- Email: [email protected]
136 Using flow cytometry for differential diagnosis of thrombocytopenia
S52
Posters / Leukemia Research 35 (2011) S27–S142
Conclusions: the analysis of BM subsets of CD34+ cells is useful in the differential diagnosis of MDS with BM blasts < 5% and non-clonal disorders. The most useful features are the number of B-cell precursors and aberrant co-expressions, but not asynchronous maturation. Support: MDS Foundation, FAPESP and CNPq 134 Hidden genomic abnormalities detected in MDS patients with a normal karyotype have prognostic relevance B. Royer-Pokora1 , A. Thiel1 , M. Beier1 , D. Ingenhag1 , K. Servan1 , M. Hein1 , U. Germing2 , B. Hildebrandt1 . 1 Institute of Human Genetics, 2 Department of Haematology, Oncology and Clinical Immunology, Heinrich Heine University Duesseldorf, Duesseldorf, Germany Approximately 50% of MDS patients have a normal karyotype. Our aim was to study whether recurrent or individual abnormalities hidden can be detected in MDS with a normal Karyotype and to determine their prognostic significance. To address this question we have analyzed 107 MDS patients of various FAB/WHO subgroups and a normal karyotype using aCGH. For this study we have used Agilent microarrays consisting of 44K, 105K or 244K 60-mer oligonucleotide probes spanning the human genome with an average spatial resolution of approximately 43, 22 and 9 kb, respectively. Array CGH identified hidden recurrent chromosomal aberrations in ≥2 cases: del5q, del7q, del21q and del4q. In 34 cases we have identified individual imbalances. Of the 107 cases 42 (39%) had hidden alterations. Larger alterations were verified by FISH and smaller alterations with Q-PCR. So far 40 of the identified aberrations were verified with other methods. Zoom in custom arrays were designed to characterize the breakpoints. In the imbalances several interesting genes were localized, involved in epigenetic and chromatin modification, such as DNMT3a, ASXL2, TET2, HDAC3, BRD8 or several genes with a role in DNA repair, e.g. DNAJC 16, 18 and 27, MLH3, RAD50, as well as members of the RAS signalling pathway, for example RAB10, 38, 39, RABL5, RAPGEF6 and RIN3. Gene functional classification using DAVID revealed an enrichment of metallopeptidases, Apoptosis genes and genes regulating haematopoiesis. Comparing the survival of patients with and without additional hidden alterations showed that those with additional genomic imbalances had an inferior survival (p = 0.002). In conclusion, hidden recurrent and individual aberrations in cytogenetically normal MDS can be identified with aCGH and these have prognostic relevance. 135 Lower bone marrow myeloid and plasmacytoid dendritic cell counts in high-risk than in low-risk MDS patients and controls 1 2 L. Saft1 , E. Bjorklund ¨ , E. Hellstrom-Lindberg ¨ , A. Porwit1 . 1 Department of Pathology, 2 Department of Medicine, Center for Experimental Hematology, Karolinska University Hospital and Institute, Stockholm, Sweden
Background: Dendritic cells (DC) are professional antigenpresenting cells and play a pivotal role in coordinating functions of the immune system by the initiation and regulation of T-cell mediated responses. Laboratory and clinical studies suggest that bone marrow (BM) failure in MDS may be, at least in part, immune-mediated. The observation that MDS patients can respond to immunosuppressive therapy sparked a renewed interest in T-cell mediated mechanisms. DCs originate from BM hematopoietic stem cells and clonal involvement of precursor DC has been demonstrated in both MDS and AML. Aims: To determine whether levels of myeloid (mDC) and plasmacytoid (pDC) dendritic cells in BM samples from MDS patients differ from those in reactive BM. Patients and Methods: BM samples from eighty MDS patients (IPSS: low-risk 25, INT-1 39, INT-2 11, high-risk 5, all WHO subtypes included) diagnosed at the Karolinska University Hospital between
