Edited by Elisa Lucchini, Trieste, Italy
(See disclaimer at the end)
The 64th ASH annual meeting left us with a huge amount of news and food for thoughts, concerning several aspects of ITP, from diagnosis to treatment, from pathogenesis to predictive factors of response. For instance, can we confidently reassure a patient with an isolated, mild thrombocytopenia? May antiplatelet antibodies not just enhance platelet clearance, but also, by their binding, affect platelet function, thus explaining the different clinical phenotypes? Growing evidence supports the role of T cells in the pathogenesis of ITP, by participating to bone marrow dysfunction and mediating the lysis of circulating platelets. However, the role of antiplatelet antibodies remains unquestioned, as testified by the results of the phase III trial with Efgartigimod, a specific FcRn inhibitor. Efforts are being made in seeking factors predictive of response to treatment, both clinical and biologic. Is ITP just a matter of bleeding risk, or should we begin to consider it as a more “systemic” disease? You can find the answers to some of these questions in the next lines and if not, online at https://www.hematology.org/meetings/annual-meeting.
Epidemiology. A retrospective cohort study evaluated the long-term outcome of 91 patients with persistent isolated mild thrombocytopenia (PIMT), i.e. patients with a platelet count between 100 and 149×109/L. Compared to subjects with a normal platelet count, PIMT patients had a higher risk of developing after several years some hematological diseases (30.8% vs 1.9%): in most cases ITP, but also myelodysplastic syndrome (MDS), myeloproliferative neoplasms (MPN) and non-Hodgkin lymphoma, as well as systemic autoimmune diseases (Ayad et al. abs n° 19). These findings, if confirmed by prospective studies, may question the threshold of “low” platelet count which deserves further evaluation, and whether those subjects need a closer follow-up. For patients with a platelet count less then 100×10^9/L, the differential diagnosis is usually straightforward, however, in some cases of mild thrombocytopenia without a clear evidence of bone marrow dysplasia, the etiology may be tough. In such cases, the use of additional parameters may help, such as platelet indices and thrombopoietin (TPO) serum levels: compare to MDS, patients with ITP have usually higher immature platelet fraction and lower TPO serum levels. (Lucchini et al. abs n° 3765).
In the diagnostic work-up, it is important not to forget about the correlation of thrombocytopenia and Helicobacter pylori (HP) infection, especially in endemic areas. Platelet count of patients who tested positive for HP by urea breath test and serology do benefit from HP eradication (Siritham et al. abs n°917).
Biology. Bone marrow (BM) in ITP participates to the pathogenesis of the disease, probably as a target of the impaired immune system. BM of 40 ITP patients in 3 different stages of the disease (newly diagnosed treatment naïve, persistent-phase treatment-naïve, and relapsed patients) and 10 healthy controls was analyzed by Cytometry by Time of Flight (CyTOF). Compared with healthy controls, ITP patients had a higher proportion of monocytes and myeloid DCs. The proportion of CD4+ naïve and CD8+ naïve T cells was decreased in BM of ITP patients, while the proportion of CD8+ effector T cells was increased, with a reduced expression of the coinhibitory molecule CTLA-4, and an enhanced expression of the costimulatory molecules CD28 and ICOS (Liu et al. abs n° 914). These findings show that in the bone marrow of ITP patients, T-cells are active and armed, but against what? May they be armed against platelet antigens? In the peripheral blood of ITP patients, terminally differentiated effector memory CD8+ T cells (TEMRA) are oligoclonally expanded and functionally active. The clonally expanded TEMRA CD8+ T cells are not virus-driven, persist over the years and correlate with disease activity. When co-cultured with platelets, CD8+ T cells bind-to and directly cause platelet activation and death (Malik et al. abs n° 915).
So, would a treatment against T lymphocytes be effective? The mTOR inhibitor, Rapamycin, was tested both in vitro and in a mouse model of ITP, showing an inhibition of CD8+ T cell activity (reduced levels of granzyme B and perforin), and an increase in platelet counts. (Ni et al. abs n° 3764). The role of T cells in the pathogenesis of ITP can also be inferred from the response achieved to B-cell directed treatments. B cell activation markers (BAFF, APRIL) resulted higher in patients who responded to splenectomy, while patients who did not had higher T and NK cell functionality and cytotoxicity (Li at al. abs n° 916).
