Addressing thrombosis concerns in immune thrombocytopenia: the role of fostamatinib in immune thrombocytopenia management

ABSTRACT

Introduction

Immune thrombocytopenia (ITP), a disease that commonly presents with an increased risk of bleeding, can also paradoxically produce an increased risk of thromboembolic events. The risk of thromboembolism can be associated with patient-related factors (e.g. co-morbidities, age and history of thrombosis), disease-related factors (e.g. a greater proportion of younger, more reactive platelets, and the presence of microparticles and pro-inflammatory cytokines) and treatment-related factors (e.g. splenectomy, thrombopoietin receptor agonists, and IVIg).

Areas covered

Aspects of the pathophysiology of ITP and the effects of treatment are discussed with emphasis on individualizing treatment based on the patient’s thromboembolic risk, treatment options and preferences.

Expert opinion

An increased understanding of the pathophysiology of ITP has led to the development of new agents such as fostamatinib, a spleen tyrosine kinase inhibitor. Further research into the factors contributing to the risks for bleeding and thromboembolic events can contribute to the development of more specific therapies for ITP and allow greater individualization of therapy based on each patient’s medical history and clinical status.

KEYWORDS:

• Bleeding
• fostamatinib
• immune thrombocytopenia
• platelets
• thromboembolism

1. Introduction

Immune thrombocytopenia (ITP) is an acquired autoimmune disease characterized by platelet counts less than 100 × 10⁹/L that can result in bleeding and other symptoms. ITP is a complex disease (Figure 1) [1]. The pathophysiology of the disease process is not fully understood, but emerging research has allowed the development of therapies which are more targeted than older therapies. Antibodies, self-reacting T-cells, complement and many other factors are involved in this complex disorder. Autoantibodies against platelet surface antigens lead to macrophage-mediated platelet phagocytosis. In addition, T-cell mediated platelet destruction contributes to the decrease in platelet count. On the replacement side of platelet homeostasis, megakaryocyte maturation is inhibited leading to decreased platelet formation [1]. The cells and pathways involved in the increased destruction and reduced formation of platelets offer multiple avenues for therapeutic intervention in the treatment of ITP.

Figure 1. A simplified depiction of the ITP pathogenesis along with the sites of intervention for new therapeutics (from Provan and Semple [1]).

These recent findings have allowed the design of therapies that increase the production of platelets (green in Figure 1) like the thrombopoietin receptor agonists (TPO-RAs) or inhibit the destruction of platelets (red in Figure 1) by targeting spleen tyrosine kinase (SYK), Fc receptors, Bruton’s tyrosine kinase (BTK), and complement among others. The advancements in basic science have helped to unravel the pathophysiology of ITP and, consequently, to design targeted therapies for the disease. In the near future, there may be as many as six classes of therapy approved for ITP.

Until recently, the management of ITP has been largely unsatisfactory due to a limited efficacy and presence of adverse effects. Patients have also not been satisfied with older treatments due to their side effects and inconvenience (e.g. route of administration) [2]. There’s an unmet need for new classes of therapy that are not immunosuppressive and do not have the same toxicities as current treatments.

Patients with ITP have been recognized as being at increased risk for thrombosis [3–5]. The rate of thromboembolic events (TEEs) in an ITP patient is three to four times higher than in the general population [6]. Additionally, some of the current treatments, such as TPO-RA or splenectomy, may increase thromboembolic risk [7–9], so choosing a therapy with little or no thrombosis risk is preferable in those patients at risk of thromboembolism. This review is focused on the factors contributing to the prothrombotic nature of ITP and the effects of treatment on thrombotic risk.

2. Risk of thrombosis in ITP patients

ITP and thrombosis are an intriguing association [6]. A patient with a low platelet count that develops a thrombosis seems illogical. Conventional thinking would seem to dictate that a patient cannot get a thrombosis if they have a low platelet count due to the role of platelets in forming a clot. Rodeghiero asked in his editorial: ‘Is ITP a thrombophilic disorder?’ [6]. It seems counterintuitive. Table 1 summarizes the factors that increase the risk of thrombosis in ITP.

Table 1. Pro-thrombotic factors in ITP.

