Polymorphism and Auto-reactive B and T Cells Subsets in Adult's Immune Thrombocytopenia (ITP) (Poly-ITP)

Aims of the study : 1) To determine whether the presence of the V158 allele of Fc gammaRIIIA gene is associated with a better outcome of Immune Thrombocytopenia (ITP) in adults and especially with a higher response-rate to rituximab. 2) To analyze the impact of therapy and especially rituximab on some B and T cells autoreactive subsets in the peripheral blood.

Patients and Methods :

Inclusion criteria : age ≥ 18 years, primary ITP defined according to the recent consensual criteria (Rodeghiero F et al. Blood 2009), active ITP defined as an ITP with a platelet count < 50 x 109/L requiring treatment, informed consent. Main exclusion criteria : secondary ITP.

Blood samples (serum, plasma, DNA) will be collected in every patient at time of inclusion (pre-treatment) and then sequentially at 3, 6 and 12 months after inclusion in patients treated with rituximab or during the remission phase for other treatments for immunological studies. When a marrow analysis is indicated, some marrow specimens will also be collected and studied and if a patient will have to undergo a splenectomy as a standard of care, some spleen specimens will also be collected. Fc gamma RIIIA V/F158 polymorphism will be assessed by means of an allele-specific PCR. The sequential analysis of anti-platelets (anti-GpIIbIIIa) antibodies producing B cells will be performed by means of flow cytometry and ELISPOT analysis. T cells subsets will be harvested in presence of GpIIbIIIa immunodominant peptides and and cytokines expression will be measured on supernatants on days 2 and 11 in vitro by using Luminex technology in order to characterize and distinguish TH1, TH2, TH17, TFH and Tregs subsets.

The primary outcome will be the overall response rate 1 year after inclusion defined by a platelet count > 30 x 109/L with at least a two-fold increase of the initial (pre-treatment) count. For the patients treated with rituximab as a standard of care, based on the overall expected response-rate (40-50%) and based on preliminary data on FcgammaRIIIA V/F158 distribution, the inclusion of 85 patients should be sufficient to show an association of the V158 allele and the response (b risk 20% and a 5%). Responders and non responders will be compared by non parametrical tests, a multivariate analysis will be than performed using a logistic regression model. The immunological data B and T cells subsets) obtained in both the responders and the non responders will be compared over time (To, M3, M6 and M12) by non parametrical matched tests.