Search

To identify novel autoantibodies in acquired aplastic anemia (AA), we screened the sera of patients with AA possessing small populations of paroxysmal nocturnal hemoglobinuria (PNH)–type cells for the presence of antibodies (Abs) which recognize proteins derived from a leukemia cell line, UT-7. Immunoblotting using proteins derived from lysates or culture supernatants of UT-7 cells revealed the presence of IgG Abs specific to an 80-kDa protein. Peptide mass fingerprinting identified this 80-kDa protein as moesin. Enzyme-linked immunosorbent assay (ELISA) using recombinant moesin showed high titers of antimoesin Abs in 25 (37%) of 67 patients with AA. Moesin was secreted from several myeloid leukemia cell lines other than UT-7, such as OUN-1 and K562, as an exosomal protein. The presence of antimoesin Abs was significantly correlated with the presence of PNH-type cells and antidiazepam-binding inhibitor-related protein-1 (DRS-1) Abs. Patients with AA that did not show any of these 3 markers tended to respond poorly to immunosuppressive therapy. These findings suggest that a B-cell response to moesin, possibly derived from hematopoietic cells, frequently occurs in patients with AA and that detection of antimoesin Abs in combination with other markers may be useful in diagnosing immune pathophysiology in patients with AA.

We investigated the clinical significance of a minor population of paroxysmal nocturnal hemoglobinuria (PNH)-type blood cells in patients with acquired aplastic anemia (AA). We quantified CD55-CD59- granulocytes and red blood cells (RBCs) in peripheral blood from 122 patients with recently diagnosed AA and correlated numbers of PNH-type cells and responses to immunosuppressive therapy (IST). Flow cytometry detected 0.005% to 23.1% of GPI-AP- cells in 68% of patients with AA. Sixty-eight of 83 (91%) patients with an increased proportion of PNH-type cells (PNH+) responded to antithymocyte globulin (ATG) + cyclosporin (CsA) therapy, whereas 18 of 39 (48%) without such an increase (PNH-) responded. Failure-free survival rates were significantly higher (64%) among patients with PNH+ than patients with PNH- (12%) at 5 years, although overall survival rates were comparable between the groups. Numbers of PNH-type and normal-type cells increased in parallel among most patients with PNH+ who responded to IST, suggesting that these cells are equally sensitive to immune attack. These results indicate that a minor population of PNH-type cells represents a reliable marker of a positive IST response and a favorable prognosis among patients with AA. Furthermore, immune attack against hematopoietic stem cells that allows PNH clonal expansion might occur only at the onset of AA.

Introduction: Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired stem cell disorder with the expansion of PIGA mutant clone(s), which is deficient in GPI-anchored proteins including CD55 and CD59. The lack of CD55 and CD59 renders PNH-type red blood cells (RBC) susceptible to complement attacks, resulting in intravascular hemolysis. Classic PNH manifests 3 major symptoms: anemia, bone marrow failure, and thrombosis. Small populations of PNH-type cells ( Methods: Patients diagnosed with PNH, AA, MDS or indistinguishable BMF were prospectively recruited to the study since 2011 in Japan. A high-resolution FCM assay was established to precisely detect a small population of PNH-type granulocytes (with FLAER) and RBCs (with anti-CD55 and CD59 antibodies) ≤ 0.01% of the total granulocyte or RBC population based on the Kanazawa method (Blood 2006 107:1308-1314). Six university laboratories across Japan were designated as regional analyzing centers and measured the percentages of PNH-type cells in the study population, as well as collected clinical and laboratory data. Periodic blind cross validation tests using a positive control sample containing 0.01% PNH-type cells and a negative control sample were performed to minimize inter-laboratory variations. Results: As of July 2016, 2,849 patients were enrolled in the study and 2,734 patients were analyzed. Nine hundred twelve patients (33.4%) were positive for PNH-type cells (≥ 0.005% PNH-type erythrocytes and/or ≥ 0.003% PNH-type granulocytes) and 238 (8.7%) patients had more than 1% of PNH-type cells. PNH-type cells were positive in 90/90 PNH (100%), 512/982 AA (52.1%), 132/822 MDS (16.1%), and 141/512 indistinguishable BMF (27.5%) patients. Among the MDS patients, PNH-type cells were positive in approximately 20% of patients with RCUD, RCMD, MDS-U, or 5q- syndrome, but not in any patients with RARS, RAEB-1, and RAEB-2 (Fig. 1). The serum LDH level increased in proportion to the PNH clone size of RBCs, and 63.3% of the patients possessing ≥1.0% RBCs showed LDH levels more than 1.5 times the upper limit of normal. Of 171 patients who completed submission of 3-year follow-up data, BMF patients with PNH-type cells showed a better response rate [CR+PR, 99/107 (92.5%)] to IST compared to those without PNH-type cells [44/64 (68.8%)] (P Conclusions: These interim analyses of the OPTIMA study demonstrate that our high-resolution FCM is reliable in detecting small populations of PNH-type cells and produces consistent results among different laboratories. The presence of PNH-type cells exclusively in patients with AA and low-risk MDS suggests a link between benign pathophysiology of BMF and an increase in the number of PNH-type cells. The better response of PNH-type cell-positive BMF to IST compared to BMF without PNH-type cells was consistent with previous reports. Our data, for the first time, prospectively confirms the significance of small populations of PNH cells in BMF patients in Japan and warrants further worldwide, prospective studies on non-Japanese patients with BMF. Disclosures Ueda: Alexion Pharmaceuticals: Honoraria, Research Funding. Nishimura:Alexion Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau. Hosokawa:Aplastic Anemia and MDS International Foundation: Research Funding. Yonemura:Alexion Pharmaceuticals: Research Funding. Obara:Alexion Pharmaceuticals: Honoraria, Research Funding. Shichishima:Alexion Pharmaceuticals, Inc. Japan: Honoraria. Ninomiya:Alexion Pharmaceuticals: Honoraria. Kawaguchi:Alexion Pharmaceuticals: Honoraria. Kanakura:Fujimotoseiyaku: Research Funding; Toyama Chemical: Research Funding; Bristol - Myers: Research Funding; Nippon Shinyaku: Research Funding; Astellas: Research Funding; Eisai: Research Funding; Pfizer: Research Funding; Chugai Pharmaceutical: Research Funding; Shionogi: Research Funding; Kyowa Hakko Kirin: Research Funding; Alexion Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau. Nakao:Alexion Pharmaceuticals: Honoraria, Research Funding.

金沢大学附属病院血液内科, A small population of CD55(-)CD59(-) blood cells was detected in a patient who developed donor-type late graft failure after allogeneic stem cell transplantation (SCT) for treatment of aplastic anemia (AA). Chimerism and PIGA gene analyses showed the paroxysmal nocturnal hemoglobinuria (PNH)-type granulocytes to be of a donor-derived stem cell with a thymine insertion in PIGA exon 2. A sensitive mutation-specific polymerase chain reaction (PCR)-based analysis detected the mutation exclusively in DNA derived from the donor bone marrow (BM) cells. The patient responded to immunosuppressive therapy and achieved transfusion independence. The small population of PNH-type cells was undetectable in any of the 50 SCT recipients showing stable engraftment. The de novo development of donor cell-derived AA with a small population of PNH-type cells in this patient supports the concept that glycosyl phosphatidylinositol-anchored protein-deficient stem cells have a survival advantage in the setting of immune-mediated BM injury.