Disease Related Factors

Increased proportion of young platelets

Presence of microparticles

Changes in clot viscoelasticity

Platelet – WBC interactions

Non-platelet targeted autoantibodies

Patient-Related Factors

Comorbidities

Post-Surgical Status

Malignancy

Central Venous Line

Trauma

Antiphospholipid Syndrome

Treatment-Related Factors

Splenectomy

TPO-RAs

Corticosteroids

TPO-RAs = thrombopoietin receptor agonists

2.1. Disease-related factors

A recent paper by Han et al. [10] showed that low platelets may sometimes be linked with thrombosis not just with ITP, but also with hereditary and acquired thrombocytopenias. This includes several congenital thrombocytopenias, antiphospholipid syndrome (APS), disease and drug-induced thrombocytopenias (including heparin-induced thrombocytopenia) and a heterogenous group of disorders characterized by thrombocytopenia such as the thrombotic microangiopathies, e.g. thrombotic thrombocytopenic purpura (TTP) [11].

Some studies on the nature of platelets and coagulation status can help explain why patients with ITP develop thrombosis. Firstly, ITP patients have a greater proportion of immature platelets than normal individuals [12,13]. Saxon et al. showed that the percentage of immature platelets was larger in blood from ITP patients (32.9 ± 10.2%, mean ± SD) than in patients with other hematologic diseases (acute lymphoblastic leukemia: 6.6 ± 3.1%; aplasia: 3.4 ± 2.0%) or normal individuals (7.9 ± 2.9%) [12].

Several studies have suggested that immature platelets may be more reactive [14–16]. Therefore, with a greater number of immature platelets, an increased risk of thrombosis could be expected. In patients with bone marrow disease, low platelet counts might cause bleeding because the platelets present are not the immature, more reactive cells seen with ITP [17].

The immature platelet fraction can be measured to assess a patient’s relative bleeding or thrombosis risk. If the immature platelet fraction is low, bleeding risk may be higher and thrombosis risk may be less compared to patients with a high fraction. This correlation needs additional validation.

On the other hand, a greater fraction of younger more hemostatic platelets, can tip the balance toward thrombosis. The clotting system is always a balance of hemostasis and thrombosis that can move from one extreme to the other.

The second pathophysiological aspect of ITP that can be related to thrombosis is the presence of microparticles. Microparticles have been shown to have procoagulant activity [18,19]. This is a somewhat controversial area of research due to some debate about the best method to measure microparticles. In this study by Álvarez-Román et al., more microparticles related to both platelets and red cells were found in ITP patients compared to controls [18]. The platelet microparticles are created by autoantibodies which mediate platelet fragmentation [20]. There are large number of platelet and red cell microparticles in ITP patients and they have been shown to increase the risk of thrombosis [18,19].

The third aspect of hypercoagulability in ITP patients relates to viscoelastic measures. Common viscoelastic measures are the thromboelastogram and rotational thromboelastometry (ROTEM; Pentapharm, Munich Germany). These are commonly used methods to assess the kinetics of clot formation and fibrinolysis. As with microparticles, viscoelastic measures have been a controversial topic in ITP research. The debate centers on how these viscoelastic measures can be used in patients with thrombocytopenia.

Some studies suggest that measures of clot firmness determined by thromboelastometry and the number of circulating immature platelets could be used to assess bleeding risk and hypercoagulability in thrombocytopenic patients, but there is not a clear consensus on this topic [21].

In the previously mentioned paper by Álvarez-Román et al., patients with ITP showed increased maximum clot firmness, reduced lysis, higher clotting time and clot formation time [18]. This may be due to increased resistance to protein C. This state is similar to that seen in APS, where the surrogate marker is activated protein C resistance. Even in the absence of antiphospholipid antibodies (APLAs) in ITP patients, they can have similar laboratory findings to APS. Markers for fibrinolysis were also measured in patients with ITP. Resistance to fibrinolysis was observed in these patients due to increased levels of plasminogen activator inhibitor-1 (PAI-1) [18].

Another important factor regarding thrombosis in ITP patients is the interaction of platelets with white blood cells. In the past, platelet function was considered as being separate from white cells, but they do interact [22]. Platelets can adhere to neutrophils and direct them to the site of vascular injury. In addition, platelets can induce neutrophil extracellular traps (NETs) formation which, in turn can activate platelets [22]. Although pathophysiologically different, APS, which is often characterized by thrombocytopenia, and ITP have multiple factors, in addition to platelets, come together to cause thrombosis, i.e. activation of endothelial cells, monocytes and APLAs [23].

