., chemotherapy, radiation therapy)
l Congenital thrombocytopenia
l Alcoholism
l Folic acid or vitamin B12 deficiency
l Bone marrow disorders (e.g. MDS, myelofibrosis, leukemia, other malignancies invading the bone marrow)
increased platelet destruction
l Certain drugs (e.g. heparin, quinine)
l alloimmune thrombocytopenia (e.g. post-transplant purpura)
l Disseminated intravascular coagulation
l Thrombotic thrombocytopenic purpura-hemolytic-uremic syndrome
l cardiopulmonary artery bypass surgery
dilutive thrombocytopenia
splenic retention
l portal hypertension
l infiltrative splenomegaly
MDS, myelodysplastic syndrome
Diagnosis
The diagnosis of primary ITP requires the exclusion of ITP due to non-autoimmune factors (Table 1) and secondary ITP (Figure 1). In particular, congenital thrombocytopenia must be considered in the evaluation of independent thrombocytopenia. The differentiation of congenital thrombocytopenia from ITP can be difficult in the absence of information on the patient’s previous platelet count and studies of family members. In the United States, secondary ITP accounts for approximately 20% of all patients with ITP. In some areas where the infection is endemic, this percentage may be higher (Figure 1). Table 2 lists the recommendations of a recent international consensus conference for the initial diagnosis of primary ITP. This recommendation does not take into account regional differences in ITP response to antibiotic therapy (e.g., low response rates for ITP after H. pylori eradication in the United States) or the cost effectiveness of screening for cases lacking relevant medical history (e.g., testing all patients for immunoglobulin levels). Effectiveness of specific treatment for ITP is an independent and most convincing diagnostic criterion.
Figure 1. Proportions of different types of ITP estimated based on the authors’ clinical experience (from Cines et al, 2009. Used with permission) SLE, systemic lupus erythematosus; APS, antiphospholipid syndrome; CVID, common variable immunodeficiency disease; CLL, chronic lymphocytic leukemia; APLS, autoimmune lymphoproliferative syndrome; post-tx After bone marrow or solid organ transplantation
Table 2. Recommendations for the initial evaluation of adult ITP patients. (Adapted from Provan et al, 2010. Used with permission)
Basic assessment
Tests that may be useful
Tests of unproven or uncertain benefit
Current medical history
Glycoprotein-specific antibodies
Prothrombin
Family history
Anti-cardiolipin antibodies
Reticulocytes
Physical examination
Anti-thyroid antibodies and thyroid function
PaIgG
Complete blood count
Pregnancy test for women of childbearing age
Platelet lifespan measurement
Reticulocyte count
Anti-nuclear antibody
Hemorrhage Time
Peripheral blood smear
Microvirus and CMV PCR quantification
Serum Complement
Immunoglobulin quantification
Bone marrow examination (if needed)
Blood group (Rh)
Direct anti-human globulin test
Helicobacter pylori
HIV
HCV
HCV, hepatitis C virus; PCR, polymerase chain reaction; CMV, cytomegalovirus; PaIgG, platelet-associated immunoglobulin G
Distinguishing between primary and secondary ITP is sometimes difficult. Patients with a clinical diagnosis of primary ITP are often found to have antinuclear antibodies, antiphospholipid antibodies, antithyroid antibodies, or direct anti-erythrocyte globulin positivity. However, with the exception of those patients with thyroid antibodies, it is rare for patients with primary ITP, to progress to another autoimmune disease with significant clinical features.
Natural course of disease
Because patients with ITP usually present with bleeding that requires intervention, there are no prospective studies of untreated adult ITP. The long-term prognosis of adults treated initially with daily oral prednisone and intravenous immunoglobulin or anti-Rh(D) globulin is difficult to determine, and an estimated 5% to 40% of patients do not receive additional therapy. Even those patients who receive several years of active treatment show a somewhat favorable prognosis. Overall, the remission rate decreases with the duration of the disease. Therefore, the IWG recommended that ITP be divided into subtypes according to disease course, i.e., “newly diagnosed ITP” for those with less than 3 months after diagnosis, “persistent” for those with disease lasting 3 to 12 months, and “chronic” for those with disease lasting longer than 3 months. The “chronic type” refers to those who have had the disease for more than 12 months.