January 2006 and November 2010 were included. In 67 patients (84%) the BM samples were taken before treatment and in 20/67 patients at least one additional follow-up biopsy was available prior to treatment start. BM samples from 13/80 (16%) patients were analyzed after therapy start. DC levels (expressed as percentage of BM cells) were measured by 4-color flow cytometry (FCM) using Lineage cocktail 1(lin)− FITC/CD123-PE/HLA-DR-PerCP/CD11cAPC antibody combination, FACS-Calibur flow cytometer and Paint-a Gate Pro software (Becton&Dickinson). pDC immunophenotype was characterized as lin−/HLA−DR+/CD123+/CD11c− and mDC as lin−/HLA−DR+/CD11c+/CD123−. In addition, BM samples from 36 hospital controls were investigated. Results: Average DC levels were significantly lower in MDS (mDC 0.039%; pDC 0.118%) as compared to the control group (mDC 0.089%; pDC 0.178%; p < 0.001). Lower DC levels were found in treated MDS patients than in untreated patients (p < 0.01). High-risk MDS and patients with higher BM blast percentages (>5%) had significantly lower DC levels than low-risk MDS patients. Progression to AML was seen in 15/80 (18.7%) patients (one del(5q), 6 RCMD, 5 RAEB-1, 3 RAEB-2); no significant difference in DC levels at diagnosis were seen in patients who developed AML in the follow-up period but in sequentially analyzed patients we could see lower DC levels at progression. Conclusions: DC levels are significantly lower in high-risk MDS and inversely correlated with the percentage of BM blasts. Our findings may indicate involvement of the DC compartment in the MDS disease process. FCM-determined DC levels may assist in the diagnostic work-up and risk-assessment of MDS patients. 136 Using flow cytometry for differential diagnosis of thrombocytopenia D. Subira, D. de Miguel, I. San Roman, ´ M. D´ıaz Morfa, N. Golbano, D. Morales, J. Arbeteta, S. Herrero, F. Fuertes, B. Pinedo. Hematology, Hospital Universitario de Guadalajara, Guadalajara, Spain Differential diagnosis between immune thrombocytopenia (ITP) and myelodysplastic syndrome (MDS) may be difficult, especially in old people. Altered immunophenotypes in maturing cells have been described in bone marrow samples from patients with MDS. However, flow cytometry immunophenotyping (FCI) detractors criticize lack of standard procedures of analyses and subjective identification of deviation from normal patterns of differentiation. Aim: To determine the utility of semi-quantitative FCI for identifying MDS within a group of patients with thrombocytopenia. Patients: From May 2010 to January 2011, 20 patients (9M/11F, median age 72, range 16–89) were studied for differential diagnosis of thrombocytopenia. Median platelet count was 45×109 (range 3.5–118); median hemoglobin was 12 g/dl, and leukocyte 6.4×109 /l. The following diagnoses were established: ITP (n = 6), refractory ITP (n = 1); MDS (n = 6); non-MDS thrombocytopenia (n = 7). Methods: FCI using 4-colour combinations. Data studied in bone marrow samples were: myeloid granularity, abnormalities in the phenotypic pattern of maturation of the myeloid (CD16/ CD13/HLADR/CD11b) and monocytic (CD36/CD14/HLADR/CD64) compartment, distribution of CD10 on mature granulocytes, quantification of CD10+ B-cell precursors and myeloid CD34+ cells. Deviations from normal patterns were calculated using reference images with the Infinicyt software program for FCI analysis. The following deviations were considered as abnormalities: >30% in myeloid granularity, >19% in myeloid maturation, >11% in monocytic maturation, < 70% of CD10 mean fluorescence intensity on granulocytes, <1% of CD10+ lymphocytes within the B-cell compartment, and >2% CD34+ myeloid cells. Results: Median number of FCI abnormalities was significantly higher in patients with MDS (n = 3.5) as compared to patients with ITP (n = 0.4) and non-MDS thrombocytopenia (n = 0.6). These
