Acquired Glanzmann's Thrombasthenia with IgG and IgA Against Activated Αiibβ3

Glanzmann's thrombasthenia (GT) is a rare haemostatic disorder characterised by an impaired function of integrin αIIbβ3. It is usually caused by mutations in ITGA2B or ITGB3 genes that encode this platelet receptor (congenital GT). More rarely, it can be triggered by autoantibodies (acquired GT). Acquired GT is often associated with the presence of another autoimmune disease or haematological malignant disorder, or is drug induced.1 Only a few cases of acquired GT have been described, with this disorder being reported secondary to conditions such as immune thrombocytopaenia (ITP) particularly following splenectomy,2-4 Hodgkin's or non-Hodgkin's lymphomas,4-6 systemic lupus erythematosus or antiphospholipid syndrome.7 Therapeutic strategies include plasma exchange, immunosuppressants, splenectomy, intravenous immunoglobulin, recombinant factor VII or rituximab, alongside the treatment of the underlying disease.5, 8-10 The clinical profile of acquired GT includes spontaneous onset of very extensive mucocutaneous bleeding, gum bleeding, ecchymoses, epistaxis, with or without more severe haemorrhagic complications, including gastrointestinal bleeding or excessive menorrhagia. These symptoms are thought to be caused by antibodies (IgG) against αIIbβ3 causing either impaired function6 or potential internalisation of this integrin.2 Immune-mediated disorders are probably the most common paraneoplastic syndromes in lymphomas.11 Although rare, acquired GT has been described with lymphoproliferative diseases and autoimmune conditions.2-7 We present the case of an 84-year-old male suffering from small lymphocytic lymphoma (SLL), who developed acquired GT. To confirm the presence of anti-platelet autoantibodies and identify their target, we developed two modified assays, using flow cytometry and ELISA. The patient presented with a 6-week history of extensive spontaneous bruising over arms, legs and gluteal region. He had no history of increased bleeding tendencies and no family history of excessive bleeding. Six years prior to presentation, he was diagnosed with extranodal marginal zone lymphoma and received six cycles of rituximab–bendamustine to complete remission. Biopsy performed 4 years later was consistent with SLL/chronic lymphocytic leukaemia (CLL), for which he was monitored on a ‘watch-and-wait’ approach. Autoimmune complications in CLL are frequent,12 but to our knowledge, this is the first case described of acquired GT in a CLL patient. On presentation, bloods showed normal platelet count and no clotting abnormalities apart from mildly prolonged partial thromboplastin time (Figure S1A). At this point, his white cell count was elevated. Aggregometry analysis revealed a severe defect in platelet aggregation with ADP, collagen, arachidonic acid, as well as ristocetin (Figure 1). A healthy control was run alongside and presented normal aggregation. The platelet function analyser (PFA-200) also revealed severely prolonged bleeding time (>300 s to ADP) in this patient. Platelet glycoprotein levels were normal. There was no defect in factor VIII and VWF antigen levels (Figure S1C). At this point, the diagnosis of acquired GT was considered. Since there was also, surprisingly, a defect in aggregation with ristocetin at the time of admission, it was initially suspected that the patient might also have autoantibodies against GPIb (acquired Bernard Soulier syndrome); however, there was no defect in platelets binding to VWF when a microfluidic assay was performed as previously described13, 14 (Figure S1D). The patient started treatment with six doses of rituximab and high-dose steroids tapered over 3 months, in addition to twice-daily platelet transfusions for the first 3 weeks. Treatment for the underlying SLL/CLL was initiated with Venetoclax with a rapid dose escalation. During the first 2 weeks of treatment, the patient developed minor gum bleeding, right arm and left leg hematoma followed by ecchymoses. After two doses of rituximab, aggregometry results considerably improved (Figure 1). The patient had no further bleeding or bruising and continued to improve clinically over the next few weeks. As reported in other acquired GT cases,4, 15 platelet counts remained normal throughout clinical follow-up. Following treatment initiation, citrated plasma was stored longitudinally and tested via modified flow cytometry and ELISA to confirm the presence of autoantibodies and investigate their target. Citrate-anticoagulated blood was collected through venepuncture using a 21-gauge needle from the patient and healthy controls, following informed consent. To investigate whether the patient had anti-platelet autoantibodies, we isolated platelets from healthy controls (n = 8) as previously described14 and incubated these with either control or patient plasma, in the presence/absence of platelet inhibitors (apyrase and prostaglandin E1) or collagen. Flow cytometry results clearly indicated that the patient presented both IgG and IgA against platelets. This was reflected by the increased percentage of platelets positive for IgG and/or IgA (Figure 2A,B). The level of detection was highest when platelets were gated using the CD42b marker, suggesting a competition between patient autoantibodies and CD41/CD61, which stain for integrin αIIbβ3. Importantly, IgG and IgA were only detected on the surface of activated platelets (isolated in the absence of inhibitors with/without stimulation with collagen) and not in their resting state (isolated in the presence of inhibitors), suggesting that the autoantibodies target the active conformation of αIIbβ3. This finding is supported by the patient's normal platelet count throughout his bleeding symptoms. Should autoantibodies bind to inactive αIIbβ3 (i.e. resting platelets), these could promote their clearance, leading to thrombocytopenia, similar to ITP. To confirm these findings, an ELISA was established, in which plates were coated with purified human αIIbβ3 activated by LIBS2,14 prior to being incubated with patient's or healthy control plasma (n = 11). As reflected in Figure 2C, the patient had increased levels of both IgG and IgA binding to activated αIIbβ3 when compared to healthy control plasma. Of note, the results might be underestimated by the presence of fibrinogen in patient/control plasma, which may compete with the autoantibodies in binding activated αIIbβ3. Serum samples were not available at the time. Longitudinal analysis revealed a decrease in the antibody titres over time, in line with the other results presented in this manuscript and with the symptom resolution. Together, these data confirm a new case of acquired GT disorder, first one reported following SLL/CLL. We used two laboratory approaches that revealed the patient presented not only IgG, but also IgA against αIIbβ3 in its active conformation. To our knowledge, this is the first report to evaluate the presence of IgA in this condition, thus providing further insight into the pathogenesis of this rare disorder and how it can be approached therapeutically. Adela Constantinescu-Bercu designed the research, performed the experiments, wrote and reviewed the manuscript. Sabina McCann performed the experiments and wrote the manuscript. Amjad Hmaid collected clinical data, wrote and reviewed the manuscript. Rens de Groot designed the research and reviewed the manuscript. Deepak Singh performed the experiments and reviewed the manuscript. Satyen H. Gohil and Mari Thomas provided clinical data and reviewed the manuscript. Marie Scully designed the research and reviewed the manuscript. John-Paul Westwood provided clinical data and reviewed the manuscript. The authors declare no conflicts of interest. The patient has confirmed consent for publication of the case. Data S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.