Treatment
Treatment principles
The treatment of ITP needs to be individualized. The conventional treatment goal is to achieve a safe platelet count of 30 x 109/L or higher in most people with minimal treatment toxicities. This treatment approach is based on three assumptions: 1) platelet counts are a reliable surrogate marker of bleeding risk; 2) medical interventions cannot alter the natural course of primary ITP; and 3) bleeding and drug toxicity adequately reflect the burden of the disease and its impact on quality of life. Each of these hypotheses will be confirmed and discussed.
Platelet counts
There is evidence that platelet counts predict severe bleeding in patients with ITP. For example, in the absence of other comorbid bleeding disorders, trauma, or surgery, the likelihood of intracranial bleeding with a platelet count above 20 x 109/L is extremely low. When platelets are below 20-30×109/L, intracranial hemorrhage is the main form of bleeding.
On the other hand, maintaining a platelet count of 30×109/L is modest for some patients. The recently published recommendations of the International Consensus Panel state, “Factors that play an important role in decisions about treatment options include the extent of bleeding, comorbidities with a tendency to bleed, the complexity of particular treatments, activity and lifestyle patterns, and tolerance to side effects of therapeutic agents. A number of potential interventions that may contribute to bleeding, such as over-care of the patient, patient expectations, and non-ITP therapeutic medications, may create bleeding risk.” Older patients and those with a prior history of bleeding are at greater risk.
Relief
The drift of B and T antigen epitopes over time, leading to an increase in high-affinity platelet-specific antibodies, provides the rationale for applying immunosuppression as an early intervention right at diagnosis. Recent studies using this approach appear to challenge the notion that the natural course of adult ITP is not altered (see the “First-line treatment of first-time patients” section below).
Quality of life
The International Consensus Panel guidelines raise the importance of patient choice. There is a growing awareness that some patients with ITP suffer from fatigue, fear of bleeding, limitations in daily activities, inability to participate in important occupational and recreational activities, and poor quality of life. Randomized controlled trials of adults with chronic primary ITP have concluded that the two recently approved thrombopoietins, Romiplostim and Eltrombopag (see “Thrombopoietins” section below), may improve quality of life for patients.
Who needs treatment
Treatment is generally not recommended when platelets are above 30 x 109/L and there is no abnormal platelet function or other coagulation disorders, surgery, trauma, concomitant anticoagulants, or injury-prone lifestyles. Consensus guidelines on target platelet values at the time of surgery in patients with ITP have been published by an international consensus panel (Table 3), although most of these are based on experience rather than formal studies, and treatment must be individualized.
Table 3: Target values for platelet counts at the time of surgery in adults with ITP
(Adapted from Provan et al, 2010. Used with permission.) These recommendations are mostly derived from expert opinion rather than formal studies. As a complement to platelet-raising therapy, antifibrinolytic drugs can be used as an adjunct to dental procedures to prevent bleeding.
Surgery
Recommended platelet counts
Dental prophylaxis (scraping, deep cleaning)
R20-30×109/L
Simple tooth extraction
R30×109/L
Complex tooth extraction
R50×109/L
Regional dental anesthesia
R30×109/L
Minor surgery
R50×109/L
Major surgery
R80×109/L
Major neurosurgery
R100×109/L
First-line treatment for first-time patients
For primary patients, corticosteroids and intravenous immunoglobulin or anti-Rh(D) globulin can be used to stop bleeding and raise platelet counts above 30-50×109/L (Table 4). Although these agents are safe and well tolerated, anti-Rh(D) globulin administration has rarely resulted in severe intravascular hemolysis, disseminated intravascular coagulation, and acute renal failure, and therefore should be avoided in patients with a potential propensity for hemolysis or in those with a positive direct anti-human globulin test not due to therapy. Several uncontrolled studies suggest that one to four cycles of oral dexamethasone as initial therapy may improve efficiency and prolong remission without increasing (without additional) toxicity. In one trial, a single course of dexamethasone 40 mg/d applied for 4 days (equivalents to 400 mg/d prednisone) resulted in sustained remission in 50% of adults at first diagnosis. In another study, one cycle of dexamethasone given every 14 days for a total of 4 cycles yielded 86% efficacy, with 74% of patients achieving complete remission with a median remission period of 8 months. We need randomized controlled studies to confirm the superiority of more aggressive treatment given during prednisone therapy. Another recent trial, randomized to patients given intravenous anti-CD20 antibody (rituximab) and dexamethasone as initial therapy, showed higher efficiency and duration compared to those treated with dexamethasone alone. Longer follow-up and controlled studies using the optimal treatment regime are needed to confirm the application of these enhanced interventions at the onset as the correct treatment philosophy.