Posters / Leukemia Research 35 (2011) S27–S142
results were not biased by the age of patients. The most informative parameters for discrimination between MDS and non-MDS were percentage of CD34+myeloid cells and CD10+B-cells, and myeloid and monocytic maturation (Table). Conclusions: With the number of patients studied, FCI seems a promising tool for differential diagnosis of thrombocytopenia. Despite the absence of megakaryocytic parameters to study, abnormalities in the myelo-monocytic compartment, and B-cell and myeloid precursors are useful to suspect MDS. Introduction of new software programs for quantitative measurements of deviations improves the objectivity of FCI analyses, but the initial selection of populations still comprises a subjective approach. Table 1
ITP (n = 7) Non-MDS (n = 7) MDS (n = 6)
Abnormal SSC/CD45
Abnormal myeloid maturation
Abnormal monocytic maturation
CD10− mature granulocytes
<1% CD10+ B-cells
>2% CD34+ myeloid cells
Median number of abnormalities
0 1 2
2 1 5
0 3 5
0 0 2
1 1 4
0 0 3
0.43 0.57 3.5
137 Standardization of flow cytometry in myelodysplastic syndromes: A report from an international consortium and the European LeukemiaNet Working Group T.M. Westers1 , R. Ireland2 , W. Kern3 , C. Alhan1 , J.S. Balleisen4 , M.C. Ben ´ e´ 5 , P. Bettelheim6 , K. Burbury7 , M. Cullen8 , J.A. Cutler9 , M.G. Della Porta10,11 , A.M. Drager1 , J. Feuillard12 , P. Font13 , U. Germing14 , D. Haase15 , U. Johansson16 , S. Kordasti2 , M.R. Loken9 , L. Malcovati10 , J.G. te Marvelde17 , S. Matarraz18 , T. Milne2 , B. Moshaver19 , G.J. Mufti2 , K. Ogata20 , A. Orfao18 , A. Porwit21 , K. Psarra22 , S.J. Richards8 , D. Subira´ 23 , V. Tindell2 , T. Vallespi24 , P. Valent6 , V.H. van der Velden17 , T.M. de Witte19 , D.A. Wells9 , F. Zettl15 , A.A. van de Loosdrecht1 . 1 Hematology, VU University Medical Center, Amsterdam, The Netherlands; 2 King’s College Hospital, London, UK; 3 MLL Munich Leukemia Laboratory, Munich, 4 Hematology Oncology and Clinical Immunology, Heinrich-Heine-University, D¨ usseldorf, Germany; 5 Medicine&CHU, Nancy Universite, Nancy, France; 6 Internal Medicine, Medical University of Vienna, Vienna, Austria; 7 Hematology, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia; 8 HMDS, St James’s University Hospital, Leeds, UK; 9 Hematologics, Seattle, WA, USA; 10 Hematology and Oncology, Fondazione IRCCS Policlinico San Matteo, 11 University of Pavia, Pavia, Italy; 12 Laboratoire d’Hematologie, CHU Dupuytren, Limoges, France; 13 Hematology, Hospital General de Valencia, Universitario Gregorio Maranon, Madrid, Spain; 14 Hematlogy Oncology and Clinical Immunology, Heinrich-Heine-University, D¨ usseldorf, 15 Hematology and Oncology, Georg-August-University, G¨ ottingen, Germany; 16 Haematology, University Hospitals NHS Foundation Trust, Bristol, UK; 17 Immunology, Erasmus Medical Center, Rotterdam, The Netherlands; 18 Servicio Central de Citometr´ıa, Centro de Investigaci´ on del C´ ancer, Instituto de Biologia Celular y Molecular del C´ ancer (CSIC/USAL) and Department of Medicine, Universidad de Salamanca, Salamanca, Spain; 19 Hematology, St Radboud University Medical Center, Nijmegen, The Netherlands; 20 Hematology, Department of Medicine, Nippon Medical School, Tokyo, Japan; 21 Pathology, Karolinska University Hospital, Stockholm, Sweden; 22 Immunology-Histocompatibility, Evangelismos Hospital, Athens, Greece; 23 Hematology, Fundaci´ on Jim´enez D´ıaz, Madrid, 24 Hematology, Hospital Universitario, Vall d’Hebron, Barcelona, Spain Flow cytometry (FC) is increasingly recognized as new forthcoming standard in the diagnosis and prognosis of myelodysplastic syndromes (MDS). Validation of current FC assays and agreements upon the technique are requisites for future clinical application. A working group within the European LeukemiaNet (ELN) was formed in 2008 to discuss and propose standards for FC in MDS bone marrow analysis (Haematologica, 2009). Thus far, three meetings were