There are also clinical aspects to the relationship between ITP and thrombosis. Five extensive epidemiologic investigations comparing venous and/or arterial thrombosis in chronic ITP versus control populations [6] showed that the adjusted incidence ratio for venous thromboembolism (VTE) (0.41–0.67 per 100 person-years) and arterial thrombosis (AT) (0.96–2.78 per 100 person-years) was higher in ITP patients than in the control population (VTE: 0.09–0.42 per 100 person year; AT (0.67–1.78 per 100 person years) [3,5,24–26]. Therefore, both venous and arterial thrombosis risk is increased in patients with ITP.

Autoantibodies targeted at antigens other than those on the platelet surface may play a role in increased thrombotic risk in ITP patients. Several studies have found that a high percentage (25–75%) of ITP patients have APLAs [27–34]. The role of these autoantibodies in thrombotic events in ITP patients is a matter of debate. Stasi et al. studied 149 newly diagnosed ITP patients for 12 months and found that 42.2% were positive for APLAs. The presence of these antibodies had no effect on the clinical course for these patients including the incidence of thrombosis [30].

In contrast, Diz-Küçükkaya et al. followed 82 ITP patients for five years and found an increased incidence of thromboses in patients that were positive for APLAs. Thirty-one patients were APLA-positive at diagnosis. Cumulative thrombosis-free survival was 39.0% in ITP patients with APLAs and 97.7% in those without [31]. Pierrot-Deseilligny et al. found the lowest incidence of APLAs (25%) in ITP patients and a positive association of APLAs with thrombosis [32].

A 2016 review and meta-analysis suggested that the thrombosis risk in ITP patients may differ depending on the type of APLA. The odds ratio was much higher for lupus anticoagulant antibodies (6.11 95% CI 3.40–10.99) than with anticardiolipin antibodies (2.14 95% CI 1.11–4.12) [35]. In addition, anti-β2GP1 antibodies were found to be present in three of five ITP patients with APLAs that had thrombotic events [33]. Antiphospholipid and multivalent autoantibodies (via FcγRIIA receptors) have also been implicated in platelet activation which can play a role in both thrombosis and platelet destruction [36].

2.2. Patient-related factors

Patient-related risk factors for thromboembolic events in ITP patients include age (over 60 years), male sex, and obesity (body mass index ≥30 kg/m2). Other risk factors are a history of TEEs, the presence of APLAs and hemolysis (for example, Evans syndrome), and comorbidities (such as cancer, hyperlipidemia, diabetes, hypertension, a coronary disease, and a chronic kidney disease) [6].

Machin et al. looked at the risk of thrombosis in ITP patients with underlying diseases or conditions compared to non-ITP patients [37]. The condition showing the highest risk in ITP patients was the post-operative state. There are circumstances, e.g. immobility, which routinely occur in patients post-operatively which increase thrombotic risk. When a patient with ITP gets admitted for surgery unrelated to ITP, they may experience immobility and increased risk of thrombosis must be considered. Other factors like an underlying malignancy, the need for a central venous line, trauma and APS were also clear thrombotic risk factors in ITP patients [37].

2.3. Treatment-related factors

ITP treatments can also play a role in increased thrombotic risk. Kristinsson et al. looked at splenectomy and the risk of thrombosis in over 8,000 American veterans with 27 years of follow up [38]. The condition associated with the largest increase in relative risk of hospitalization or death was sepsis. Heightened awareness of infection risk and a low threshold for treating infections are needed in splenectomized patients. This was demonstrated by the increased risk associated of pneumococcal pneumonia, septicemia, and meningitis that was observed in this study. There was also increased risk of several cardiovascular conditions after splenectomy, i.e. deep venous thrombosis, pulmonary embolism, coronary artery disease, myocardial infarction, and ischemic stroke. By extension, splenectomy may increase the risk of venous and arterial thrombosis in patients with ITP.

Although when comparing across studies, TPO-RAs may have a better response rate (65.7% at one month) [39] than fostamatinib (43% at 24 weeks) [40], TPO-RAs may also increase thrombosis risk. Cooper et al. reviewed data from several studies of TPO-RAs in ITP (randomized, controlled clinical trials and/or long-term rollover studies) [8]. The incidence of thromboembolic events (TEEs) reported in these studies ranged from 0.9–9.4% in studies up to seven months duration and from 2.6–8.9% in studies of 2–8 years duration [41–50].