Table 4. First-line treatment options for adults with primary ITP (Adapted from Provan et al, 2010. Used with permission)
Recommended treatment strategies
Approximate treatment effectiveness
Approximate treatment effective time
Toxicity
Duration of sustained effectiveness
Corticosteroid
Dexamethasone 40mg/d×4d every 2-4 weeks for 1-4 cycles
Effective initially in up to 90% of patients
Several days to several weeks
Varies with length of use: mood swings, weight gain, irritability, anxiety, insomnia, Cushing’s face, buffalo back, diabetes, fluid retention, osteoporosis, skin thinning, alopecia, hypertension, gastrointestinal distress and ulcers, immunosuppression, psychosis, cataracts, opportunistic infections, adrenal insufficiency; hypertension, anxiety. Reduced tolerability to the same dose. The rate of adverse reactions may be reduced when using short-term therapy.
Efficacy up to 50%-80%, up to 80% applied for 3-6 cycles (in 2-5 years of follow-up)
Methylprednisolone 30mg/kg/d x 7d
Up to 95%
4.7d vs 8.4d (high-dose methylprednisolone [HDMP] vs prednisone)
23% of patients still had stable platelet counts (>50×109/L) at 39 months
Prednisone (Long) 0.5-2mg/kg/d x 2-4w
Effective in 70%-80% of patients initially
Several days to weeks
Still indeterminate: estimated 10-year disease-free survival 13%-15
Anti-Rh(D) immunoglobulin
50-70ug/kg
Initial efficacy similar to IVIG (dose dependent)
4-5d
Common: hemolytic anemia (dose-limiting toxicity), fever/chills
Rare: intravascular hemolysis, disseminated intravascular coagulation, renal failure, rare death
Mostly lasts 3-4 weeks, but some can last for months
Intravenous immunoglobulin (IVIG)
0.4g/kg/d x 5d or 1g/kg/d x 1-2d
Effective initially in up to 80% of patients; half of patients achieve normal platelet counts
Rapid; mostly effective in 24h; typical 2-4d
Headache common: often moderate but sometimes severe
Transient neutropenia, renal insufficiency, aseptic meningitis, thrombosis, fever, chills, malaise, nausea, diarrhea, blood pressure changes, tachycardia
IVIG preparations may contain small amounts of IgA and may occasionally cause allergic-like reactions in patients lacking IgA; IVIG with IgA removed may be used
Usually transient; platelet counts return to pretreatment levels 2-4 weeks after treatment; may persist for several months in rare individuals
Hospitalization and emergency treatment
Patients with ITP require hospitalization if they have 1) visceral bleeding or severe cutaneous mucosal bleeding, 2) platelet counts below 10 x 109/L with a history of severe bleeding or poor compliance, and 3) platelets 10-20 x 109/L and current therapy has not been effective (especially those for whom follow-up is expected to be difficult). Most patients with ITP can be treated out of hospital. For patients who cannot undergo splenectomy and do not have a positive direct anti-human globulin test, we use corticosteroids in combination with intravenous immunoglobulin or anti-Rh(D) globulin (in Rh-positive individuals) for emergency treatment until platelets exceed 30×109/L or bleeding stops. Platelet transfusion is commonly used in cases of life-threatening severe organ bleeding or suspected intracranial hemorrhage. For patients who fail initial therapy, a combination of intravenous immunoglobulin and anti-Rh(D) globulin, acting on different IgG-Fcg receptors in combination with vincristine and methylprednisolone, may be effective. Recombinant factor VIIa can be tried in a very small number of patients who do not respond to other treatments and need to stop bleeding as soon as possible (e.g., intracranial hemorrhage). We use e-aminocaproic acid or tranexamic acid as an adjunct to control oral bleeding or nasal bleeding, topical thrombin and fibrin glue for tooth extraction, and progesterone preparations for menorrhagia. It is also important to give general measures to reduce the risk of bleeding, such as discontinuation of medications that weaken platelet function, and control of blood pressure.