Background: Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired hematopoietic stem cell disorder caused by the clonal expansion of the phosphatidylinositol glycan class A (PIGA) mutant hematopoietic stem cells, which results in a deficiency in glycosylphosphatidylinositol-anchored proteins (GPI-APs). Through the high-resolution flow cytometry-based method, GPI-APs deficient blood cells (i.e., PNH-type cells) are often detectable in patients with bone marrow failure syndromes (BMF), such as aplastic anemia (AA) and low-risk types of myelodysplastic syndromes (MDS). Sugimori et al reported that when BMF patients possessed increased PNH-type cells, the patients had a good prognosis and showed a high response rate to immunosuppressive therapies, suggesting that detection of PNH-type cells is potentially useful in determining an optimal treatment for BMF patients. Thus, we conducted a nationwide, multi-center prospective observational investigation, the OPTIMA study. Methods: From July 2011, we start recruiting the patients with BMF that were diagnosed at various hematology clinics throughout Japan to the OPTIMA study. The primary endpoint of this study was to determine the prevalence of BMF patients with PNH-type cells and to clarify the clinical significance of the presence and quantitative changes of these cells with regard to the clinical features. Six different university laboratories were assigned as regional analyzing centers. The percentage of PNH-type cells was measured by the high-resolution flow cytometry-based method, originally established in Kanazawa University. At six individual laboratories, cross validations were conducted twice a year to minimize the inter-laboratory variations in the detection sensitivities, cutoff values, etc. The liquid FLAER method (≥0.003%) and cocktail method (≥0.005%) with CD55 and CD59 antibodies were used for the detection of PNH-type granulocytes and erythrocytes, respectively. Results Between July 2011 and May 2015, a total of 2328 patients were enrolled to this study, and we analyzed 2212 patients who were eligible for the interim analysis. Of these patients, 74 (3.3%) were diagnosed with PNH, 690 (31.2%) with AA, 592 (26.8%) with MDS, and 856 (38.7%) with undiagnosed BMF. Using high-resolution flow cytometry-based method, 755 (34.1%; 95.9% in PNH, 52.8% in AA, 18.2% in MDS, and 24.8% in undiagnosed BMT) patients had ≥0.005% PNH-type erythrocytes and ≥0.003% PNH-type granulocytes. Overall, 181(8.2%) patients had ≥1% of both PNH-type erythrocytes and granulocytes; the prevalence in each disease subset was 68/74 (91.9%) in PNH, 67/690 (9.7%) in AA, 22/592 (3.7%) in MDS, and 24/856 (2.8%) in undiagnosed BMF. Regarding FAB and WHO classifications of MDS subtype, no patients with RARS (0/22), RAEB-1 (0/37) or RAEB-2 (0/23) had PNH-type cells. In contrast, 20.4% (56/275) patients with RCMD, 18.3% (26/153) patients with RCUD and 50% (2/4) patients with del (5q) MDS possessed increased PNH-type cells. Blood samples from 75 (65 with and 10 without PNH-type cells) patients were analyzed three years after the first examination. Of 65 PNH (+) patients, PNH-type cells disappeared in 4 (6.2%), while the percentage remained stable in 61 (93.8%). All of the 10 PNH (-) at the enrollment were also negative for PNH-type cells in 3 years. Conclusions: A high-resolution flow cytometry-based method that enables the detection of minimal PNH-type cells below 0.01% was successfully transferred from Kanazawa University to other laboratories in Japan. Our interim analysis confirmed previous findings that PNH-type cells were detectable in patients with 52.8% of AA and 18.2% of MDS patients. Regarding FAB and WHO classifications of MDS subtype, PNH-type cells were not detected in any of MDS RARS, RAEB-1 or RAEB-2 patients. Further analysis are required to determine the clinical significance of the minimal level of PNH-type cells as well as chronological changes in the PNH-type cell percentage, especially in terms of their relation to response to immunosuppressive therapy. Disclosures Ninomiya: Alexion Pharmaceuticals: Honoraria. Ando:Eisai Co., Ltd.: Honoraria, Research Funding. Yonemura:1. Chugai Pharma, 2. Alexion Pharma, 3. Japan Blood Products Organization, 4. OHARA Pharma: Research Funding. Kawaguchi:Alexion Pharmaceuticals: Honoraria. Ueda:Alexion Pharma: Research Funding. Nishimura:Alexion Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau.

金沢大学医学部附属病院内科, Objective: Although a number of reports have documented a significantly increased incidence of HLA-DR15 in aplastic anemia (AA), the exact role of HLA-DR15 in the immune mechanisms of AA remains unclear. We herein clarify the difference between DRB1*1501 and DRB1*1502, the two DRB1 alleles that determine the presentation of HLA-DR15, in the pathophysiology of AA. Materials and Methods: We investigated the relationships of the patients* HLA-DRB1 allele with both the presence of a small population of CD55-CD59- (PNH-type) blood cells and the response to antithymocyte globulin (ATG) plus cyclosporin (CsA) therapy in 140 Japanese AA patients. Results: Of the 30 different DRB1 alleles, only DRB1*1501 (33.6% vs 12.8%, pc < 0.01) and DRB1*1502 (43.6% vs 24.4%, pc < 0.01) displayed significantly higher frequencies among the AA patients than among a control. AA patients possessing HLA-DR15 tended to be old, and especially, the frequency of DRB1*1502 in patients 40 years of age and older (52.4%) was markedly higher than that in those younger than 40 years old (16.2%, pc < 0.01). Only DRB1*1501 was significantly associated with the presence of a small population of PNH-type cells and it also showed a good response to ATG plus CsA therapy in a univariate analysis. A multivariate analysis showed only the presence of a small population of PNH-type cells to be a significant factor associated with a good response to the immunosuppressive therapy (p < 0.01). Conclusions: Although both DRB1*1501 and DRB1*1502 contribute to the development of AA, the methods of contribution differ between the two alleles. © 2007 International Society for Experimental Hematology.

To identify candidate antigens in aplastic anemia (AA), we screened proteins derived from a leukemia cell line with serum of an AA patient and identified diazepam-binding inhibitor-related protein 1 (DRS-1). Enzyme-linked immunosorbent assay (ELISA) revealed high titers of anti–DRS-1 antibodies (DRS-1 Abs) in 27 (38.0%) of 71 AA patients displaying increased paroxysmal nocturnal hemoglobinuria (PNH)–type cells (PNH+), 2 (6.3%) of 32 PNH– AA patients, 5 (38.5%) of 13 PNH+ myelodysplastic syndrome (MDS) patients, and none of 42 PNH– MDS patients. DRS-1 gene was abundantly expressed in myeloid leukemia cell lines and in CD34+ cells derived from healthy individuals. Stimulation of T cells from an AA patient displaying high DRS-1 Abs with a putative CD4+ T-cell epitope (amino acid residues [aa's] 191-204) presented by HLA-DR15, which overlapped with a hot spot (aa's 173-198) of DRS-1 Ab epitopes, gave rise to T cells cytotoxic for L cells (murine fibroblasts) that were transfected with DRB1*1501 and DRS-1. Enzyme-linked immunospot assay demonstrated increased frequency of T-cell precursors specific to the DRS-1 peptide in other HLA-DR15+ AA patients displaying high DRS-1 Ab titers. These findings indicate that DRS-1 may serve as an autoantigen eliciting immune attack against hematopoietic stem cells in a subset of AA patients characterized by increased PNH-type cells.