S53
convened (Amsterdam, Munich and London) with representatives from 22, mainly European institutes. Major goals of the working conferences were: a. to define the role of FC in the diagnosis and prognosis of MDS related to currently validated systems (including cytogenetic and molecular markers); b. to define minimal requirements to assess dysplasia by FC in known or suspected MDS; c. to define how these data should be interpreted objectively; d. to consider the specificity of FC in MDS related to a series of other (non) clonal hematological diseases. The group concluded that standardization of FC in MDS may not only contribute to the improvement of diagnosis in MDS and in patients with unexplained cytopenias, but it may also refine prognostic scoring systems. That is, FC may identify MDS subgroups with an altered disease course not determined by current methods. Additionally, replicate FC assessments are highly recommended, not only in inconclusive cases, but also to monitor the course of the disease in untreated, mainly low risk MDS patients, and to monitor treatment with currently available drugs. Since MDS is a heterogeneous disease, it is unlikely that a single marker supports its diagnosis; consequently an accumulation of FC aberrancies might confer the diagnosis of MDS. Minimal requirements to analyze dysplasia by FC were defined. The proposed core markers should enable a categorization of the results of FC analysis as normal or as (possibly) in agreement with MDS. A FC report should include a description of validated FC abnormalities such as aberrant marker expression on myeloid progenitor cells (e.g. lineage infidelity marker expression); in addition, dysgranulopoiesis and dysmonocytopoiesis might be scored if ≥2 abnormalities are evident. The working group is dedicated to initiate further studies to establish robust FC diagnostic and prognostic markers and marker patterns in MDS, with the ultimate goal to refine and improve diagnosis and prognostic scoring systems. Nevertheless, it is stressed that FC analysis should be part of the integrated diagnosis rather than a separate technique. 138 A preliminary study on the value of periodic acid-Schiff stain in erythroblasts in myelodysplastic syndrome Z. Xiao, L. Liu. Institute of Hematology and Hospital of Blood Diseases, Chinese Academe of Medical Sciences, Tianjin, China Objective: To investigate the value of periodic acid-Schiff (PAS) stain in erythroblasts in judgement of dyserythropoiesis, diagnosis and differential diagnosis of myelodysplastic syndrome (MDS). Methods: The PAS stain in bone marrow (BM) erythroblasts in 406 MDS pateints, 207 non-severe aplastic anemia (NSAA), 144 immune thrombocytopenic purpura (ITP), 67 megaloblastic anemia (MegA), 76 iron deficiency anemia (IDA), 50 paroxysmal nocturnal hemoglobinuria (PNH), 50 acute erythroid leukemia (AEL) and other related laboratory tests data in MDS patients were analyzed retrospectively. Results: The PAS-positive detection rate in MDS (53.0%) was significantly higher than in NSAA (14.5%), ITP (27.1%) and PNH (16.0%), but significantly lower than in AEL (84.0%) (All the P = 0.000). There was no significant difference in the PAS-positive detection between MDS and MegA (46.3%), or MDS and IDA (40.8%) (P = 0.310; P = 0.052). The PAS-positive rate (Median, M = 1%) and PAS-positive scores (M = 2) in erythroblasts in MDS was significantly lower than in AEL (M = 8%; M = 17), and significantly higher than in NSAA (M = 0%; M = 0), ITP (M = 0%; M = 0), PNH (M = 0%; M = 0), MegA (M = 0%; M = 0), and IDA (M = 0%; M = 0) (All the P < 0.05). The cut-off value of PAS-positive rate and score for the diagnosis of MDS from the other groups except AEL were 0.5% and 0.5, whose sensitivity and specificity were 60.8% and 74.4% respectively. MDS patients were divided into PAS-positive and PAS-negative groups according to PAS reaction, the percentage of erythroid cell in BM was