The effects of TPO-RAs have also been studied in ITP patients with APS. A retrospective cohort study in 11 centers in the French national network for adult ITP, analyzed the efficacy and safety of TPO-RA in systemic lupus erythematosus (SLE) associated ITP [51]. Eighteen patients with SLE-ITP were treated with TPO-RAs. Of these patients, ten (55%) had APLAs, five (27%) showed definite APS.

After a median follow-up of 14.7 months with TPO-RAs, there were four arterial thrombosis events in four patients and two venous thrombosis in a patient without APS. The authors concluded that patients with SLE should be systematically screened for antiphospholipid antibodies (APLAs) before TPO-RA initiation. In patients positive for APLAs, alternative therapy should be discussed (if possible), especially in patients with definite APS or suboptimal adherence to anti-coagulation

Another aspect of ITP treatment and thrombotic risk is the effect of corticosteroid treatment. Corticosteroids are widely recognized as a first-line treatment for newly diagnosed ITP patients. A retrospective study of ITP patients with thrombotic events found that the vast majority (19 of 26) were on steroid therapy at the time of the event [52]. In addition, a study was conducted to assess the relationship between eltrombopag and thrombotic events using a large population extracted from the Food and Drug Administration Adverse Event Reporting System (FAERS) database. This study showed an association between thrombotic events and eltrombopag use. Moreover, concomitant use of eltrombopag and glucocorticoids appeared to result in an even higher incidence of thrombotic events [47].

3. Patient factors affecting treatment selection

An important factor to remember when treating ITP is that this is not a malignancy. This is one of the reasons why ITP could be overlooked or undertreated. Although ITP presents low risk for morbidity and mortality, the consequences of a major bleeding event, such as intracranial hemorrhage, could be fatal [53]. Some patients can have clinical and quality-of-life impacts that can affect their daily routines and in some cases the therapy may cause more problems than the disease.

When choosing therapy for an ITP patient there are several factors to consider: Does the patient actually need treatment? What is the more suitable treatment for this patient? What treatment would patient prefer based on the potential effects on the quality of life at home and at work? After the treatment options are explained to the patient, physicians should consider the patient’s preferences and lifestyle.

As their physician, what do you think is the best treatment? Knowing their comorbidities and their history of bleeding, what is their likelihood of bleeding? The treatment decision should be a two-way dialogue between physician and patient. The need for dialogue is stressed heavily in the new treatment guidelines [39,54]. Physician-patient dialogue enhances compliance with therapy and makes the patients feel more involved in the treatment process.

This last point has been highlighted in the latest international guidelines. The updated American Society of Hematology guidelines [39] and the updated international consensus report on ITP treatment [54] are more patient-centric than previous versions. Guidelines are likely to become even more patient-focused in the future based on patient surveys and studies such as the iWISh data [2] and the positive effects of more communication with patients.

When an ITP patient must be treated, there are several factors that should be considered. Bleeding risk must be assessed since bleeding risk is higher in ITP patients over the age of 60 [55]. Another important risk factor is the presence of comorbidities [55]. If a patient has one or more comorbidities, e.g. autoimmune diseases (e.g. SLE), hypertension or diabetes, these can also increase the risk of bleeding with ITP.

Other adverse effects of ITP treatments should also be considered when selecting appropriate therapy. The most common adverse events associated with fostamatinib in the clinical trials were diarrhea, hypertension, nausea, dizziness, and increases in serum liver enzymes [40]. The adverse effects most commonly found in the eltrombopag clinical trials were bleeding, headache, nasopharyngitis, diarrhea and vomiting [41]. With romiplostim, the most common adverse events were headache, fatigue, epistaxis, arthralgia, and contusion [46]. The potential effects of these adverse effects on each patient’s comorbid conditions should be considered when choosing a therapy. In light of the observations above, the demographics of ITP patients are important in selecting appropriate therapy.

Figure 2(a) shows the results of a retrospective study of ITP incidence in France [56]. Previously, it was believed that ITP most common in females of reproductive age, but now it is well recognized that ITP is more common in older patients [56–58]. This study shows that there is a greater incidence of ITP in females in their reproductive years than in males of the same age, but the incidence for both sexes in these age groups (18–28, 30–39 and 40–49 years) is less than that seen in people over the age of 60. As in the previous study [55]. this shows that over the age of 60 the incidence increases dramatically for both sexes, most common population of patients needing treatment for ITP. There is a peak in incidence in young children as well, but often these patients do not need treatment.