Background: A subset of patients with acquired aplastic anemia (AA) require low-dose cyclosporine (CsA) to maintain good hematopoietic function even after achieving a remission with immunosuppressive therapy (IST). Such CsA-dependent AA offers opportunities to clarify the immune mechanisms underlying bone marrow (BM) failure, but little is known about the mechanism(s) underlying this condition except for its close association with a particular HLA-class II allele DRB1*15:01. Regulatory T cells (Tregs), defined as CD4+CD25highFoxP3+ T cells, have been shown to be decreased in patients with severe AA at diagnosis and to recover to normal levels in association with hematological recovery after successful IST. An insufficient Treg recovery may account for dependency on CsA in a particular subset of AA patients, and also in patients with idiopathic pure red cell aplasia (PRCA), another typical immune-mediated BM failure of which remission is known to depend on CsA in virtually all patients. To test this hypothesis, we determined the percentage and absolute number of Tregs, as well as Th1, Th2 and Th17 cells, in various types of immune-mediated BM failure, using an anti-FoxP3 antibody (Ab) and anti-Helios Ab that allows enumeration of the functional Tregs more precisely than conventional methods. Methods: Different CD4+ Treg subsets were enumerated using a multicolor flow cytometry assay. Peripheral blood mononuclear cells (PBMCs) were stimulated by phorbol myristate acetate (PMA) and ionomycin in the presence of monensin for four hours, then were subjected to flow cytometry to determine the percentage of CD4+ T cells positive for cytoplasmic interferon-γ (Th1), interleukin 4 (Th2) and interleukin 17 (Th17). CsA-dependent AA was defined as AA with two or more episodes of relapse of pancytopenia that occurred during more than one year of therapy, followed by a good response to CsA. All six patients with CsA-dependent AA had normal blood cell counts and HLA-DR15 (DRB1*15:01 in five and DRB1*15:02 in one). Results: The normal ranges of each T cell subset percentage and the absolute numbers determined in 47 healthy volunteers (HVs) were: FoxP3+Helios+ Tregs 6.5±1.6%/ 47.7±13.8/µl, Th1 22.3±6.6%/ 151.2± 45.5/µl, Th2 5.5±1.6%/ 38.3±170.9/µl and Th17 1.7± 0.7%/ 11.2± 4.7/µl. The expression of FoxP3 by CD4+ T cells decreased with time after sampling, and this assay therefore necessitated analyses within 16 hours after sampling using fresh PBMCs. Compared to healthy volunteers, the percentage and absolute number of FoxP3+Helios+ Treg cells were significantly lower in the five patients with severe AA at diagnosis (3.7±1.1%, 22.7±6.8/µl) and in six CsA-dependent AA patients (3.4±0.8%, 13.8±12.3/µl) (Figure). In contrast, the Treg percentages were comparable to those of the HVs in four AA patients on CsA in convalescence (5.8±1.0%) and nine patients in remission off CsA (6.3±1.1%) while the absolute numbers of FoxP3+Helios+Tregs (27.3±19.2/µl, 32.3±13.3/µl) in these groups were lower than those of the HVs. On the other hand, the percentages of all other CD4+ T cell subsets were comparable among the four different groups and HVs. The apparent decrease in FoxP3+Helios+ Tregs despite normal blood cell counts in patients with CsA-dependent AA prompted us to screen nine patients with idiopathic PRCA whose erythropoietic functions were all dependent on CsA. The FoxP3+Helios+ Tregs were decreased in both the percentage and absolute count (3.1±0.5%, 12.1±5.9/µl) in all of these patients, including two patients whose PBMCs were examined before CsA therapy. As a result, the ratios of Th1/Treg and Th17/Treg were markedly higher in newly-diagnosed AA (12.5±3.4, 0.9±0.8), CsA-dependent AA (12.6±12.1, 1.1±1.5) and PRCA (10.6±7.7, 0.7±0.3) patients than in AA patients in convalescence (6.1±4.3, 0.4±0.1) and those in remission (3.8±2.3, 0.4±0.2), which were comparable to the HVs Th1/Treg and Th17/Treg ratios (3.3±0.9, 0.3±0.2). Conclusions: Decreased percentages and absolute counts of Tregs, as well as increased Th1/Treg and Th17/Treg ratios, are common features in patients with CsA-dependent AA and PRCA. A persistent decrease in Tregs may serve as a good marker for predicting AA relapse in patients responding to CsA. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.

[Background] Paroxysmal nocturnal hemoglobinuria (PNH) is a hematopoietic stem cell disorder derived from an acquired mutation of the phosphatidylinositol glycan class A (PIGA) gene in the hematopoietic stem cells which results in the expansion of glycosylphospatidylinositol-anchored protein (GPI-AP)-deficient (PNH-type) hematopoietic cells. PNH-type blood cells are also observed in patients with bone marrow failure (BMF). PNH is conventionally diagnosed when patients have >1% of GPI-AP-deficient erythrocytes and granulocytes determined by flow cytometry. Analyses with high resolution flow cytometry by several different groups have shown that patients with aplastic anemia (AA) or low-risk types of myelodysplastic syndromes (MDS) have small percentages of PNH-type erythrocytes, granulocytes, and/or other lineages of blood cells and that these patients respond better to immunosuppressive therapies compared with BMF patients lacking PNH-type cells. In order to determine the prevalence and clinical significance of PNH-type cells in BMF patients, we conducted a nationwide multi-center prospective observational investigation, the OPTIMA study. [Methods] From July 2011, Japanese patients with PNH, AA, MDS or BMF of uncertain origin have been prospectively enrolled into the study. Six laboratories in different cities in Japan were assigned as regional analyzing centers and measured the percentages of PNH-type cells in the study population as well as collecting clinical and laboratory data. The high-resolution flow cytometry assessments used a liquid fluorescein-labeled proaerolysin (FLAER) method and a cocktail method with anti-CD55 and anti-CD59 antibodies for the detection of PNH-type granulocytes and erythrocytes, respectively. Periodic blind cross validation tests using a standard blood sample containing 0.01% PNH-type cells and a normal control were conducted to minimize inter-laboratory variations. From analysis of 68 healthy individuals >0.003% of PNH-type granulocytes and >0.005% of PNH-type erythrocytes were considered to be abnormal (Sugimori et al, Blood, 2006). [Results] As of May 2014, flow cytometry data have been collected from 1685 patients and are included in this interim analysis. Of these patients, 65 (4%) were diagnosed with PNH, 523 (31%) with AA, 459 (27%) with MDS, and 638 (38%) with BMF of unknown etiology. Overall, 154 (9%) patients had ≥1% of both PNH-type erythrocytes and granulocytes: 63 (97%) patients with PNH; 57 (11%) with AA; 18 (4%) with MDS; and 16 (3%) with BMF of unknown etiology. In total, 545 (32%) patients had ≥0.005% PNH-type erythrocytes and ≥0.003% PNH-type granulocytes. These consisted of the followings; all 65 (100%) patients with PNH; 264 (51%) with AA; 76 (17%) with MDS; and 140 (22%) with BMF of unknown origin. Lactate dehydrogenase (LDH) levels ≥1.5 × upper limit of normal range were seen in 14/329 (4%) patients with 0.005-1% PNH-type erythrocytes, 23/62 (37%) patients with 1-10% PNH-type erythrocytes, and 69/71 (97%) patients with ≥10% PNH-type erythrocytes. Periodic blind validation tests revealed that inter-laboratory differences in absolute measurements of PNH-type cells were always within 0.02%. [Conclusion] A high-resolution flow cytometry-based method, based on the Kanazawa method, that enables the detection of very low percentages of PNH-type cells was successfully transferred to 6 laboratories across Japan. Our results demonstrated that the proportion of patients identified as having small percentages of PNH-type cells differed depending on diagnosis (PNH, AA, MDS, or unknown BMF) and that elevated LDH levels (>1.5 x upper limits of normal range) were more frequently associated with higher percentages of PNH-type erythrocytes. Our findings suggest that the high resolution method is helpful as a diagnostic tool in BMF syndromes, including AA, MDS, and PNH, and may prove useful in understanding the pathophysiology of these disorders. Disclosures Noji: Alexion Pharma: Honoraria. Shichishima:Alexion Pharmaceuticals, Inc; and Medical Review Company: Honoraria, Research Funding. Obara:Alexion Pharma: Research Funding. Chiba:Alexion Pharma: Research Funding. Ando:Alexion Pharma: Research Funding. Hayashi:Alexion Pharma: Research Funding. Yonemura:Alexion Pharma: Research Funding. Kawaguchi:Alexion Pharma: Honoraria. Ninomiya:Alexion Pharma: Honoraria, Research Funding. Nishimura:Alexion Pharma: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau. Kanakura:Alexion Pharma: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau.