Has Eltrombopag an immunomodulatory effect on T cells? Eltrombopag in vitro suppresses the proliferation of T cells, probably due to its iron-chelating action. In fact, Deferoxamine induces the same cell cycle arrest to G1 phase, but in a less potent way. This effect is not seen with Romiplostim or rhTPO, that do not exhibit iron-chelating property. Furthermore, the addition of iron to Eltrombopag-treated T-cells restores cell proliferation. In vivo, patients who respond to Eltrombopag, display a reduction in the frequency of CD8+ TEMRA cells, suggesting an immunomodulatory effect of the drug (Tan et al. abs n° 2451).
Risk of bleeding and thrombosis in ITP.
ITP is mainly a bleeding disorder, but the thrombotic risk associated with the disease is not negligible, especially for some patients undergoing specific treatments. Venous Thromboembolism (VTE) is a non-negligible complication in patients with chronic ITP affecting 1.4 per 100 person-years. Low platelet count doesn’t seem to protect from VTE, but in the majority of cases patients had concurrent risk factors for thromboembolism. Arterial thromboembolism is also a non-infrequent complication (2 per 100 person-years), and often these patients are sub-optimally managed, likely due to the concern of the physician in treating them. The use of anticoagulation is therefore fundamental to prevent thrombosis in such patients: ITP management must aim to obtain the safest platelet count which allows the use of anticoagulation and/or antiplatelet agents (Wang et al. abs n° 22, 24). To support that, a French report showed that, in such patients, the use of one antiplatelet agent (acetylsalicylic acid) does not increase the risk of bleeding, compared to using two (Ollier et al. abs n° 23).
Bleeding symptoms are highly variable in ITP patients, and patients with the same platelet count may have different bleeding diathesis. Bleeding diathesis may be related to the epitope-specificity of the anti-platelet autoantibody: each epitope-specific antibody can differently alter platelet physiology and functionality. Anti-GPIIb/IIIa antibodies that bind to either the head or tail region of aIIb increased the expression of phosphatidylserine, the opposite effect is obtained with autoantibodies that bind to bIIIa. Anti-GPIIb/IIIa with other epitope specificity alter platelet spreading area and adhesion (Shaver et al. abs n°913). These findings show that, besides enhancing platelet clearance, anti-platelet antibodies may also impair platelet functionality. Bleeding can also manifest in a more subtle way, and also microbleeds may be dangerous and lead to clinical consequences, for example thinking about cerebral microbleeds (CMB) in children. Cerebral MRI found CMB in 5/38 children with ITP, all chronic; the presence of CMB correlated with higher bleeding score, while there was no correlation with disease duration, cognitive impairment, fatigue and health-related quality of life (Hart et al. abs n°21).
Is ITP just a matter of hemostasis? ITP doesn’t seem to only affect platelet count: fatigue, memory and concentration issues are symptoms patients often complain of. Is there a cognitive impairment in ITP? Episodic memory was the most damaged when evaluated with the CANTAB tool, and risk factors for such impairment were higher organ bleeding score, older age at diagnosis and at time of cognitive test, and a major number of ITP treatments received (Vladescu et al. abs n° 2447). Patients enrolled in the phase 1/2 Rilzabrutinib trial, who underwent the Cogstate Brief Battery test, showed a cognitive impairment both for psychomotor/attention, and for learning/memory skills (Kuter et al. abs n° 3773). Moreover, ITP patients are more likely to suffer from psychiatric conditions, such as depression, anxiety, obsessive-compulsive disorder and fatigue if compared to the general population (Mannering et al. abs n° 3768).
The efficacy and safety of intravenous Efgartigimod (a specific FcRn inhibitor) in adult patients with persistent or chronic ITP was evaluated in the phase III, placebo controlled ADVANCE trial. 131 heavily pretreated participants were randomized 2:1 to receive intravenously Efgartigimod or Placebo for 12 weeks. Efgartigimod showed superiority in terms of sustained response (platelet ≥50×10^9/L in ≥4 of 6 visits between weeks 19 and 24, without rescue therapy), which was achieved in 22% of patients (vs 5% in the placebo group), response rate according to IWG criteria (51% vs 20%), and onset of platelet response, without raising concerns in terms of toxicity (Broome et al. abs n°3, selected for Plenary Session presentation).
