Targeted treatment of immune thrombocytopenia in CTLA‐4 insufficiency: a case report

Immune thrombocytopenia (ITP) is characterized by thrombocytopenia (peripheral blood platelet count < 100 × 109/l) which can be primary (without underlying causes) or secondary.1 Secondary ITP can be associated with infections, drugs, lymphoproliferative disorders or autoimmune diseases. The pathophysiology of ITP is complex and involves multiple arms of the immune system, including dysregulation of B cells, helper T cells, and cytotoxic T cells. Accumulating evidence also points towards loss of immune tolerance from downregulation of regulatory T (Treg) cells.2 ITP treatment in the front-line setting often includes corticosteroids and intravenous immunoglobulin (IVIG). Refractory ITP is treated with thrombopoietin receptor agonists (TPO-RA), rituximab, or splenectomy. Third-line agents are protean and lack large randomized trials.3, 4 A 22-year-old male presented to our institution with petechiae, oral purpura, and blood-filled acneiform lesions. He had a known diagnosis of ITP since age 18, requiring treatment at diagnosis and at age 20. Both times, he had a robust response to treatment, receiving IVIG plus oral corticosteroids with the first episode and corticosteroid monotherapy with the second. His comorbidities included common variable immune deficiency based on a history of hypogammaglobulinaemia, impaired pneumococcal polysaccharide antibody response and recurrent infections. He also had a history of hypogonadotropic hypogonadism and chronic spontaneous urticaria. He was receiving subcutaneous testosterone enanthate and monthly IVIG replacement, and had not had any change to his medications prior to the most recent presentation. The full blood count at admission showed a platelet count of 1 × 109/l (Fig 1). The rest of the haematological, renal, and hepatic profiles were unremarkable. Virology screening, including hepatitis C, human immunodeficiency virus, Epstein–Barr virus, and cytomegalovirus were negative. On the seventh day of admission, he underwent a computed tomography scan of the chest, abdomen, and pelvis. There was no lymphadenopathy, though the spleen size was near the upper limit of normal at 13 cm. He was initially treated with IVIG at 1 g/kg daily for two days with dexamethasone 40 mg daily for four days. At the end of dexamethasone treatment, his platelet count remained at 1 × 109/l. He developed gross haematuria with a new Escherichia coli urinary tract infection, worsening epistaxis and bleeding oral purpura. Hence, he received antibiotics, tranexamic acid mouthwash, and started methylprednisolone at 1 mg/kg daily on the sixth day of admission. Despite the second course of corticosteroids, he continued to show severe thrombocytopenia. He had a significant family history of autoimmune cytopenia. His sister had a history of ITP, warm autoimmune haemolytic anaemia, autoimmune neutropenia, and Addison’s disease. His maternal uncle and grandfather both had a history of ITP requiring splenectomies. The patient had whole-exome sequencing as part of his paediatric endocrinopathy workup in 2016. This analysis had revealed a cytotoxic T lymphocyte-associated antigen-4 (CTLA4) missense variant (CTLA4 c.160G>A, p.(Ala54Thr)) of uncertain significance at that time. This variant has not been published in population databases (gnomAD).5 It was subsequently published in a case report of one patient with features of CTLA-4 insufficiency.6 The functional significance of the variant was then confirmed in another study where CTLA-4 protein expression and CTLA-4-mediated CD80 transendocytosis were both decreased.7 Sequencing of our patient’s affected sister subsequently revealed the same variant in CTLA4. Immunophenotyping of our patient further supported a diagnosis of CTLA-4 insufficiency, demonstrating absolute CD3+ T cell lymphopenia with normal B and NK (natural killer) cells. T cell phenotyping showed expanded PD1+ T cells. The CD4+PD1+ T cells were severely elevated, suggesting expansion of T follicular helper (TFH) cells (Fig 2A). B cell phenotyping showed decreased switched and mildly increased unswitched memory B cells with expansion of autoreactive CD21lowCD38low B cells (Fig 2B). Hence, sirolimus was initiated on the nineth day of admission. He received a loading dose of 6 mg, followed