Background Paroxysmal nocturnal hemoglobinuria (PNH) is a disease derived from an acquired mutation of the phosphatidylinositol glycan class A (PIGA) gene in the hematopoietic stem cells. In some cases with aplastic anemia (AA) or low-risk types of myelodysplastic syndromes (MDS), it is known that glycosylphosphatidylinositol-anchored protein deficient (PNH-type) cells can be often detected at low frequencies (about 0.01%) through the high-resolution flow cytometry-based methFod. Because these patient groups are reported to have a good reactivity towards immunosuppressive therapies as opposed to the other patient groups lacking PNH-type cells, detection of these cells is potentially useful in determining a treatment plan for the patients with bone marrow failure syndromes. To confirm this preliminary information, a large cohort study is needed. Method A nationwide multi-center prospective observational study (OPTIMA) was started in July 2011 to determine the prevalence of patients with bone marrow failure syndromes who carried PNH-type cells and to clarify the significance of the presence and quantitative changes of these cells with regard to the clinical features. Each of the six laboratories in different universities was assigned as a regional analyzing center. The percentage of PNH-type cells was measured by the high-resolution flow cytometry-based method, originally established in Kanazawa University. At six individual laboratories, cross validations were conducted to minimize the inter-laboratory variations in the detection sensitivities, cutoff values, etc. The liquid FLAER method (≥0.003%) and cocktail method (≥0.005%) with CD55 and CD59 antibodies were used for the detection of PNH-type granulocytes and erythrocytes, respectively. Results Quality of the assay was managed in all the laboratories by periodic blind validation tests using standard blood samples containing 0.01% PNH-type cells. Until July 2013, 1214 cases were examined; 461 (38%) were positive for PNH-type cells and 141 (11.6%) had ≥1% PNH-type cells. Out of 1214, 783 patients were diagnosed to have AA (n=386), MDS (n=341), and PNH (n=56) based on the case report forms. PNH-type cells were detected in 56.2%, 19.1% and 100% of patients with AA, MDS and PNH, respectively. In a half of patients having ≥1% PNH-type cells, lactate dehydrogenase levels exceeded the ≥1.5×upper limits of normal. Conclusion Our study has successfully established the high-resolution flow cytometry-based method that enables the detection of minimal PNH-type cells (below 0.01%). Also, by implementing a uniform protocol to six individual laboratories across the country, a system has been established for the patients to undergo the detection test with equal accuracy in all of these laboratories. Further accumulation of case studies and prolonged observations are required to determine the clinical significance of the minimal PNH-type cells, especially in terms of its relation to response to immunosuppressive therapy. Disclosures: Obara: Alexion Pharmaceuticals: Honoraria. Chiba:Alexion Pharmaceuticals, In: Research Funding. Sugimori:Alexion Pharma: Honoraria. Noji:Alexion Pharmaceuticals: Honoraria. Yonemura:Alexion Pharma: Research Funding. Ando:Alexion Pharma: Research Funding. Kawaguchi:Alexion Pharmaceuticals: Honoraria. Shichishima:Alexion Pharmaceuticals, In: Honoraria, Membership on an entity’s Board of Directors or advisory committees, Research Funding. Ninomiya:Alexion Pharma: Honoraria. Nishimura:Alexion Pharma: Research Funding, Speakers Bureau. Kanakura:Alexion Pharmaceuticals: Research Funding, Speakers Bureau. Nakao:Alexion Pharmaceuticals, In: Research Funding.

Abstract 1703 Myleodysplastic syndrome (MDS) is a heterogeneous group of clonal hematopoietic diseases marked by various cytopenias including neutropenia and anemia. Hematologic improvement in response to T-cell depleting agents and immunosuppressants in a subset of MDS patients suggests that hematopoietic impairment may result from autoreactive T-cell activation. Conversely, suppression of the T-cell compartment may be permissible to the development of malignant myeloid clones. The suspicion that T-cell homeostatic balance may be disrupted in MDS has spawned numerous studies investigating mechanisms of immune suppression and autoimmune regulation. T-cell autoimmunity is regulated in part by regulatory T-cells (Tregs), which express T cell receptors (TCRs) that recognize self-antigens and function to maintain immune homeostasis. Indeed, higher numbers of Tregs have been found in higher risk MDS patients, and those MDS subtypes with higher blast percentages. However, the role of Tregs in lower risk patients is less clear, but may still represent an important component of pathogenesis in early disease. Because of the autoimmune nature of low risk MDS, we hypothesized that alterations in the Treg compartment may play a role in lower risk MDS disease progression. To address this, we analyzed Treg phenotype and numbers in MDS patients largely classified as low risk by the International Prognostic Scoring System (IPSS) and 41 age-matched controls using cell surface markers as analyzed by flow cytometry. Tregs were identified as CD3+CD4+CD25+CD127dim. The use of CD127 as a Treg marker was validated by intracellular staining with FoxP3. Naïve and memory phenotypes within the Treg compartment were analyzed using CD45RA and CD27 expression. Naïve Tregs (TregN) were characterized as CD45RA+CD27+, central memory Tregs (TregCM) as CD45RA−CD27+, and effector memory Tregs (TregEM) as CD45RA−CD27−. Two distinct subgroups of MDS patients were identified: one with elevated total numbers of Tregs, and a second with a marked shift in Treg memory phenotype away from the common TregCM phenotype and toward the rare TregEM phenotype. These two subgroups of patients were distinct from each other, and appeared to be independent of MDS WHO subtype, IPSS score, karyotype, neutropenic state, and thrombocytopenic state. MDS patients with a shift toward TregEMwere associated with anemia (p=0.0433). This association was not found in MDS patients with total increases in Treg numbers. Next, we sought to determine if there are differences in overall survival (OS) of MDS patients with either increased total numbers of Tregs or in patients with a shift toward TregEM phenotype. Patients with greater than 50 Tregs/μl did have reduced OS compared to patients with less Tregs, but this difference was not statistically significant (p=0.1387). In contrast, a highly significant difference was observed in MDS patients with a shift toward the TregEM phenotype (p=0.0200), with patients having greater than 2.5TregEM/μl displaying the worst OS (Figure 1). Figure 1 Overall Survival in lower risk MDS patients with Low, Intermediate, or High absolute numbers of TregEM cells Figure 1. Overall Survival in lower risk MDS patients with Low, Intermediate, or High absolute numbers of TregEM cells Lastly, because an increased TregEM compartment is associated with reduced OS in MDS patients, we sought to compare the suppressive capacities of TregN, TregCM, and TregEM cells to better understand the functional consequences of Treg memory shift. CD4+ T-cells were isolated by negative selection, and then TregN, TregCM, and TregEM cells were sorted using CD45RA and CD27 expression. Conventional CD4+ responder cells were then labeled with Carboxyfluorescein succinimidyl ester (CFSE) and co-cultured with increasing ratios of each isolated Treg population. Cellular division in response to CD3/CD28 stimulation was measured 5 days later by analyzing CFSE dilution. TregCM appear to be the least suppressive, while TregN and TregEM cells were more potently suppressive. These results indicate that a shift in Treg phenotype toward TregEM may result in a more suppressive Treg compartment without obvious increases in total Treg numbers. Taken together, this data suggests that in lower risk MDS, the phenotypic makeup of the Treg compartment may be more important than total Treg numbers. In addition, this data suggests that increased TregEM cells may be a novel prognostic indicator in low risk MDS. Disclosures: Komrokji: Celgene: Honoraria, Research Funding, Speakers Bureau. List:Celgene: Consultancy.

Abstract 1158 Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired hematopoietic stem cell disorder characterized by a global deficiency of glycosylphospatidyl inositol-floated proteins (GPI-AP) on the membrane of all cell types derived from the affected stem cell. While the PNH phenotype is non-malignant, outgrowth of GPI-AP (-) clones is mediated by a selective advantage in the context of a permissive condition, such as autoimmune-mediated bone marrow suppression as seen in aplastic anemia (AA). The GPI-AP (-) phenotype results from mutations of the PIG-A gene, located at Xp22.2, which participates in GPI synthesis. Because the gene is X-linked, a single frame-shift or missense inactivating mutation results in loss of enzyme function resulting in GPI-AP deficiency on the cell surface. Due to × chromosome inactivation, the frequency of PNH is similar in men and women. New cytogenetic technologies such as single nucleotide polymorphism array (SNP-A)-based karyotyping show superb resolution allowing for detection of submicroscopic chromosomal defects not seen on metaphase karyotyping. We applied this technology to identify clonal aberrations in patients with PNH (N=32). The average size of the PNH by percentage of CD55/CD59 double negative granulocytes was 73%. Among this cohort, we identified 3 patients with a clonal GPI (-) hematopoietic cell population resembling canonical PNH, but who did not harbor an identifiable mutation in PIG-A. Patients had classic symptoms of hemolysis and fairly typical clinical course. By quantitative PCR, these patients also lacked expression of PIG-A. During this molecular screen, a microdeletion of Xp22.2 encompassing the PIG-A locus, was found in GPI (-) sorted myeloid cells in these 3 patients using GPI (+) T cells as the comparator. This finding underscores the acquired somatic nature of this deletion in these patients. The deletions were 555kb, 506kb and 616kb in length. A commonly deleted region (CDR) of 437kb included PIGA as well seven known neighboring genes (ASB9, ASB11, FIGF, PIR, BMX, ACE2, and TMEM27) and predicted LOC100288981. One patient with the Xq22.2 deletion also had copy neutral loss of heterozygosity (CN-LOH) encompassing 9p24.3p21.3; the JAK2 V617F mutation was found in this patient. One of the remaining 29 patients had trisomy 8, 9 and 21, suggesting a concurrent diagnosis of myelodysplastic syndrome. The PNH phenotype in the male patients (N=2) is easily explained by a hemizygous deletion of the X-linked PIG-A gene. However, in the female patient, the deletion must have occurred on the active (un-lyonized) × chromosome leading to loss of PIG-A transcription and thereby GPI (-) clonal expansion. Nested PCR reactions were performed to amply individual exons of the PIG-A gene and the presence of a wild-type PIG-A gene on the inactive × chromosome was verified by exon amplification, cloning, and sequencing. Our observation indicates that in addition to mutations, the PNH phenotype may arise from an alternate molecular mechanism, namely a segmental chromosomal deletion involving the PIGA locus in addition to inactivating mutations. While three instances of small intragenic deletions, including 2 Disclosures: No relevant conflicts of interest to declare.

