Immune thrombocytopenia (ITP) is a disorder caused by autoantibody-mediated platelet destruction and decreased platelet production. GPIIbIIIa is an immunodominant antigen in ITP, although other platelet surface proteins are frequently targeted. Approximately 40 percent of ITP patients respond to treatment with rituximab, a monoclonal antibody directed against CD20 on the surface of B cells, although the response is transient in a significant number of these patients.1 Most patients who fail rituximab are cured by splenectomy, indicating that the spleen is the major site of autoantibody production in ITP.
During the humoral response to an antigen, B cells differentiate and proliferate in germinal centers of secondary lymphoid organs such as the spleen. B-cell differentiation involves a branched pathway that leads to the formation of antibody-secreting plasma cells (PCs) or memory B cells (Figure). PCs are non-proliferating and lack CD20. Plasmablasts (PBs) are PC precursors that also lack CD20 and secrete antibody, but are distinguished from PCs by their capacity to proliferate.2 Additionally, they contain surface immunoglobulin and greater amounts of HLA-DR than PCs.
Matthieu Mahévas et al. in the laboratory of Claude-Agnès Reynaud at Paris Descartes University analyzed B-cell populations in splenectomy samples derived from 10 ITP patients who did not respond to treatment with rituximab and five ITP patients who were rituximab-naïve. Additionally, they analyzed normal spleens obtained from patients who had died of stroke. They found that in contrast to normal spleens, spleens from rituximab-naïve ITP patients contained numerous active germinal centers. They used the proliferation marker Ki67 and HLA-DR to discriminate PBs from PCs by flow cytometry. Compared with normal spleens, PBs, but not PCs, were significantly increased in the spleens from rituximab-naïve ITP patients. In four of the five rituximab-naïve ITP spleens, anti-GPIIbIIIa antibody-secreting cells were identified by ELISPOT assay.
The rituximab-treated ITP patients had been splenectomized within the six months after the last rituximab infusion. B cells were barely detectable in peripheral blood of these patients, consistent with previous reports on the duration of response to rituximab. In contrast, the spleen samples in these rituximab-treated ITP patients showed significant numbers of PCs. Memory B cells were rare, and PBs were not observed. Additionally, anti-GPIIbIIIa antibody-secreting cells were identified in seven out of the 10 patients by ELISPOT assay.
The lack of PBs and the small numbers of memory B cells in rituximab-treated ITP patients suggested that the observed antibody-secreting cells were long-lived PCs that were produced before treatment with rituximab. Seven out of 10 of these patients responded to splenectomy, suggesting that autoreactive, long-lived PCs in ITP are often restricted to the spleen. To test this hypothesis, the authors performed transcriptional analysis on PCs and PBs that were sorted from rituximab-naïve and rituximab-treated ITP spleens and from normal spleens using HLA-DR expression as a discriminatory marker. Compared with PBs from rituximab-naïve ITP patients, PCs from rituximab-treated ITP patients displayed a gene expression profile consistent with a quiescent, long-lived phenotype. This phenotype included overexpression of anti-apoptotic genes, negative regulators of the cell cycle, including members of the Kruppel-like factor family, and genes related to the unfolded protein response and underexpression of genes involved in positive regulation of the cell cycle and activating molecules (e.g., toll-like receptors).
In contrast, PCs from normal spleens displayed an intermediate expression profile, including some anti-apoptotic genes shared with long-lived PC and some cell-cycle genes expressed by PB. This observation was confirmed by single-cell expression analysis. At the single-cell level, PCs from rituximab-naïve ITP patients also displayed an intermediate expression phenotype. Because the spleen is an inflammatory milieu in rituximab-naïve ITP patients, the authors suggested ITP spleen does not favor the formation of long-lived PCs, per se.
In Brief
Splenectomy performed in patients with ITP represents a unique opportunity to study the cell population in an autoimmune process. The results of the study by Mahévas et al. indicate that the spleen is the site of an active B-cell response in patients with ITP. The authors identified antibodysecreting, evidently long-lived PCs that survived in the absence of germinal center B cells, dividing memory B cells, and PB. The results suggest that treatment with rituximab may paradoxically lead to the formation of long-lived splenic PCs in some ITP patients.
