Erythroid hypoplasia or aplasia is a hematological condition observed in including idiopathic pure red cell aplasia (PRCA), thymoma-associated PRCA and aplastic anemia. Myelodysplastic syndrome (MDS) with erythroid hypoplasia/aplasia in bone marrow is a rare type of MDS that was not included in existing classifications of MDS. Patients with erythroid hypoplasia/aplasia have common characteristics; transfusion dependencies, immunologic abnormalities and successful immunosuppressive therapies with cyclosporin A (CsA). Thus, we may regard erythroid hypoplasia/aplasia as one of hematological disease entities. However, pathogenic mechanisms of erythroid hypoplasia/aplasia have not been fully elucidated, although T-lymphocyte-mediated inhibition of erythropoiesis is suspected to be the most possible mechanism of the pathogenesis. Recently, we reported that oligoclonal expansion of CD8+/perforin+ T cells was observed in patients with thymoma-associated PRCA and the oligoclonality was exclusively detected in CD8+ T cells, but not CD4+ T cells. To clarify the pathogenetic role of the T-cells, we analyzed the T-cell subsets and therapeutic responses in patients with erythroid hypoplasia/aplasia in bone marrow.
Among 253 patients with MDS diagnosed at Hiroshima University Hospital between 2000 and June 2011, 12 patients (4.7%) showed erythroid hypoplasia/aplasia. A total of 22 patients with erythroid hypoplasia/aplasia, including 8 MDS with erythroid hypoplasia/aplasia, 3 idiopathic PRCA, 3 thymoma-associated PRCA and 8 aplastic anemia, were enrolled in this study. All patients were treated with CsA and improvement in anemia in this study followed the International Working Group (IWG) 2006 criteria. For T-cell subset analysis, mononuclear cells (MNCs) were purified from bone marrow (BM) or peripheral blood (PB) of the patients. MNCs were stained with fluorescent (FITC, PE, PerCP or APC)-conjugated antibodies for CD8, perforin, CCR7, CD62L, CD27, CD28 and CD45RO, CD45RA and were subjected to flow cytometric analysis. As controls, 30 patients with MDS without erythroid hypoplasia/aplasia and 30 patients without BM abnormalities were also analyzed.
Among 22 patients with erythroid hypoplasia/aplasia, 10 patients (4 MDS with erythroid hypoplasia/aplasia, 1 idiopathic PRCA, 3 thymoma-associated PRCA and 2 aplastic anemia) responded to CsA therapy within 2 to 8 weeks. The median blood hemoglobin concentration increased from 6.5 g/dL at the baseline to 9.3 g/dL with treatment, with a median increase of hemoglobin of 2.8 g/dL from the baseline. We attempted to compare the T-cell subsets between CsA-responders and non-responders. All of 3 thymoma-associated PRCA showed good response to CsA therapy, suggesting that the oligoclonal expansion of a CD8+/perforin+ T-cell subset may be associated with the responses to immunosuppressive therapy. Thus, we focused on a T-cell subpopulation expressing CD8+/perforin+. Intriguingly, the CD8+/perforin+ T cells were significantly increased in the CsA-responders (44.3 ± 9.6%, n=10) compared to the non-responders (19.0 ± 9.3%, n=12, P<0.0001), normal BM controls (16.9 ± 7.0%, n=30) and MDS without erythroid hypoplasia/aplasia (15.1 ± 7.0%, n=30). Among the CD8+/perforin+ T cells, CD27+/CD62L+/−/CCR7low/CD28low/CD45RA++/CD45RO+ population was prominent, which is consistent with an effector memory T (TEM) cell subset described by Decrion et al.
Our study reveals that CD8+/perforin+ T cell subset is a large population in the patients with CsA-responsive erythroid hypoplasia/aplasia. It is suggested that CD8+/perforin+ T cell subset may have functions to reduce erythroid progenitors via immunological mechanisms. The mechanisms may be easily suppressed by immunosuppressive therapies. In conclusion, expansion of CD8+/perforin+ T cell subset predicts response to cyclosporin A therapy in patients with erythroid hypoplasia/aplasia. The disease entity of “erythroid hypoplasia/aplasia in bone marrow with expansion of CD8+/perforin+ T cell subset”, including MDS, PRCA with or without thymoma and aplastic anemia, may have common pathogenetic mechanisms.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.