Abstract 2777 Poster Board II-753 Telomeres are repeated nucleotide sequences at the ends of chromosomes that serve to preserve genetic integrity in hematopoietic progenitors (HPs). Abnormal telomere maintenance resulting from mutations in telomerase or other telomere repair genes have an established pathogenetic role in a subset of acquired and congenital forms of bone marrow failure (BMF). Excess telomere shortening of myeloid elements was reported previously in MDS, however it is unclear whether such abnormalities in telomere maintenance play a pathogenetic role in disease susceptibility or represent a disease specific phenotype arising from accelerated cellular proliferation of the myeloid compartment. Alterations in telomere maintenance and telomerase activity in T cells may offer insight into genetic susceptibility since these cells are not derived from the malignant clone. We examined telomere length and replication history in neutrophils and T cells in newly diagnosed MDS cases (n=66) and compared results to unrelated, age-matched controls (n=63) lacking a prior history of cancer. Telomere length in peripheral blood mononuclear cells was assessed by quantitative real-time PCR (qRT-PCR) and found to be significantly shorter among MDS cases compared to controls. Peripheral blood concentration of the senescence protein stathmin was analyzed in a subset of 48 patients and 48 controls and found to be higher in MDS cases compared to controls after adjustment for age and sex (mean 4.24 pg/ml ± 7.25 in cases vs mean 2.45 pg/ml ± 2.84 in controls, p=0.06) with no change observed with age. Mechanisms of cellular senescence and telomere attrition in T-cells may be related to antigen-driven or homeostatic proliferation. Therefore, telomere length was compared to an independent genetic marker of proliferation, i.e., the concentration of T cell receptor excision circles (TRECs). TRECs are episomal DNA fragments excised during TCR gene rearrangement within the thymus that do not transfer to daughter cells. Therefore, TREC DNA copy number has been shown to be diluted during each round of division and tends to be reduced with age due to impaired generation of new thymic T-cell emigrants. In controls, TREC concentration declined with age consistent with cellular proliferation and loss of thymic function. For every year increase in age, log TREC values decreased by 0.05 DNA copies (p=0.0012). More aggressive proliferation was evident in T cells from cases compared to controls with a 2-fold more rapid decline in TREC copy number each year (0.099 unit decrease in TREC copies per year among cases, p Disclosures: No relevant conflicts of interest to declare.

Abstract 1085 Poster Board I-107 Background Benign types of bone marrow failure such as aplastic anemia (AA) and refractory anemia (RA) of MDS defined by the FAB classification occasionally present various chromosomal abnormalities, although the clinical significance of each abnormality is unclear. Small populations of CD55-CD59- granulocytes and erythrocytes are detected in approximately 50% of patients with AA and 15% of patients with RA (Br J Haematol 2009). Analyzing the relationship between the presence of a certain chromosomal abnormality and PNH-type cells may be useful in deducing the clinical significance of the karyotype abnormalities because such PNH-type cells represent a good marker for the benign bone marrow failure with immune pathophysiology. Patients and Methods From 1995 through 2009, peripheral blood from 2487 patients with AA or RA was examined for the presence of PNH-type cells using high sensitivity flow cytometry. Chromosome data on bone marrow cells were available in 1513 patients. The proportion of patients possessing PNH-type cells was determined in a subset of AA or RA defined by various chromosomal abnormalities. The sorted PNH-type granulocytes from some patients with chromosomal abnormalities were subjected to FISH analysis to determine the origin of the cells. Results Chromosomal abnormalities were detected in 16% of the patient population, 11% (59/546) in patients possessing PNH-type cells (PNH+ patients) and 19% (186/967) in patients without increased PNH-type cells (PNH- patients). The most frequent chromosomal abnormalities were trisomy 8 (+8) (3.6%), deletion of 20q (del(20q)) (2.4%), deletion of 13q (del(13q)) (1.0%), monosomy 7 (-7) (0.5%), deletion of 5q (del(5q)) (0.3%) in order of descending prevalence. The proportion of PNH+ patients in each patient group defined by chromosomal abnormality was 24%, 14%, 73%, 0% and 0% respectively. The proportion of PNH+ patients in patients with normal karyotype was 35%. Notably, patients with del(13q) showed a remarkably high prevalence of PNH-type cells (73%), even significantly higher than the prevalence of the patients without chromosomal abnormalities (P Conclusions The presence of del(13q) in AA or RA represents benign bone marrow failure associated with a good response to immunosuppressive therapy, while the presence of -7 and del(5q) is thus considered to be related to a non-immune pathophysiology. AA and RA with +8 or del(20q) comprise a small subset of immune-mediated bone marrow failure. Disclosures No relevant conflicts of interest to declare.

Abstract 3193 Poster Board III-130 Paroxysmal nocturnal hemogloginuria (PNH) is a clonal genetic disorder associated with multiple mutations in the X-linked phosphatidylinositol glycan class A (PIG-A) gene that causes glycosyl phosphatidylinositol (GPI) anchor protein (AP) deficiency in patients with PNH. Conditional PIG-A gene inactivation (PIG-A (-/-)) in hematopoietic cells of mice does not induce overt proliferative advantage to the HPC compartment. In humans, the molecular basis of clonal expansion of GPI-AP deficient cells is unknown but may relate to selective pressure from the bone marrow microenvironment conducive to GPI-AP survival. Alternatively, coexistence of additional genetic anomalies that confer growth potential has been speculated. A single mutation in the Janus Kinase (JAK)-2 gene has been reported to be the underlying molecular mechanism for myeloproliferative diseases including polycythemia vera (PV), essential thrombocythemia, and myelofibrosis. Substitution of a valine for phenylalanine destabilizes the auto-inhibitory activation domain of JAK2 leading to myeloproliferation. Genetic evidence and in vitro functional studies indicate that V617F gives hematopoietic precursors proliferative and survival advantages. A high proportion of patients with myeloproliferative disorders and myelodysplastic syndrome (MDS) carry this dominant gain-of-function mutation of JAKV617F. The clinical syndromes of PNH and JAKV617F mutations often present as overlapping clinical syndromes raising the possibility that acquired JAKV617F mutations confer growth and survival potential to PNH clones in at least a subset of patients. To address this possibility, 21 PNH patients were screened for the presence of JAKV617F mutations. We show here that the features of Budd-Chiari syndrome (BCS) as well as a predisposition to myelodysplastic syndrome can be caused by an acquired co-existing PIGA and JAKV617F activating mutation. Of the 21 cases examined, PNH with JAK2V617F mutations co-existed in three cases (14.3%). Thus far, two patients have been well characterized. Both cases were male, ages 51 (case 1) and 65 year-old (case 2) with normal cytogenetics and de novo classic PNH manifested by Budd-Chiari syndrome. Flow cytometry assay using CD55, CD59, and CD48 (GPI-APs) and FLAER to detect PNH type cells showed that 90% of granulocytes (case 1) and almost 100% of granulocytes (case 2) were GPI-AP deficient cells. Interestingly, T-cells in both cases displayed a normal phenotype suggesting that the PIGA gene mutation was selectively acquired in a common myeloid progenitor (CMP). From sorted populations of PNH+ and normal type granulocytes (case 1 only) or T cells (case 2), the JAK2V617F mutant clone was found to co-exist with the PIGA mutation using allele-specific polymerase chain reaction (PCR). Real-time quantitative PCR using primers and probe to the junction of exon 5-6 coding region demonstrated drastically low to absent mRNA transcript expression of PIGA only in PNH-type granulocytes in both cases. Individual RT-PCR of the coding region of each exon (2-6) confirmed that transcription was absent in case 1 but limited to regions encoded by exons 2,4, and 6 in case 2. These results suggest that case 2 possesses a splicing mutation involving exon 5 to 6. In conclusion, JAK2V617F mutation is known to cause the constitutive activation of the JAK-STAT signaling pathway, and leads to autonomous cell growth in a cytokine-independent expansion of HPCs. While it has been reported that JAK2V617F mutation occurs in roughly 50% of primary Budd-Chiari syndrome and in 15 patients of 163 (9.2%) (Hoekstra et al, Journal of Hepatology, 6:19, 2009) of patients with PNH, these findings implicate co-existing JAK2V617F /PIG-A mutations may contribute clinically to specific manifestations of splenic venous thrombosis or hepatic vein thrombosis in association with intravascular coagulation and myelodysplasia. Disclosures No relevant conflicts of interest to declare.