Second-line treatment
Given the toxic side effects of long-term corticosteroid use, the dosage of steroids should be minimized after effective first-line treatment. If there is no sign of improvement in the first month of treatment, or if severe steroid-related toxic reactions occur (Table 5), the treatment regimen should be changed. A small number of patients are effective on very small amounts of corticosteroids, and even fewer patients may be treated with alternate-day therapy. Danazol alone or in combination with azathioprine, as well as aminophenazone, are used as alternatives to steroids. However, for most patients, treatment options including splenectomy, rituximab or thrombopoietin are available after failure of first-line therapy.
Splenectomy
For decades, splenectomy has been considered the standard of care for patients with ITP who are unresponsive or intolerant to prednisone. Two-thirds of patients maintain stable long-term remission, and other patients who achieve partial remission can be treated with less applied salvage therapy. Even the small percentage of patients who do initially fail to respond to splenectomy therapy can eventually maintain a safe platelet count without supportive therapy. When performed by an experienced surgeon, the complication rate and mortality rate of transperitoneal splenectomy is low, and the incidence of fatal sepsis is reduced if the recommended vaccination schedule is subsequently given and antibiotics are administered at the early signs of systemic febrile disease. The efficiency rate is lower in elderly patients, but there is no reliable indicator that can predict the effectiveness of treatment, with the possible exception of indium-labeled platelets, but this method is also not widely used. Recently, late complications of splenectomy have received increasing attention, including atherosclerosis, pulmonary hypertension, and impaired immune surveillance mechanisms. Although splenectomy largely offers the highest chance of remission and cost effectiveness, few internists recommend and few patients choose splenectomy as the preferred second-line treatment.
Anti-CD-20 monoclonal antibodies
A single course of rituximab (375 mg/m2・w for 4 weeks) results in complete remission (defined here as a platelet count >150×109/L) at 1 year in nearly 40% of patients, a third at 2 years, and 15%-20% of patients remain in complete remission at 5 years. Platelet counts begin to rise within 1-2 weeks of treatment, suggesting effective clearance of platelet autoantibodies, but more durable effects are seen months after treatment. Those patients who initially achieve complete remission remain effective after relapse. In contrast, those in partial remission usually relapse within a year and generally have no lasting effect on re-treatment. A small uncontrolled study showed that a smaller dose (100 mg/w for 4 weeks) was equally effective despite a later onset of action. Serious side effects on infusion are uncommon, but respiratory distress may occur. Rituximab is considered contraindicated in patients with active hepatitis B. Rituximab has been found to cause persistent abnormalities in T and B cells and an impaired response to glycolytic vaccine antigens, but the clinical significance of this finding is uncertain. The role of rituximab in these actively treated patients with progressive multifocal leukoencephalopathy is uncertain. We need additional long-term safety data to determine the risk factors for the development of this progressive fatal toxicity and to quantify the risk. Other anti-CD-20 antibodies are in development or entering clinical trials.
Thrombopoietin receptor agonists
Two thrombopoietin receptor agonists (TRAs), Romiplostim and Eltrombopag, have been approved by the FDA for use in patients with ITP who have been treated with corticosteroids as initial therapy and who still require treatment. In some parts of the world, their use is still limited to patients who need treatment after splenectomy,.