Posters / Leukemia Research 35 (2011) S27–S142
Conclusions: the analysis of BM subsets of CD34+ cells is useful in the differential diagnosis of MDS with BM blasts < 5% and non-clonal disorders. The most useful features are the number of B-cell precursors and aberrant co-expressions, but not asynchronous maturation. Support: MDS Foundation, FAPESP and CNPq 134 Hidden genomic abnormalities detected in MDS patients with a normal karyotype have prognostic relevance B. Royer-Pokora1 , A. Thiel1 , M. Beier1 , D. Ingenhag1 , K. Servan1 , M. Hein1 , U. Germing2 , B. Hildebrandt1 . 1 Institute of Human Genetics, 2 Department of Haematology, Oncology and Clinical Immunology, Heinrich Heine University Duesseldorf, Duesseldorf, Germany Approximately 50% of MDS patients have a normal karyotype. Our aim was to study whether recurrent or individual abnormalities hidden can be detected in MDS with a normal Karyotype and to determine their prognostic significance. To address this question we have analyzed 107 MDS patients of various FAB/WHO subgroups and a normal karyotype using aCGH. For this study we have used Agilent microarrays consisting of 44K, 105K or 244K 60-mer oligonucleotide probes spanning the human genome with an average spatial resolution of approximately 43, 22 and 9 kb, respectively. Array CGH identified hidden recurrent chromosomal aberrations in ≥2 cases: del5q, del7q, del21q and del4q. In 34 cases we have identified individual imbalances. Of the 107 cases 42 (39%) had hidden alterations. Larger alterations were verified by FISH and smaller alterations with Q-PCR. So far 40 of the identified aberrations were verified with other methods. Zoom in custom arrays were designed to characterize the breakpoints. In the imbalances several interesting genes were localized, involved in epigenetic and chromatin modification, such as DNMT3a, ASXL2, TET2, HDAC3, BRD8 or several genes with a role in DNA repair, e.g. DNAJC 16, 18 and 27, MLH3, RAD50, as well as members of the RAS signalling pathway, for example RAB10, 38, 39, RABL5, RAPGEF6 and RIN3. Gene functional classification using DAVID revealed an enrichment of metallopeptidases, Apoptosis genes and genes regulating haematopoiesis. Comparing the survival of patients with and without additional hidden alterations showed that those with additional genomic imbalances had an inferior survival (p = 0.002). In conclusion, hidden recurrent and individual aberrations in cytogenetically normal MDS can be identified with aCGH and these have prognostic relevance. 135 Lower bone marrow myeloid and plasmacytoid dendritic cell counts in high-risk than in low-risk MDS patients and controls 1 2 L. Saft1 , E. Bjorklund ¨ , E. Hellstrom-Lindberg ¨ , A. Porwit1 . 1 Department of Pathology, 2 Department of Medicine, Center for Experimental Hematology, Karolinska University Hospital and Institute, Stockholm, Sweden
Background: Dendritic cells (DC) are professional antigenpresenting cells and play a pivotal role in coordinating functions of the immune system by the initiation and regulation of T-cell mediated responses. Laboratory and clinical studies suggest that bone marrow (BM) failure in MDS may be, at least in part, immune-mediated. The observation that MDS patients can respond to immunosuppressive therapy sparked a renewed interest in T-cell mediated mechanisms. DCs originate from BM hematopoietic stem cells and clonal involvement of precursor DC has been demonstrated in both MDS and AML. Aims: To determine whether levels of myeloid (mDC) and plasmacytoid (pDC) dendritic cells in BM samples from MDS patients differ from those in reactive BM. Patients and Methods: BM samples from eighty MDS patients (IPSS: low-risk 25, INT-1 39, INT-2 11, high-risk 5, all WHO subtypes included) diagnosed at the Karolinska University Hospital between