Background: An increase in CD4+FoxP3+ regulatory T cells (Tregs) has been associated with disease progression in myelodysplastic syndrome (MDS). Tregs provide peripheral tolerance through naturally occurring negative regulatory mechanism that control self-reactivity and autoimmunity. MDS is a heterogeneous hematopoietic disease linked to several distinct pathophysiologic mechanisms that occurs primarily in older individuals. Through the work of our laboratory and others, the balance of CD4 and CD8 T cells has been shown to be disturbed in MDS patients that respond to immunosuppressive therapy. IST offers hematologic benefit and improved survival in a subpopulation of MDS patients with evidence of a T-cell autodominant immune process. It is critical to clarify the correlation between dysregulated naïve and memory CD4+ T cells and dysfunction of Tregs since it has been reported that the supply and function of naïve Tregs is important for protection against autoimmune diseases mediated by T-cells. Methods: We analyzed the direct correlation among FoxP3, CD25 and CD127 expression, and that of CD45RA and CD27 to define Tregs with naïve, central memory, and effector memory phenotypes in 45 MDS patients and 17 healthy controls. We developed an 8-color flow cytometry staining approach that included the following phenotypic markers to define the cell populations: antibodies to CD3, CD4, CD25, CD127 (IL-7Rα), CD45RA, CD27, FoxP3, and aqua fluorescent dye for live:dead cell discrimination. Results: A total of 23 men and 22 women were enrolled from the Bone Marrow Failure Rare Disease Research Network with a diagnosis of MDS with a median age of 69 years old. Based on International Prognostic Scoring System classification, 90% were in a lower risk category (low and intermediate −1 risk). The controls were 17 healthy individuals that donated blood to the Southwest Florida Blood Services (SFBS) and consisted of 8 men and 9 women (median age 52 years old). We found that bright CD25 expression and dim expression of CD127 were most closely associated with FoxP3 expression in healthy controls. The transcription factor FoxP3, which is the most specific phenotypic marker of Tregs, displayed a distinct association with CD25 and CD127 in naïve and memory cells. Therefore, naïve and memory Tregs were defined by FoxP3 expression and the following phenotypes: naïve Tregs (CD45RA+/CD27+/FoxP3+), central memory Tregs (CD45RA−/CD27+/FoxP3+), and effector memory Tregs (CD45RA−/CD27−/FoxP3+). The mean percentage of FoxP3+ Tregs in MDS patients did not differ from healthy controls (5.24% and 4.83%, respectively P=0.94). Younger age, < 61 years of age, has been strongly associated with responsiveness to IST and T-cell-mediated autoimmunity. When the Tregs were examined in patients divided into two groups based on an age of 60 years old, a higher percentage of FoxP3 positive cells was observed in the older age group compared to the younger group (mean, 7.46% and 4.98%, respectively, P=0.12) of MDS patients. Interestingly, the percentage of FoxP3+ Tregs tended to be inversely correlated with the lymphocyte count (P=0.08) and the CD4/CD8 ratio (P=0.1) in patients. When naïve and memory Tregs were compared between cases and controls, these Treg populations demonstrated distinct skewing with effector memory Tregs dramatically higher in some patients. MDS patients were then divided into two groups based on the percentage of effector memory Tregs. Using a cutpoint equal to the mean + 2 S.D of controls, the patients were grouped by the percentage of effector memory (EM) Tregs into two groups: higher EM-Tregs (mean 39.3% ± 12.5%) and lower EM-Tregs (mean 6.39% ± 5.16%). The patient group with higher EM-Tregs was significantly associated with the presence of anemia (P=0.01) compared to the group with lower effector memory Tregs Conclusions: Memory Tregs have been reported to express higher levels of integrin alphaE, display differential tissue distribution, and a higher turnover rate compared to naïve Tregs. Memory Tregs differentiate from naïve CD25−FoxP3− non-Tregs in the peripheral blood after antigenic challenge under tolerogenic conditions. Our findings suggest that the conversion of effector memory Tregs may be important during development of anemia in MDS patients with autoimmune-mediated bone marrow failure.

Small populations of CD55−CD59− blood cells are detectable in approximately 50% of all acquired aplastic anemia (AA) patients and the presence of such PNH-type cells are also associated with a good response to immunosuppressive therapy (Sugimori C, et al. Blood 2006). In most patients showing 0.1% to 1.0% PNH-type cells at the diagnosis of AA, the PNH-type cell proportion remains unchanged over 3 years even after successfully responding to immunosuppressive therapy (Mochizuki K, et al. ASH 2006). Although these findings suggest that small populations of PNH-type cells are derived from a limited number of PIG-A mutants without any proliferative advantage, this hypothesis has not yet been verified at the molecular level. To appropriately address this issue, we studied 3 patients with AA who showed 0.14 to 1.6% PNH-type granulocytes. The CD55−CD59− granulocytes were sorted from these patients 2 different times at a minimum of 6 month intervals and then they were subjected to a PIG-A gene analysis. Five exons were amplified using 6 different primer sets and each amplified product was then subcloned into E. coli . At least 5 transformed clones for each amplified product were randomly plucked and subjected to sequencing. Single mutations were thereafter detected in all 3 patients as shown in Table 1. The same single mutations were then detected in the CD55−CD59− granulocytes from the patient obtained 6 months after the first examination. Two patients were in a state of hematologic remission at from 1 to 7 years after the first examination of their peripheral blood and the proportion of PNH-type cells remained stable during this period in all patients. Response to immunosuppressive therapy was not evaluable in one patient because he rejected ATG therapy. These findings indicate that PNH-type granulocytes from patients with AA are therefore clonal populations derived from single hematopoietic stem cells (HSCs) with a PIG-A mutation. If an AA patient has many HSCs with PIG-A mutations before the development of AA, then the immune system attack against HSCs should allow for the survival of the PIG-A mutants leading to the generation of a polyclonal PNH-type cell population. The presence of clonal PNH-type cells at the time of AA diagnosis suggests that the number of HSCs with a PIG-A mutation in healthy individuals may therefore be much lower than we expected and the paucity of PIG-A mutant HSCs may therefore account for the absence of increased number of PNH-type cells in approximately 20% of all AA patients who apparently respond to immunosuppressive therapy. | Patient | Age | Gender | Proportion of PNH-type granulocytes | PIG-A mutation | Response to IST | |:------- | --- | ------ | ----------------------------------- | ------------------------ | --------------- | ------------- | | 1 | 64 | M | 0.147% | 593 bp (exon 2) | T insertion | PR | | 2 | 82 | M | 1.629% | 3′splice site (intron 1) | G to A change | PR | | 3 | 76 | M | 0.161% | 276 bp (exon 2) | G deletion | not evaluable | Table 1

