Thrombotic thrombocytopenic purpura (TTP) is characterised by profound thrombocytopenia, haemolytic anaemia and organ dysfunction (cardiac, neurological or renal). The current treatment strategy includes plasma exchange, corticosteroid therapy, rituximab and caplacizumab, a bivalent humanised 'nanobody' targeting the A1 domain of factor Willebrand, thereby inhibiting platelet adhesion. Several response profiles to this first line of treatment have been observed: 1) durable remission profile (defined by the absence of thrombocytopenia, renal failure or clinical worsening for at least 30 days after the first day of normalisation of platelet levels), 2) refractory profile (defined by a platelet level after 4 days of intensive treatment of less than twice the initial level, associated with a persistently high level of lactate dehydrogenase), 3) relapse profile (defined by the recurrence of neurological manifestations, renal failure and/or thrombocytopenia < 100,000/mm3 for at least 2 days without any other cause identified after durable remission). The prognosis for TTP remains burdened by a mortality rate of around 10% and a relapse rate of 40-50%. The cellular players in the pathogenic process, responsible for the production of autoantibodies in this particularly severe disease, are still poorly characterised. This high-affinity memory B lymphocyte response requires cooperation with T lymphocyte players, namely follicular helper T lymphocytes. The involvement of these specific lymphocyte populations has been highlighted in the literature on other autoimmune diseases mediated by pathogenic autoantibodies, but has been little studied in TTP. The current high mortality rate in TTP suggests that a better understanding of immunopathological processes is required. Based on the model of other autoimmune diseases, this project should make it possible to identify the presence, at diagnosis, of circulating memory B lymphocytes specific for ADAMTS13, and also circulating autoreactive follicular helper T lymphocytes. After treatment with rituximab, depletion of circulating ADAMTS13-specific memory B lymphocytes is expected.