January 2006 and November 2010 were included. In 67 patients (84%) the BM samples were taken before treatment and in 20/67 patients at least one additional follow-up biopsy was available prior to treatment start. BM samples from 13/80 (16%) patients were analyzed after therapy start. DC levels (expressed as percentage of BM cells) were measured by 4-color flow cytometry (FCM) using Lineage cocktail 1(lin)− FITC/CD123-PE/HLA-DR-PerCP/CD11cAPC antibody combination, FACS-Calibur flow cytometer and Paint-a Gate Pro software (Becton&Dickinson). pDC immunophenotype was characterized as lin−/HLA−DR+/CD123+/CD11c− and mDC as lin−/HLA−DR+/CD11c+/CD123−. In addition, BM samples from 36 hospital controls were investigated. Results: Average DC levels were significantly lower in MDS (mDC 0.039%; pDC 0.118%) as compared to the control group (mDC 0.089%; pDC 0.178%; p < 0.001). Lower DC levels were found in treated MDS patients than in untreated patients (p < 0.01). High-risk MDS and patients with higher BM blast percentages (>5%) had significantly lower DC levels than low-risk MDS patients. Progression to AML was seen in 15/80 (18.7%) patients (one del(5q), 6 RCMD, 5 RAEB-1, 3 RAEB-2); no significant difference in DC levels at diagnosis were seen in patients who developed AML in the follow-up period but in sequentially analyzed patients we could see lower DC levels at progression. Conclusions: DC levels are significantly lower in high-risk MDS and inversely correlated with the percentage of BM blasts. Our findings may indicate involvement of the DC compartment in the MDS disease process. FCM-determined DC levels may assist in the diagnostic work-up and risk-assessment of MDS patients. 136 Using flow cytometry for differential diagnosis of thrombocytopenia D. Subira, D. de Miguel, I. San Roman, ´ M. D´ıaz Morfa, N. Golbano, D. Morales, J. Arbeteta, S. Herrero, F. Fuertes, B. Pinedo. Hematology, Hospital Universitario de Guadalajara, Guadalajara, Spain Differential diagnosis between immune thrombocytopenia (ITP) and myelodysplastic syndrome (MDS) may be difficult, especially in old people. Altered immunophenotypes in maturing cells have been described in bone marrow samples from patients with MDS. However, flow cytometry immunophenotyping (FCI) detractors criticize lack of standard procedures of analyses and subjective identification of deviation from normal patterns of differentiation. Aim: To determine the utility of semi-quantitative FCI for identifying MDS within a group of patients with thrombocytopenia. Patients: From May 2010 to January 2011, 20 patients (9M/11F, median age 72, range 16–89) were studied for differential diagnosis of thrombocytopenia. Median platelet count was 45×109 (range 3.5–118); median hemoglobin was 12 g/dl, and leukocyte 6.4×109 /l. The following diagnoses were established: ITP (n = 6), refractory ITP (n = 1); MDS (n = 6); non-MDS thrombocytopenia (n = 7). Methods: FCI using 4-colour combinations. Data studied in bone marrow samples were: myeloid granularity, abnormalities in the phenotypic pattern of maturation of the myeloid (CD16/ CD13/HLADR/CD11b) and monocytic (CD36/CD14/HLADR/CD64) compartment, distribution of CD10 on mature granulocytes, quantification of CD10+ B-cell precursors and myeloid CD34+ cells. Deviations from normal patterns were calculated using reference images with the Infinicyt software program for FCI analysis. The following deviations were considered as abnormalities: >30% in myeloid granularity, >19% in myeloid maturation, >11% in monocytic maturation, < 70% of CD10 mean fluorescence intensity on granulocytes, <1% of CD10+ lymphocytes within the B-cell compartment, and >2% CD34+ myeloid cells. Results: Median number of FCI abnormalities was significantly higher in patients with MDS (n = 3.5) as compared to patients with ITP (n = 0.4) and non-MDS thrombocytopenia (n = 0.6). These
Posters / Leukemia Research 35 (2011) S27–S142