Acquired aplastic anemia (AA) is thought to be caused by the immune system attack against hematopoietic stem cells. However, there is no direct evidence that an immune system attack against normal hematopoietic stem cells leads to development of AA. The immune system attack may be directed toward abnormal stem cells given the fact that some patients with myelodysplastic syndrome respond to immunosuppressive therapy. Although the presence of a small population of CD55−CD59− blood cells represents a reliable marker for the immune pathophysiology of AA, little is known regarding when and how such paroxysmal nocturnal hemoglobinuria (PNH)-type cells appear in patients with AA. The development of AA with a small population of PNH-type cells was recently observed in an allogeneic stem cell transplant (SCT) recipient. This patient, a 59-year-old male, who had been treated with allogeneic peripheral blood stem cell transplantation (PBSCT) from an HLA-compatible sibling for treatment of very severe AA in March 2002, developed severe pancytopenia in December 2005. Late graft failure (LGF) without residual recipient cells was diagnosed based on the results of a chimerism analysis. Sensitive flow cytometry failed to reveal any increase in the proportion of CD55−CD59− PNH-type blood cells. The patient underwent a second PBSCT from the original donor without preconditioning in February 2006. Although his pancytopenia was completely resolved by day 20, his blood counts gradually decreased from day 60 without any apparent complications. Flow cytometry revealed small populations of PNH-type granulocytes in his peripheral blood (Figure 1). Both the PNH-type and normal phenotype granulocytes were of donor origin. PIG-A gene analyses showed the PNH-type granulocytes in the patient to be a clonal stem cell with an insertion of thymine at position 593 (codon 198). Similar results were obtained from the sorted PNH-type granulocytes obtained 6 months later. The patient was treated with horse antithymocyte globulin and cyclosporine. The patient required no further transfusions after 88 days of the therapy and remains well as of August, 2007. The small population of PNH-type cells was not detectable in any of 50 SCT recipients showing stable engraftment or in an AA patient suffering graft rejection after a SCT. These findings suggest that some factors expressed by the patient induced an immune system attack against autologous hematopoietic cells, leading to de novo development of donor-cell derived AA. This is the first evidence that an immune system attack against normal hematopoietic stem cells results in AA associated with a clonal expansion of a PIG-A mutant which may originally be present in the donor bone marrow. Figure Figure

Several studies revealed that a small number of glycosylphosphatidylinositol anchored protein-deficient cells are detectable in peripheral blood of healthy individuals but these PNH-type cells do not persist for a long time because they are originated from committed progenitor cells rather than primitive stem cells with PIG-A gene mutations. We have studied more than 700 patients with AA for the presence of CD55-CD59- granulocytes and red blood cells using highly sensitive flow cytometry and detected 0.003% or more PNH-type cells in about 50% of patients. In patients whose proportions of PNH-type are very low, the appearance of PNH-type cells may also be transient because they are not derived from true PIG-A mutant stem cell clones. In an attempt to characterize hematopoietic clones from which small populations of PNH-type cells are originated, we studied changes in the proportion of PNH-type granulocytes, clinical findings, and PIG-A mutations over 3–7 years on 52 AA patients (18 males and 33 females). The initial examination revealed less than 0.1% (0–0.075%) PNH-type granulocytes in 21 patients (Group A) and more than 0.1% (0.101–36%) PNH-type granulocytes in 31 patients (Group B). In three (14%) patients of Group A, PNH-type granulocytes became undetectable 2.9, 1.3 and 1.6 years after the initial examination. 0.007 % granulocytes became detectable 2.5 years after the first examination which produced a negative result in one patient. Among the other 17 (81%) patients of Group A, proportions of PNH-type granulocytes remained stable at 0.03–0.41% over 3–6 years. By contrast, 5 (16%) patients of Group B showed an apparent increase in the proportion of PNH-type cells (from 1.46%, 1.63%, 3.31%, 23%, and 36% to 79%, 21%, 83%, 55%, and 78%, respectively) 2 to 4 years after the initial examination. Four of the five developed clinical signs of hemolysis. The proportion of PNH-type cells in Group B patients remained stable in 33 (63%) patients and declined in 12 (23%) patients. When we sorted a small (0.04%) population of PNH-type granulocytes from a patients of Group A and analyzed PIG-A gene, only one point mutation at position 479, C to T was revealed in the exon 2. The finding that the proportion of PNH-type cells remained stable over 3 years in most Group A patients suggests that very low proportions of PNH-type cells are derived from primitive hematopoietic stem cell clones with PIG-A mutation. Among AA patients displaying small populations of PNH-type cells, only those whose proportion of PNH-type cells are more than 1% may be at a high risk of developing hemolytic PNH.

It is well known that HLA-DRB1 alleles corresponding to HLA-DR15 such as DRB1*1501 and DRB1*1502 are associated with susceptibility to acquired aplastic anemia (AA) and paroxysmal nocturnal hemoglobinuria (PNH). However, little is known about how these DRB1 alleles contribute to the development of AA. We previously reported that DRB1*1501 is significantly associated with both an increase in the proportion of PNH-type cells and good response to a immunosuppressive therapy (IST) including antithymocyte globulin (ATG) plus cyslosporine (CsA) or CsA alone, while DRB1*1502, whose DR molecules differ from those derived from DRB1*1501 in only one amino acid, is not associated with either of the two factors. To gain insight into the reason for different contributions to the development of AA by the two DRB1 alleles, we determined DRB1 alleles and the proportion of PNH-type cells of 77 patients with recently diagnosed AA and analyzed the relationship of DRB1 alleles and the presence of increased PNH-type cells with response to ATG plus CsA therapy. Fifty-five of 77 patients (71%) improved with ATG plus CsA therapy. When the factors favorably affecting the response to IST in the AA patients were examined under a multivariate analysis, only the presence of PNH-type cells was significantly associated with the response to IST. Kaplan-Meier curves showed that there were significant differences in the probability of the response to IST between the DRB1*1501 + 1502 − patients and either the DRB1*1501 − 1502 + patients (P Figure Figure

Although there is a large body of evidence that immune mechanisms play an essential role in the development of acquired aplastic anemia (AA), knowledge of autoantigens eliciting the immune system attack is limited. In an immune-mediated subset of AA patients characterized by the presence of HLA-DRB1*1501 and an increase in the proportion of paroxysmal nocturnal hemoglobinuria (PNH)-type blood cells, not only T-cell response but also B-cell response to antigens derived from hematopoietic stem cells may take place. To test this hypothesis, we screened sera of such immune-mediated AA patients for the presence of IgG antibodies recognizing a megakaryoblastic leukemia cell line, UT-7, using immunofluorescence staining. Since the sera from several patients stained cytoplasm of UT-7 cells, lysates of the cell line were subjected to Western blotting using the same sera. A clear band of 80 kDa protein was revealed by the sera from several patients. Peptide mass finger printing of the protein extracted from the single band identified this protein as moesin. Moesin is a membrane-cytoskeleton linker protein in cytoplasm and involved in the formation of microvilli, cell adhesion sites, and ruffling membranes. When sera of a large number of AA patients were screened for the presence of antibodies specific to recombinant moesin protein with Western blotting and ELISA, significantly high titer of the antibodies was detected in 35 of 110 (32%) AA patients. The prevalence of anti-moesin antibodies in AA patients showing increased PNH-type cells (41%) was significantly higher than those (12%) without PNH-type cells ( p =0.002). Anti-moesin antibodies were detectable in 15 of 39 (38%) recently diagnosed AA patients before therapy and in 20 of 71 (28%) patients who had a long history of AA. Four of 6 newly diagnosed patients showing anti-moesin antibodies responded to ATG+cyclosporine therapy and restored good hematopoietic function. Since B lymphocytes reportedly secrete moesin as a form of exosome (J Biol Chem, 2003), we examined supernatants of various leukemia cell lines incubated for 1 hour for the presence of moesin. The membrane-linker protein was detectable in the culture supernatant of 4 (UT-7, K562, OUN-1 and NB4) of 10 myeloid leukemia cell lines. These findings indicate that B-cell response to moesin may represent immune pathophysiology of AA and that moesin secreted from hematopoietic progenitor cells as a form of exosome may be efficiently processed by antigen-presenting cells, leading to induction of antibodies and possibly T cells specific to moesin in immune-mediated AA.