Romistatin is a “polypeptidome” consisting of four identical peptides that specifically bind and activate the cMPL Fc segment of the thrombopoietin receptor, prolonging its half-life. The drug is administered subcutaneously once a week (1-10ug/kg). In a placebo-controlled, two-sample, double-blind, randomized phase III trial, 63 splenectomized patients and 62 non-splenectomized patients were administered the study drug for 6 months. At the final 6-8 weeks of this study, 61% of patients treated with romiplostim and 38% of those in the non-splenectomy group and 38% of those in the splenectomy group achieved the treatment goal of a platelet count of 50 x 109/L or higher for at least 6 weeks and did not apply rescue therapy. In contrast, only 1 of 42 study subjects in the placebo group achieved this goal. Many romiplostim-treated patients reduced or discontinued corticosteroid-based maintenance therapy and had improved quality of life. In an ongoing unblinded extension study, most patients treated with romiplostim had more stable platelet levels, some of whom had been treated with romiplostim for nearly 5.5 years.
Eltrombopta is administered orally at a dose of 25-75 mg/d. It must be taken 1 h before or 2 h after a meal and should not be administered within 4 h of the use of drugs or products containing polyvalent cations, such as acid suppressants, dairy products, or micronutrient supplements. The initial dose for patients in Southeast Asia should be reduced by 50% of the normal dose. Dose adjustment is required due to inhibition of resulvastatin or other OATP1B1 carriers by eltrombopta. The efficacy of eltrombopta was observed to be similar to that of romiplostim in clinical trials. In a phase III clinical trial lasting 6 weeks, 114 subjects were randomized to eltrombopta or placebo. Approximately 59% of patients in the treatment group and 16% of patients in the placebo group achieved a treatment goal of a platelet count of 50 x 109/L or greater, respectively, at week 6. Patients treated with eltrombopta were less likely to bleed or require rescue therapy and could have less adjuvant therapy. Similar results were reported in another 6-month phase III clinical trial. In an ongoing unblinded extension study, treatment response was reported to last up to 2 years.
Romilastine and eltrombopta were generally well tolerated. In some patients, an increase in myeloid reticulin was observed within the first 1 year of TRA treatment. However, treatment inefficacy due to myelopathic signs is rare and reversible in most cases. More long-term studies are needed to illustrate the incidence and natural course of TRA-associated myelofibrosis and to compare these findings with the effects of other ITP drug treatments on bone marrow histology.
Clinical trials have found that nearly 10% of patients who did not continue Romistatin or Eltrombopta had platelets fall back below pre-initiation levels. Such toxicity may be avoided or reduced by tapering off TRAs, increased monitoring, and early addition of other ITP therapeutic agents.
To date, there is no reliable evidence to confirm whether TRA increases the incidence of thromboembolic complications. When all studies on romiplostim were analyzed together, there was no significant difference in the incidence of thromboembolism between the treatment and placebo groups (8 and 10/100/year, respectively). In a comprehensive analysis of a study of etriboplat treatment in ITP patients, 17 of the 377 patient-years treated with etriboplat developed thromboembolic complications. Although no such events were documented in the placebo-treated group, the use of this drug was limited to 26 patient-years. Most TRA-related thromboembolism occurred in patients with preexisting atherosclerosis or thrombotic risk factors and with normal or reduced platelet counts. Additional studies on the incidence of TRA-associated thromboembolism and identification of potential risk factors have been approved in light of recent evidence that eltrombopta may increase the risk of venous thrombosis in patients with hepatitis C and, in some patients, a predisposition to thrombosis in ITP itself.
To date, the risk of TRA treatment-related leukemia has been reported to be low. The incidence of hematologic neoplasms was low and similar in the treatment and placebo groups in controlled trials of romiplostim and eltrombopar. In a study of patients with myelodysplastic syndrome-associated thrombocytopenia, romiplostim caused a transient and reversible increase in circulating naive cells in a small number of individuals, but the incidence of leukemic transformation was not significantly different between the treatment and placebo groups.
Abnormal laboratory tests of the hepatobiliary system were found in 13% of patients in the eltrombopta-treated group, suggesting that liver function should be monitored regularly. In a preclinical trial, cataract formation was observed in juvenile rodents given eltrombopta at several times the clinical dosage in humans. Until the risk factors for cataract development at the clinically recommended dose of eltrombopta are clear, regular eye examinations must be performed.