results were not biased by the age of patients. The most informative parameters for discrimination between MDS and non-MDS were percentage of CD34+myeloid cells and CD10+B-cells, and myeloid and monocytic maturation (Table). Conclusions: With the number of patients studied, FCI seems a promising tool for differential diagnosis of thrombocytopenia. Despite the absence of megakaryocytic parameters to study, abnormalities in the myelo-monocytic compartment, and B-cell and myeloid precursors are useful to suspect MDS. Introduction of new software programs for quantitative measurements of deviations improves the objectivity of FCI analyses, but the initial selection of populations still comprises a subjective approach. Table 1
ITP (n = 7) Non-MDS (n = 7) MDS (n = 6)
Abnormal SSC/CD45
Abnormal myeloid maturation
Abnormal monocytic maturation
CD10− mature granulocytes
<1% CD10+ B-cells
>2% CD34+ myeloid cells
Median number of abnormalities
0 1 2
2 1 5
0 3 5
0 0 2
1 1 4
0 0 3
0.43 0.57 3.5
137 Standardization of flow cytometry in myelodysplastic syndromes: A report from an international consortium and the European LeukemiaNet Working Group T.M. Westers1 , R. Ireland2 , W. Kern3 , C. Alhan1 , J.S. Balleisen4 , M.C. Ben ´ e´ 5 , P. Bettelheim6 , K. Burbury7 , M. Cullen8 , J.A. Cutler9 , M.G. Della Porta10,11 , A.M. Drager1 , J. Feuillard12 , P. Font13 , U. Germing14 , D. Haase15 , U. Johansson16 , S. Kordasti2 , M.R. Loken9 , L. Malcovati10 , J.G. te Marvelde17 , S. Matarraz18 , T. Milne2 , B. Moshaver19 , G.J. Mufti2 , K. Ogata20 , A. Orfao18 , A. Porwit21 , K. Psarra22 , S.J. Richards8 , D. Subira´ 23 , V. Tindell2 , T. Vallespi24 , P. Valent6 , V.H. van der Velden17 , T.M. de Witte19 , D.A. Wells9 , F. Zettl15 , A.A. van de Loosdrecht1 . 1 Hematology, VU University Medical Center, Amsterdam, The Netherlands; 2 King’s College Hospital, London, UK; 3 MLL Munich Leukemia Laboratory, Munich, 4 Hematology Oncology and Clinical Immunology, Heinrich-Heine-University, D¨ usseldorf, Germany; 5 Medicine&CHU, Nancy Universite, Nancy, France; 6 Internal Medicine, Medical University of Vienna, Vienna, Austria; 7 Hematology, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia; 8 HMDS, St James’s University Hospital, Leeds, UK; 9 Hematologics, Seattle, WA, USA; 10 Hematology and Oncology, Fondazione IRCCS Policlinico San Matteo, 11 University of Pavia, Pavia, Italy; 12 Laboratoire d’Hematologie, CHU Dupuytren, Limoges, France; 13 Hematology, Hospital General de Valencia, Universitario Gregorio Maranon, Madrid, Spain; 14 Hematlogy Oncology and Clinical Immunology, Heinrich-Heine-University, D¨ usseldorf, 15 Hematology and Oncology, Georg-August-University, G¨ ottingen, Germany; 16 Haematology, University Hospitals NHS Foundation Trust, Bristol, UK; 17 Immunology, Erasmus Medical Center, Rotterdam, The Netherlands; 18 Servicio Central de Citometr´ıa, Centro de Investigaci´ on del C´ ancer, Instituto de Biologia Celular y Molecular del C´ ancer (CSIC/USAL) and Department of Medicine, Universidad de Salamanca, Salamanca, Spain; 19 Hematology, St Radboud University Medical Center, Nijmegen, The Netherlands; 20 Hematology, Department of Medicine, Nippon Medical School, Tokyo, Japan; 21 Pathology, Karolinska University Hospital, Stockholm, Sweden; 22 Immunology-Histocompatibility, Evangelismos Hospital, Athens, Greece; 23 Hematology, Fundaci´ on Jim´enez D´ıaz, Madrid, 24 Hematology, Hospital Universitario, Vall d’Hebron, Barcelona, Spain Flow cytometry (FC) is increasingly recognized as new forthcoming standard in the diagnosis and prognosis of myelodysplastic syndromes (MDS). Validation of current FC assays and agreements upon the technique are requisites for future clinical application. A working group within the European LeukemiaNet (ELN) was formed in 2008 to discuss and propose standards for FC in MDS bone marrow analysis (Haematologica, 2009). Thus far, three meetings were
S53