Is discontinuation of thrombopoietin receptor-agonists feasible in the real-life setting? A retrospective Spanish experience enrolled 133 patients, 75 treated with Eltrombopag and 13 with Romiplostim. 4/13 patients could discontinue Romiplostim, without need of further therapy, while 25/75 could discontinue Eltrombopag and maintain the remission. Globally, 33% of patients could successfully discontinue the TPO-RA. They also described a cohort of patients who switched from one TPO-RA to another, and treatment-free responses were observed also in this group (Ramirez Lopez et al. abs n°1127). A report from the German prospective study (RISA) of patients treated with Eltrombopag, confirmed the efficacy and safety of the drug. Notably, treatment was discontinued in 128/313 (42%) of patients, in most cases (43%) due to lack of efficacy, but 12% of patients discontinued the drug due to complete response (Meyer et al. abs n° 1130).
The efficacy of the combination of eltrombopag and rituximab over eltrombopag alone was reported in a single center, retrospective experience: 69 patients were treated in the combination arm and 53 in the eltrombopag group. Patients treated in the combination arm had higher rate of response at 12 and 24 weeks (72% vs 54% and 70% vs 50%) and also higher rates of treatment-free remission after eltrombopag discontinuation (25% vs 9%) (Yanmei et al. abs n° 2453). Eltrombopag may also be combined with other agents: an ongoing Chinese randomized phase II study is testing the efficacy of Eltrombopag plus Diacerein (which has shown to sensitize Megakaryocytes to TPO-RAs) versus Eltrombopag alone. Patients in the combination arm had a higher rate of initial responses, duration of response and less bleeding events (Sun et al. abs n° 3767).
Earlier use of Romiplostim was explored in a retrospective study, enrolling 78 patients with ITP diagnosis less than 12 months. Patients included in this study were treated either with Romiplostim or with immunosuppressive drugs (steroids, IVIg, Rituximab, Azathioprine, Cyclosporine, Danazol); those belonging to the first group reached a slightly higher median platelet count, had less bleeding events and could discontinue steroids earlier (McDonald et al. abs n° 1128). The growing availability of new drugs, targeting different mechanisms of ITP pathogenesis, and the exciting perspective of combining them, will deeply change ITP management in the next years. However, the issue of refractory ITP remains a challenging scenario, even though limited to a small proportion of patients. Data from the CARMEN-France registry, including all cases of incident ITP in adults since 2013, detected 32 (3%) patients refractory to both eltrombopag and romiplostim, and 24 (2.2%) refractory to both TPO-RAs and Rituximab. This group of patients is at high risk of bleeding (experienced by 100% of them), infection (25-29%), venous (8-9%) and arterial (3-4%) thrombosis, and hospitalization. Despite infrequent, multi-refractory ITP is a high-burden disease, with a non-negligible risk of infections (Moulis et al. abs n° 2449).
Factors predictive of response to treatment. A study of 275 patients with primary ITP explored the impact of body mass index (BMI) on ITP outcome. Patients were divided into 4 groups based on the degree of the overweight. Patients with higher BMI were more likely to require treatment, were more steroids dependent and needed more lines of therapy. No difference was found in terms of the level of response, time to response, refractoriness, bleeding and thrombotic complications (Xiao et al. abs n° 20). By evaluating lymphocyte subsets, cytokines, antiplatelet antibodies, lymphocytes subpopulations and some clinical parameters, a predictive scoring model of response to TPO-RAs versus corticosteroids, was proposed. Baseline levels of TGF-beta1 were positively correlated with the efficacy of TPO-RA, while baseline TPO levels were not. IP-10 (interferon-gamma induced protein 10) levels were good to predict corticosteroid efficacy. The predictive score attributed 1 point to each parameter, and based on the sum, the likelihood to respond to one treatment was determined (Xu et al. abs n° 1133).
ITP, vaccination and COVID-19. A retrospective study of 442 ITP patients found that 18/60 newly diagnosed had COVID-19-related-ITP: in 5 patients related to the infection and in 13 patients related to COVID-19 vaccine. Vax-ITP patients were older and responded less well to therapy. 69/382 patients had a relapse of ITP, in 52% related either to COVID-19 infection (1 case) or vaccination (35 cases) (Auteri et al. abs n° 3761). Another report did not find a higher ITP relapse rate following SARS-CoV-2 vaccination, compared to ITP relapse rate before COVID-19 pandemic (Stefani et al. abs n° 2444). The rate of ITP relapse during COVID-19 infection or vaccination appears low also in children and young adults, and usually the decrease in platelet counts is transient (MacWhirter et al. abs n° 1138).
DISCLAIMER: This selection of abstracts should not be intended as an appreciation of their relevance or as an encouragement to use their content to support clinical decisions.