Third-line treatment
Several small uncontrolled studies have shown several immunosuppressive agents to be effective alone or in combination; these include azathioprine, cyclophosphamide, mescaline, and cyclosporine (Table 5). However, because of their poor safety profile, these drugs are generally used only in patients who do not respond to or are unsuitable for first- and second-line therapy. Vincristine analogs are rarely used and hematopoietic stem cell transplantation is chosen in rare cases. New drugs that act on antibody production, platelet production, or clearance are in development.
Table 5. Second- and third-line treatment options for adult ITP (adapted from Provan et al, 2010. Used with permission)
Recommended treatment strategies
Approximate treatment efficacy
Approximate duration of treatment effectiveness
Toxicity
Duration of sustained effectiveness
Azathioprine 1-2 mg/kg (maximum amount: 150 mg/d)
Up to 2/3 of patients; some reports of 40%
Slow; may require continued use for 3-6 months
Incidence of and generally mild: weakness, hyperhidrosis, elevated transaminases, severe neutropenia with infection, pancreatitis
Continued remission after discontinuation of therapy in 1/4 patients
Cyclosporine A 5mg/kg/d x 6d followed by 2.5-3mg/kg/d (blood concentration controlled at 100-200ng/ml)
Dose-dependent. Small percentage up to 50%-80%
3-4 weeks
In most patients, the following may occur to varying degrees, moderately but transiently: elevated blood creatinine, hypertension, malaise, sensory abnormalities, gingival hyperplasia, myalgia, dyspepsia, hirsutism, tremor
Low-dose maintenance remission (at least two years) required in more than half of those who are effectively treated
Cyclophosphamide (1-2 mg/kg orally daily for at least 16 weeks) or IV (0.3-1 g/m2 given as 1-3 doses every 2-4 weeks)
24%-85
1-16 weeks
Mostly mild to moderate: neutropenia, acute deep vein thrombosis, nausea, vomiting
Up to 50% sustained remission
Danazol 200 mg 2-4 times daily
67% complete or partial remission; some reports of 40%
3-6 months
Common side effects: acne, facial hairiness, increased cholesterol, amenorrhea, elevated transaminases
46% can sustain remission for a median period of 119±45 months with a median duration of treatment of 37 months with danazol
Aminophenone 75-100mg
50% efficacy
3 weeks
Common but treatable/reversible: abdominal distention, anorexia, nausea, methemoglobinuria, hemolytic anemia in G6PD deficient individuals
Continued remission after the end of treatment in 2/3 of those with effective treatment
Mycophenolate 100 mg twice daily for at least 3-4 weeks
Up to 75%; complete remission up to 45%
4-6 weeks
Mild and infrequent: headache (most common and dose-limiting), back pain, abdominal distention, anorexia, nausea
Lasted only a short time after interruption of treatment
Rituximab 375 mg/ m2/w x 4 (lower doses may also be effective)
60%; complete remission rate of 40%
1-8 weeks
Low incidence, often mild to moderate, with fever/chills, rash, or throat pruritus on initial infusion
More severe reactions include serum sickness and (rarely) bronchospasm, allergic reactions, pulmonary embolism, renal artery spasm, infection, hepatitis C virus activity leading to fulminant hepatitis progression Rarely progressive multifocal white matter encephalopathy
15-20% of treatment effective individuals may be in sustained remission for >3-5 years. Treatment effective individuals may require retreatment after months to years
Splenectomy
80% effective; nearly 2/3 consistently effective
1-24 days
Bleeding, peripancreatic hematoma, subdiaphragmatic abscess, wound infection, death, pneumococcal infection, fever, severe sepsis, thrombosis
Sustained remission without any subsequent treatment for 5-10 years in nearly 2/3 of patients
TPO receptor agonists
Eltrombopag 25-75 mg orally daily
Elevated platelet levels (platelet count >50 x 109/L on day 43 of the study): 70% in the 50 mg dose group and 81% in the 75 mg dose group
Increased platelet counts on day 15 in more than 80% of patients receiving the 50mg or 75mg dose of Eltrombopag
Adverse events in at least 20% of patients: headache
Treatment-related serious adverse events: 13% experienced an increase in myeloid reticulin, interruption of therapy leading to further platelet reduction, thrombosis, liver function abnormalities
Continued use of the drug for up to 1.5 years