Figure 2. (a) the incidence of ITP in France (2009–2011). white bars = females diagnosed with ITP; black bars = males diagnosed with ITP. Asterisks denote statistically significant differences between the sexes (from moulis et al. [56]). (b) The percentage of ITP cases which were secondary to other conditions with age in France (2009-2011). (From Moulis et al. [56]). (c) Gastrointestinal and central nervous system bleeds in ITP patients by age. White bars = gastrointestinal bleeds; black bars = central nervous system bleeds. There was a statistically significant correlation between gastrointestinal bleed bleeds and age (p = 0.003) and central nervous system bleeds and age (p = 0.02). (From Moulis et al. [56]).

Figure 2(b) shows the percentage of ITP patients who require treatment that have secondary ITP, i.e. ITP as a result of one or more comorbidities [56]. The older a person becomes, the more likely they are to have comorbidities and the incidence of secondary ITP increases. Comorbidities need to be considered when evaluating treatment in especially in older patients as these can increase bleeding risk [55].

Further evidence of increased bleeding risk in older patients was confirmed in this study [56]. The incidence of both central nervous system and gastrointestinal bleeding increases in these patients, especially after age 60 (Figure 2(c)) [56]. In general, ITP patients over 60 have a high risk of bleeding, and consequently require treatment, but often have comorbidities to consider.

This information is summarized in this algorithm for second-line treatment of ITP (Figure 3). In addition to determining the patient’s risk factors for bleeding, risk factors for thrombosis should be assessed as part of the decision for initial treatment. Additional factors that should be considered are the patient’s age, comorbidities, history of arterial or venous thrombosis, and if they are Lupus anticoagulant positive, including APLAs. If these factors are present, then alternative agents to TPO-RAs should be considered. If these factors are not present, then the treatment discussion can involve TPO-RAs, especially the newer agents. These assessments need to be performed on a regular basis to determine if the patient has developed risk factors during treatment and if an alteration in treatment is appropriate.

Figure 3. An algorithm for selection of second-line treatment for ITP.

This treatment algorithm was developed considering an ITP patient who has developed a thrombosis.

If an ITP patient requires anticoagulation, their platelet count is more than 50 × 10⁹/L and they are stable, they can be treated like a patient without ITP as long as there are regular reviews to assess bleeding risk and platelet count [59].

However, if the patient’s platelet count is less than 50 × 10⁹/L, with or without treatment for ITP, (patients with platelet counts of 20–30 × 10⁹/L may not need treatment), anticoagulation treatment should be considered. If the patient has atrial fibrillation and a very high bleeding risk, left atrial appendage closure should be considered, if possible and available. If the patient has indications of thromboembolism, ITP treatment should be intensified and accompanied by anticoagulation. The same course applies if left atrial appendage closure is not possible [59].

This last step in this algorithm is open to discussion. Treatment for VTE should last for at least three months. If it was idiopathic, the dose of direct oral anticoagulants could be decreased. Full-dose anticoagulation should be considered for patients with atrial fibrillation and low bleeding risk.

Antiplatelet medications may also play a role in ITP patients. Similar to the algorithm for anticoagulation, for ITP patients with stable platelet count > 50 × 10⁹/L, standard antiplatelet therapy may be used with surveillance [59]. If the indication for antiplatelet therapy is ischemic stroke and the platelet count is < 50 × 10⁹/L, the ITP treatment can be intensified, and single-agent antiplatelet therapy may be added. If the indication is acute coronary syndrome, stents should be considered which requires a minimum period of dual antiplatelet therapy. Patients with extreme bleeding risk should be considered for coronary artery bypass grafting. In this circumstance, ITP treatment should be intensified, and a single antiplatelet medication added.

4. Low incidence of thromboembolic events under fostamatinib treatment

With ITP, there’s a higher risk of venous thrombosis but risk of arterial thrombosis is not significantly increased [6,60]. However, as noted above, age, splenectomy, personal and treatment risk factors may put some ITP patients at particularly high risk of venous thrombosis, three to four times higher than a normal individual. ITP patients treated with TPO-RAs have shown indirect evidence of a much higher risk of both arterial and venous thrombosis – up to three or four times than that reported in untreated patients [60–62].