by 2.5 mg daily. On the 10th, 11th, and 12th day of admission, his platelets rose to 5, 31, and 46 × 109/l. At this point, his bleeding had ceased and he was discharged. CTLA-4 is a transmembrane T cell inhibitory receptor expressed on activated T cells and constitutively on regulatory T cells. It inhibits T cells by competing with CD28 for binding of CD80/CD86 and inhibiting the costimulatory signal downstream.8 Haploinsufficiency of CTLA-4 can arise from heterozygous germline pathogenic variants, leading to loss of T lymphocyte self-tolerance. Patients with CTLA-4 haploinsufficiency can manifest with an immune dysregulation disorder, sometimes called CTLA-4 haploinsufficiency with autoimmune infiltration (CHAI).8 A large study of CTLA-4-insufficient patients describes 133 individuals who variably presented with autoimmune cytopenia, hypogammaglobulinaemia, recurrent infections, organomegaly, and sequelae of organ infiltration.9 In this cohort, 46% had ITP, 84% had hypogammaglobulinaemia, and 33% had endocrine involvement. Targeted treatment and associated outcomes in CTLA4-insufficient patients have been reported.8, 9 Responses were seen with inhibition of CD28 signalling through the mechanistic target of rapamycin (mTOR) inhibitor, sirolimus, and with CTLA-4 replacement using CTLA-4–Fc fusion proteins such as abatacept or belatacept. Sirolimus has been used in the relapsed/refractory ITP and autoimmune cytopenia setting.10-12 However, many of the patients included in these studies had autoimmune lymphoproliferative syndrome, and CTLA4 variant status was not tested. One recent study that focussed on treatment of CTLA-4-insufficent patients included an analysis of cytopenic patients treated with more traditional refractory ITP agents.7 In this report, 30 patients received rituximab-containing regimens with an 83% response rate. The authors pointed out that rituximab eliminates CD80- and CD86-positive B cells that would activate effector T cells in the setting of CTLA-4 insufficiency. Therefore it is a reasonable second-line choice. Experience with TPO-RA is scarce, and the report documented only one responder out of four patients who received TPO-RA. In our patient, the platelet count began to recover almost immediately after initiation of sirolimus. Three months later, despite a sustained complete response, he developed intolerance to sirolimus (fatigue and diarrhoea) and was transitioned to abatacept. Following abatacept initiation, the patient’s CD4+PD1+ T cells approached normalization (Fig 2A), in keeping with previous reports of TFH response to abatacept in CTLA-4-insufficient patients.13 At the most recent follow-up, he continues to have normal haematopoietic cell counts. Sequencing of family members is under way. We describe a case of ITP in a young man who was refractory to first-line corticosteroid and IVIG treatment. The recognition that he had a novel variant of CTLA4 led to selection of treatment based on mechanistic understanding of CTLA-4 insufficiency. To the best of our knowledge, this is the first report of this specific CTLA4 variant with haematological and immunological phenotyping, informing management of this patient and his family. It also highlights the importance of having a high index of suspicion for causal monogenic variants in the right clinical context, and the ability of genetic sequencing to provide precision treatment. We thank Drs Talal Chatila and Gulbu Uzel for their clinical advice regarding this case. We thank the CAUSES Study for identifying the variant. The investigators of the CAUSES Study include Shelin Adam, Christele Du Souich, Alison Elliott, Anna Lehman, Jill Mwenifumbo, Tanya Nelson, Clara Van Karnebeek, and Jan Friedman; it is funded by Mining for Miracles, British Columbia Children’s Hospital Foundation (grant number F15-01355) and Genome British Columbia (grant number F16-02276). CMB is supported by a Michael Smith Foundation for Health Research Health Professional-Investigator Award and a Providence Health Care Research Institute Early Career Clinician Investigator Award. CMBL wrote the manuscript. AS collected the flow cytometry data. DLM and CAUSES collected the DNA sequencing data. AB, AK, PP, HAL, HM, and CMB critically reviewed the manuscript. The authors declare no potential conflicts of interest regarding the present work.