TPO receptor agonists
Romiplostim 1-10ug/kg weekly subcutaneous injection
Total efficiency: non-splenectomy group: 88%; splenectomy group: 79%
1-4 weeks (platelet count <30×109/L rises to >50×109/L in patients)
Adverse events in at least 20% of patients: headache, fatigue, rhinorrhea, arthralgia and contusions (incidence similar to placebo group)
Treatment-related serious adverse events: increased bone marrow reticulin, interruption of therapy leading to further reduction in platelets, thrombosis
Continued use of the drug for up to 4 years
Vincristine analog: total vincristine dose 6 mg (1-2 mg weekly infusion); total vincristine dose 30 mg (10 mg weekly infusion), in some patients either vincristine or vincristine can be infused
Varies widely, with transient efficacy achieved in 10%-75% of cases
5-7 days
Neuropathy, especially in elderly with repeated multiple applications; neutropenia, fever, infusion site inflammation/thrombophlebitis
6 of 9 patients can have normal platelet counts after 3-36 months (mean 10 months) of long-term treatment
Summary and future research directions
ITP is a group of immune-mediated abnormal thrombocytopenic syndromes, but they vary in pathogenesis, natural course, complications, and response to therapy. As our understanding of disease pathogenesis has improved, the term “primary ITP” has been restricted to patients whose underlying etiology is unknown and whose pathogenic pathways have not been elucidated, and the proportion of such patients is decreasing. The role of thrombocytopenia remains uncertain and cannot be monitored clinically, but in most patients, TRAs control antibody-mediated platelet clearance. It is possible to apply efficient instrumentation to screen for and clone autoantibodies, identify their antigenic epitopes, and find external antigens that predispose platelet antibodies to formation in order to explore rational therapeutic approaches. Identification of B and T cell clones that play a role in the development of the disease could enhance drug monitoring and lead to more alternative treatments.
The therapeutic approach to ITP continues to evolve. In the absence of markers for bleeding risk, pathogenesis, and therapeutic effectiveness, platelet counts, although less precise, may serve as alternative markers. The International Working Group report suggested that patients with recalcitrant ITP “may still experience spontaneous remission and that strong therapy should be deferred if possible. This recommendation is challenged by studies that have shown high rates of complete remission with early administration of strong treatments, although the long-term benefits and toxicity of these treatments still need to be determined in controlled studies with long-term follow-up.
Second-line therapy, which is predicated on the presence of bleeding and platelet reduction to 30×109/L, is based on the use of prednisone followed by splenectomy, which is cost-effective and effective in avoiding bleeding-related death. However, the toxicity of corticosteroids and the long-term effects of splenectomy influence the choice of patients and internists, and this approach is associated with reduced quality of life in treatment-naive individuals. Rituximab has relatively mild side effects, but the effects of multicycle immunosuppressive therapy have not been rated in large numbers of patients or after follow-up and compared with splenectomy. TRAs offer a promising new treatment option that challenges the importance of maintenance therapy and expense, but long-term safety studies are still needed. New drugs that act on antibody production, platelet formation and clearance are in development. Our understanding and treatment of ITP has progressed dramatically in recent years and will continue to do so.
Public Statement
Public Statement of Conflict of Interest: Both GlaxoSmithKline and Amgen use DBC as a reference. Drugs not approved for use: Only intravenous gammaglobulin, anti-Rh(D) globulin, romiplostim, and eltrombopta are approved for use in the treatment of ITP. All other drugs discussed in this article for the treatment of ITP are for non-conventional use.
Correspondence
Adam Cuker, MD, MS, Hospital of the University of Pennsylvania, 3 Dulles, 3400 Spruce Street, Philadelphia, PA 19104; TEL: (215) 615-8015; FAX: (215) 615 -6599; E-mail: [email protected]