Recent publications on the risk of thrombosis in patients with ITP and the procoagulant profile of patients with ITP have shown that that TEEs in ITP patients may be reduced with some new therapies (e.g. fostamatinib) compared to older therapies. Fostamatinib is indicated for the treatment of chronic immune thrombocytopenia in adult patients who are refractory to other treatments. It was approved by the FDA in 2018 [63] and by the EMA in 2020 [64]. Fostamatinib inhibits spleen tyrosine kinase (SYK) which plays a key role in antibody-mediated phagocytosis. SYK is required for different Fc receptors signaling in macrophages, which is integral to phagocytosis of antibody-coated platelets (Figure 4) [65].

Figure 4. Fostamatinib mechanism of action. inhibition of spleen tyrosine kinase (SYK) by fostamatinib plays a key role in antibody-mediated phagocytosis of platelets.

The clinical trial program for fostamatinib included two double-blind placebo-controlled phase 3 trials (FIT-1 and FIT-2) [40] and one open label long-term study (over 5 years: FIT-3) [8,66,67]. Recently, Cooper and the FIT clinical study investigators published a paper in which they reviewed and summarized the incidence of TEEs in ITP patients treated with fostamatinib in a registration and long-term extension studies [8].

Cooper et al. conducted a standardized Medical Dictionary for Regulatory Activities (MedDRA) query search for TEEs in the safety database for fostamatinib trials, which included 146 adult patients in the phase 3 studies for ITP (FIT-1, FIT-2, and FIT-3) [8]. All patients enrolled in these clinical trials had persistent or chronic ITP and had failed at least one treatment prior to entry into the study. The patients were treated for up to five years (mean treatment length: 19 months, range 1–62 months) with a total fostamatinib exposure of 229 patient-years.

In this analysis [8], most of the patients (87%) had one or more risk factors for TEEs and more than half (58%) had two or more risk factors for TEEs. The median number of risk factors was 2 with the range was from 0 to 7 risk factors. Only 25% of the patients were ≥65 years old, 40% were male and 30% had a body mass index ≥ 30. The comorbidities for these patients included diabetes (10%), cancer (5%), cardiovascular disease excluding hypertension (25%) and hypertension (35%). Most of the patients (94%) had been treated with corticosteroids prior to entering the clinical trial. Almost half had received TPO-RAs (47%) and 35% had undergone a splenectomy.

Out of these 146 patients, one developed a TEE [8]. The incidence of TEEs in these fostamatinib clinical trial patients was very low (0.7%). The patient who developed a TEE was a 61-year-old female with a ten-year history of ITP. She suffered a mild transient ischemic attack which was deemed unrelated to the study drug by the investigator. The transient ischemic attack resolved following treatment with intravenous fluids and increase in antihypertensive medication. The patient had received methylprednisolone and splenectomy as treatment for her ITP. She had several risk factors for TEEs, such as age, obesity and comorbidities as atherosclerosis, hypertension, and chronic obstructive pulmonary disease.

In the same paper, Cooper et al. reviewed the literature on TEEs in TPO-RA trials [8]. Eleven publications were identified that reported results of randomized, controlled, registration trials and associated long-term extension studies of TPO-RAs in ITP [41–50]. The publications were reviewed for the incidence of TEEs with TPO-RAs. The incidence in the short-term studies ranged from 0–9.4%. The highest incidence was in a clinical trial with avatrompopag [45]. The incidence of thromboembolic events in the long-term studies of ITP treated with TPO-RAs was generally higher ranging from 2.6% (avatrombopag) [45] to 8.9% (romiplostim) [50]. By comparison, for fostamatinib in studies up to 5 years duration, the incidence was 0.7% [8]. This low incidence of TEEs with fostamatinib was seen even though the phase III and extension study included heavily pre-treated patients with long standing ITP and many patients at high risk for TEEs.

A small (n = 34) placebo-controlled, double-blind, parallel-group, phase 3 fostamatinib study conducted in Japan showed that although the overall response rate was relatively low (45%, n = 22) [68]. Most of patients that responded to fostamatinib had a sustained response out to 24 weeks of treatment (36%, n = 22). In this study, there were no thromboembolic events in either the fostamatinib or placebo treatment groups.