convened (Amsterdam, Munich and London) with representatives from 22, mainly European institutes. Major goals of the working conferences were: a. to define the role of FC in the diagnosis and prognosis of MDS related to currently validated systems (including cytogenetic and molecular markers); b. to define minimal requirements to assess dysplasia by FC in known or suspected MDS; c. to define how these data should be interpreted objectively; d. to consider the specificity of FC in MDS related to a series of other (non) clonal hematological diseases. The group concluded that standardization of FC in MDS may not only contribute to the improvement of diagnosis in MDS and in patients with unexplained cytopenias, but it may also refine prognostic scoring systems. That is, FC may identify MDS subgroups with an altered disease course not determined by current methods. Additionally, replicate FC assessments are highly recommended, not only in inconclusive cases, but also to monitor the course of the disease in untreated, mainly low risk MDS patients, and to monitor treatment with currently available drugs. Since MDS is a heterogeneous disease, it is unlikely that a single marker supports its diagnosis; consequently an accumulation of FC aberrancies might confer the diagnosis of MDS. Minimal requirements to analyze dysplasia by FC were defined. The proposed core markers should enable a categorization of the results of FC analysis as normal or as (possibly) in agreement with MDS. A FC report should include a description of validated FC abnormalities such as aberrant marker expression on myeloid progenitor cells (e.g. lineage infidelity marker expression); in addition, dysgranulopoiesis and dysmonocytopoiesis might be scored if ≥2 abnormalities are evident. The working group is dedicated to initiate further studies to establish robust FC diagnostic and prognostic markers and marker patterns in MDS, with the ultimate goal to refine and improve diagnosis and prognostic scoring systems. Nevertheless, it is stressed that FC analysis should be part of the integrated diagnosis rather than a separate technique. 138 A preliminary study on the value of periodic acid-Schiff stain in erythroblasts in myelodysplastic syndrome Z. Xiao, L. Liu. Institute of Hematology and Hospital of Blood Diseases, Chinese Academe of Medical Sciences, Tianjin, China Objective: To investigate the value of periodic acid-Schiff (PAS) stain in erythroblasts in judgement of dyserythropoiesis, diagnosis and differential diagnosis of myelodysplastic syndrome (MDS). Methods: The PAS stain in bone marrow (BM) erythroblasts in 406 MDS pateints, 207 non-severe aplastic anemia (NSAA), 144 immune thrombocytopenic purpura (ITP), 67 megaloblastic anemia (MegA), 76 iron deficiency anemia (IDA), 50 paroxysmal nocturnal hemoglobinuria (PNH), 50 acute erythroid leukemia (AEL) and other related laboratory tests data in MDS patients were analyzed retrospectively. Results: The PAS-positive detection rate in MDS (53.0%) was significantly higher than in NSAA (14.5%), ITP (27.1%) and PNH (16.0%), but significantly lower than in AEL (84.0%) (All the P = 0.000). There was no significant difference in the PAS-positive detection between MDS and MegA (46.3%), or MDS and IDA (40.8%) (P = 0.310; P = 0.052). The PAS-positive rate (Median, M = 1%) and PAS-positive scores (M = 2) in erythroblasts in MDS was significantly lower than in AEL (M = 8%; M = 17), and significantly higher than in NSAA (M = 0%; M = 0), ITP (M = 0%; M = 0), PNH (M = 0%; M = 0), MegA (M = 0%; M = 0), and IDA (M = 0%; M = 0) (All the P < 0.05). The cut-off value of PAS-positive rate and score for the diagnosis of MDS from the other groups except AEL were 0.5% and 0.5, whose sensitivity and specificity were 60.8% and 74.4% respectively. MDS patients were divided into PAS-positive and PAS-negative groups according to PAS reaction, the percentage of erythroid cell in BM was