To determine the clinical significance of a minor population of glycosylphosphatidylinositol-anchored membrane protein-negative (GPI-AP−) cells in acquired aplastic anemia (AA), we quantified CD55− CD59− granulocytes and red blood cells (RBCs) in peripheral blood from 122 patients with recently diagnosed AA and correlated numbers of GPI-AP− cells and responses to immunosuppressive therapy (IST). The highly sensitive flow cytometry detected 0.005% to 23.1% of GPI-AP− cells in 68.0% of AA patients. Sixty-eight of 83 (91%) patients with an increased ratio of GPI-AP− cells (PNH+) responded to antithymocyte globulin (ATG) + cyclosporin (CsA) therapy, whereas 18 of 39 (48%) without such an increase (PNH−) responded. Failure-free survival rates were significantly higher (64%) among PNH+ than PNH− patients (12%) at 5 years (Figure) although overall survival rates were comparable between the groups. Numbers of GPI-AP− and GPI-AP+ cells increased in parallel among most PNH+ patients who responded to IST, suggesting that these cells are equally sensitive to immune attack. These results indicate that a minor population of GPI-AP− cells represents a reliable marker of a positive IST response and a favorable prognosis among patients with AA. Furthermore, immune attack against hematopoietic stem cells that allows PNH clonal expansion might occur only at the onset of AA. Figure Figure

Modern flow cytometry techniques using monoclonal antibodies (mAbs) specific to glycosylphosphatidylinositol-anchored membrane proteins have revealed an increase in the proportion of paroxysmal nocturnal hemoglobinuria (PNH)-type cells in a varying (30 to 50) percentage of acquired aplastic anemia (AA) patients who do not manifest any signs of hemolysis. However, the clinical significance of such a small proportion of PNH-type cells remains uncertain because previous studies have used a variety of flow cytometric assays with different mAbs and different definitions of an abnormal increase in the proportion of PNH-type cells, and also, few studies have focused on the correlation of presence of increased PNH-type cells with response to immunosuppressive therapy (IST) based on a large number of newly diagnosed AA patients. We previously demonstrated, using two-color flow cytometry capable of precisely determining the proportion of PNH-type cells less than 0.1%, that such a minor population of PNH-type cells represents a reliable marker for presence of immune pathophysiology in refractory anemia patients (Blood,100: 3897–3902, 2002; Blood,102:1211–1216, 2003). To determine if the PNH-type cells serve as a marker for immune pathogenesis in AA as well, we examined peripheral blood in 99 newly diagnosed AA patients (67 severe AA and 32 moderate AA) for presence of CD55−59−CD11b+ granulocytes and CD55−59−glycophorin A+ RBCs and studied correlation of response to ATG + cyclosporine (CsA) therapy with presence of increased PNH-type cells. When the influence of time from blood collection until treatments of cells with mAbs on the outcome of PNH-type-cell detection were examined, the percentages of PNH-type cells remained unchanged for 2 days for granulocytes and for 6 days for RBCs from the time of blood collection. Minor (>0.003% for CD11b+ granulocytes and >0.005% for glycophorin A+ RBCs) population of PNH-type cells were detected in 66 (66.7%) patients. The proportions of PNH-type cells in patients with increased PNH-type cells (PNH+ patients) were 0.005–0.01% in 10 (15.2%) patients, 0.01–0.1% in 20 (30.3%), 0.1–1.0% in 25 (37.9%), and 1.0–8.67% in 10 (15.2%). When clinical characteristics of PNH+ patients were compared with that of PNH− patients, PNH+ patients (mean, 54 year-old) were significantly older than PNH− patients (mean, 42 year-old, [Italic]P[/Italic]=0.042). 55 of 66 (83.3%) PNH+ patients improved with IST and achieved PR or CR whereas 17 of 33 (52.9%) PNH− patients responded. Kaplan-Meier analysis showed that PNH+ patients had a significantly better chance of response to IST than PNH- patients (Figure). These results indicate that a minor population of PNH-type cells defined by the two-color flow cytometry represents a good marker for response to IST in AA patients. Since CD55−59−glycophorin A+ RBCs less than 0.1% remain constant and detectable in blood stored at 4 ?C for up to 6 days, testing blood cells, particularly RBCs, using this sensitive flow cytometry is useful in identifying high responders to IST in AA patients. Figure Figure

Treatment of patients refractory to immunosuppressive therapy is a major problem in the management of aplastic anemia (AA). Some patients who are ineligible for allogeneic stem cell transplantation have been successfully treated with anabolic steroids although little is known about the common characteristics of these patients. In order to characterize high responders to danazol, we retrospectively studied 44 AA (13 male and 31 female) patients who were treated with 300 mg of danazol daily for at least 3 months. All patients had been previously treated with ATG+CsA or CsA alone without appreciable effects before danazol therapy. Peripheral blood of these patients was examined for the presence of PNH-type cells and clonality in granulocytes using sensitive flow cytometry (Blood100: 3897, 2002) and the improved HUMARA assay (Blood102: 1211, 2003). 24 of 44 (55%) patients attained PR or CR according to the response criteria proposed by Camitta. All 3 lineage cells increased in the responders. The rate of response in female patients (21/31, 68%) was significantly higher than that (3/13, 23%) in male patients (P=0.0058). There was no difference in the response rate between older (>50 years) patients (75%) and younger (

A large body of evidence has suggested that acquired aplastic anemia (AA) of patients carrying HLA-DR15 is a kind of organ-specific autoimmune disease where hematopoietic progenitor cells in bone marrow are attacked by CD4+ T cells recognizing endogenous antigens. We recently identified diazepam-binding inhibitor-related protein 1 (DRS-1) as a candidate autoantigen capable of provoking immune system attack against hematopoietic progenitor cells in AA (Blood, 2004). Although in other organ-specific autoimmune diseases such as insulin-dependent diabetes mellitus and primary biliary cirrhosis, cytoplasmic proteins including glutamic acid decarboxylase 65 and pyruvate dehydrogenase complex have been shown to serve as autoantigens and mediate organ damages by CD4+ T cells, it remains unclear whether a peroxisomal protein like DRS-1 can be processed in hematopoietic progenitor cells, presented by HLA-DR15, and eventually serve as a target antigen of specific CD4+ T cells, leading to killing of hematopoietic progenitor cells themselves. To clarify these issues, we established a CD4+ T-cell line specific to a DRS-1 peptide (amino acid residues 191–204) from an AA patient carrying HLA-DR15 who had exhibited a high titer of anti-DRS-1 antibody as well as a high frequency of T-cell precursors specific to DRS-1, and then examined the cytotoxicity of the DRS-1-specific T-cell line against (1) autologous lymphoblastoid cell line (LCL) cells transfected with full length DRS-1 cDNA using a lentiviral vector, (2) myeloid leukemia cell lines carrying HLA-DR15 (KH88 and SAS413) and a leukemia cell line not carrying HLA-DR15 (K562), and (3) CD34+ progenitor cells from normal individuals. When all leukemia cell lines and LCL cells were examined for DRS-1 expression using Western blotting with specific monoclonal antibodies, DRS-1 protein was detected in DRS-1-transfected LCL cells, KH88 and K562, but not in nontransfected LCL cells and SAS413. Overexpression of DRS-1 gene by the CD34+ cells from normal individuals was ascertained by real-time PCR. In the 51Cr release assay, DRS-1-specific T cells showed cytotoxicity against only DRS-1-transfected LCL cells and KH88 in a dose-dependent manner (Figure), indicating that the T cell line requires presence of both DRS-1 and HLA-DR15 on target cells to exert cytotoxicity. When the DRS-1-specific T cells were incubated with CD34+ cells isolated from normal individuals with or without HLA-DR15 at an 10:1 ratio for 4 hours and cultured in a methylcellulose medium supplemented with colony-stimulating factors, the numbers of CFU-GM and BFU-E colonies derived from an HLA-DR15+ individual were 60.0% and 52.9% of a control whereas those derived from an HLA-DR15− individual were 90.1% and 88.2%. These findings indicate that hematopoietic progenitor cells in individuals with HLA-DR15 can present DRS-1 through the DR molecule and a breakdown of immune tolerance to DRS-1 may lead to development of AA. Download : Download high-res image (173KB) Download : Download full-size image

Some patients with aplastic anemia (AA) and approximately 40% of patients with refractory anemia (RA) of myelodysplastic syndrome exhibit karyotypic abnormalities in bone marrow dividing cells. Although some of the patients undergo evolution to acute myeloid leukemia (AML), others follow a clinical course similar to AA patients without chromosomal abnormalities. Except for several abnormalities such as −7 and 5q-, the clinical significance of such chromosomal abnormalities in bone marrow failure patients remains unclear. We recently developed a reliable HUMARA assay capable of detecting a clonal population in granulocytes which constitutes 30% or more of total granulocytes (Blood. 2003;102:1211–1216). Studying correlation between chromosomal abnormalities and the presence of clonality may help in understanding the pathogenetic role of chromosomal abnormalities in AA and RA. We thus analyzed 50 acquired AA and 28 RA female patients who were heterozygous for the HUMARA gene. Chromosomal abnormalities such as add(5)(q13), 9q–9q+ and del(7)(q14q22) were found in 8% of AA and 21% of RA patients. Clonality was detected in 38% of AA patients and 39% of RA patients. Incidence of chromosomal abnormalities in patients with clonality (27%) was higher than that in patients without clonality (4%, p