The mechanism of action of fostamatinib may help explain why the incidence of TEEs is so low. Fostamatinib inhibits SYK. The role of SYK in thrombosis is shown in Figure 5 [8]. Platelet interaction with leucocytes and endothelial cells induces bidirectional signals that contribute to both proinflammatory and prothrombotic outcomes and can mediate initiation and progression of TEEs. Experimental evidence suggests that SYK is a key signaling molecule positioned at the nexus of both the inflammatory and thrombotic responses.

Figure 5. Pro-inflammatory and pro-coagulation pathways affected by spleen tyrosine kinase (SYK). Fcγ = fragment, crystallizable region; IgG = immunoglobulin G; P = phosphoryl group (from Cooper et al. [8]).

SYK controls signaling downstream of FcyRIIA gamma globulin receptor, glycoprotein VI (GPVI) receptor and C-type lectin-like II receptor CLEC-2 involved in thrombus formation. By inhibiting SYK, fostamatinib blocks all three pathways responsible for pathogenic platelet aggregation without affecting other platelet receptors involved in maintaining hemostasis, such as the ADP receptors, avoiding undue bleeding [69,70].

Here is an illustrative case from Hospital La Paz (Madrid, Spain) where fostamatinib was successfully used as a treatment in a patient that had failed numerous treatments and had suffered a deep venous thrombosis while on a TPO-RA. A 64-year-old woman diagnosed with ITP in December 2018. Previous treatments for ITP included corticosteroids, immunoglobulin, eltrombopag, romiplostim, rituximab, and mycophenolate. Since June 2021, she was refractory to all treatments with a platelet count of 21,000/µL. At that time, she was treated with weekly immunoglobulin infusions. Regarding adverse events of her previous treatments, she had suffered a vertebral fracture during treatment with corticosteroids. She also suffered a deep venous thrombosis with a pulmonary embolism during treatment with eltrombopag in 2019 which required anticoagulant treatment for three months. Currently, she is being treated with fostamatinib and her platelets have been above 50,000/µL. Since she changed to fostamatinib, her treatments with immunoglobulin can be given less frequently (every 3–6 weeks). This case demonstrates the role of fostamatinib treatment in a refractory ITP patient with a history of a thromboembolic event while on another treatment.

Recently there was a publication on the potential of fostamatinib to treat patients with COVID-19. The effects of fostamatinib in this setting may shed some light on the mechanisms of fostamatinib in reducing the propensity for thromboembolic events in patients with ITP. Fostamatinib reduced the cytokine storm, platelet activation, NETosis, and lung injury. R406, an active metabolite of fostamatinib, was able to block NETosis induced by plasma from COVID-19 patients [71]. A double-blind, placebo-controlled clinical trial in hospitalized adult patients with COVID-19 showed that addition of fostamatinib to the standard of care had positive effects on clinical outcomes compared to placebo [72]. Analysis of samples from this study suggested that fostamatinib inhibited production of pro-inflammatory mediators associated with severe COVID-19 [73]. There are other ongoing clinical trials in patients with COVID-19 treated with fostamatinib [74–76].

5. Conclusions

ITP is not only an autoimmune bleeding disorder, but also a thrombotic disorder. Thrombotic events are increased in patients with ITP due to: ITP-related factors (immature platelets, microparticles and possibly pro-inflammatory cytokines), treatment-related factors (splenectomy, and treatment with intravenous immunoglobulin (IVIG), corticosteroids or TPO-RA), and patient-related factors [3,59,77].

Fostamatinib is an oral SYK inhibitor that reduces platelet destruction by blocking SYK signaling in macrophages. It is indicated for chronic ITP adult patients who are refractory to other treatments. The pivotal phase III studies (FIT-1 and FIT-2) and open-label extension (FIT-3) study showed efficacy and safety with fostamatinib treatment for up to five years. Moreover, these studies showed a very low incidence of TEEs, only one patient developed a TEE in the clinical trials. This compares to up to 9.4% of ITP patients with TEEs during TPO-RA treatment. SYK inhibition with fostamatinib does not appear to increase the risk of TEEs for patients with ITP and may decrease the risk of TEEs even among patients with prothrombotic risk factors. Ongoing studies indicated also that fostamatinib may be useful in patients with severe COVID-19.

6. Expert opinion

The pathophysiology of ITP is complex and not completely understood. ITP is a collection of disorders that are grouped under the umbrella term, primary ITP. Current evidence indicates that disorders of B cells, T cells, complement and other factors contribute to the pathogenesis of ITP. Determining the type of defect underlying a patient’s ITP would be clinically useful in determining an individualized course of treatment for that patient. Additional research into the pathophysiology of ITP and the underlying cellular defects could provide pathways for the creation of more specific targeted therapies that could selectively address an individual patient’s ITP.

In addition, this research could provide biomarkers to demonstrate the predominant underlying mechanism in each patient to guide selection of targeted therapies with reduced toxicity. Likewise, therapeutic strategies could be developed to address both increasing platelet production on the one hand and decreasing platelet destruction on the other which could effectively treat patients that do not respond to a single agent.

While commonly thought of as a bleeding disorder, paradoxically, ITP patients have been found to be at increased risk for TEEs [3–5]. These risks can be related to the disease itself, the treatment selected and the patient’s underlying co-existing conditions. Greater understanding of ITP pathophysiology could help ascertain the disease components related to TEEs and whether differences in the underlying defect responsible for the patient’s ITP could increase or decrease thromboembolic risk. Determination of the thrombotic risk for an individual patient is important for ITP management and this risk is not currently being fully explored for each patient. A standardized system for stratifying thromboembolic risk in patients undergoing treatment for ITP would be advantageous in optimizing and individualizing therapy. Several current treatments predispose patients to TEEs (TPO-RAs, corticosteroids, IVIG and splenectomy) especially those patients with underlying conditions that already put them at increased risk. It is useful to have therapies available like fostamatinib, that can help minimize and possibly reduce the risk of TEEs.

Physicians treating ITP need to think carefully about type of therapy they choose taking into consideration the patient’s characteristics such as comorbidities and treatment preferences, such as route and frequency of administration. Individualizing the treatment by choosing the optimal therapy for a specific patient is often a difficult decision. This choice is complicated by incomplete understanding of ITP pathophysiology and the limited number of treatment options.

When individualizing therapy for an ITP patient, the patient’s bleeding risk, co-morbidities, thromboembolic risk, and effects of treatment on the patient’s quality of life must all be taken into consideration. Quality of life is a key determinant for many patients with ITP. Treatments may have a negative effect on quality of life especially if they precipitate TEEs or require frequent visits to the clinic. This quality of life component should be discussed with patients when considering treatments, for example, when comparing TPO-RAs to SYK inhibition and comparing oral or self-administered subcutaneous administration to clinic-administered intravenous administration.

3The risks for TEEs in ITP can be disease-related (immature platelets and inflammatory cytokines), patient-related (e.g. history of TEE, obesity, immobility), or treatment-related (e.g. corticosteroids and TPO-RAs). Fostamatinib, an oral SYK inhibitor that is a newer treatment for ITP, does not increase the incidence of TEEs in ITP patients in general and may decrease them in ITP patients with comorbid conditions. A greater understanding of the pathophysiologic mechanisms of ITP has allowed the development of improved treatments for ITP and will continue to do so in the future.

Article highlights

• Immune thrombocytopenia (ITP), usually considered a bleeding disorder can also carry increased risk of thromboembolism

• The relative risk of thromboembolism in ITP depends on patient-related and disease-related factors and can be influenced by the choice of ITP treatment.

• Greater understanding of the pathophysiology of ITP has led to the development of new agents, e.g. fostamatinib and can lead to greater personalization of treatment for each patient.

• Agents such as fostamatinib can be a therapeutic alternative for patients in which thromboembolic risk must be minimized.

Declaration of interest

D Provan acknowledges the following competing interests: research support from Amgen, and Novartis; honoraria from Amgen, Novartis, SOBI, UCB, and argenx; consultancies with UCB, MedImmune, Ono, SOBI, argenx, and Takeda. J Thachil has received honoraria from Amgen, Grifols and Novartis. MT Alvarez-Roma has served as a speaker or an advisory board member with Novartis, Bayer, Takeda, Roche, Pfizer, Octapharma, Amgen, Novartis, CSL Behring and Sobi. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or material discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Previous presentations

This material was previously presented in a symposium at The International Society on Thrombosis and Haemostasis 2022 Congress, London, 9 July 2022 (Sponsored by Grifols).

Acknowledgments

Michael K. James, PhD, CMPP (Grifols) is acknowledged for medical writing assistance and Roser Mir, PhD and Jordi Bozzo, PhD, CMPP for editorial assistance.

Additional information

Funding

The preparation and submission of this manuscript were funded by Grifols.