|Year : 2014 | Volume
| Issue : 1 | Page : 40-47
Validity of new flow cytometric protocol in diagnosis of low-grade myelodysplastic syndromes
B Jamoh Yusuf1, P Goggolidou2, T Milne3, AA Abba1, AG Bakari1, A Sa'idu4, AI Dutse5, BA Gwaram5, K Ashfaq6, SA Abubakar1
1 Department of Medicine, Ahmadu Bello University Teaching Hospital, Zaria, Nigeria
2 School of Life Sciences, Kingston University, London, UK
3 Department of Haemato-Oncology, King's Path, King's College Hospital, London, UK
4 Department of Medicine, Federal Medical Center, Gombe, Nigeria
5 Department of Medicine, Aminu Kano Teaching Hospital, Kano, Nigeria
6 Centre for Clinical Practice, National Institute of Health and Clinical Excellence MidCity Place, London, UK
|Date of Submission||05-Aug-2013|
|Date of Acceptance||24-Sep-2013|
|Date of Web Publication||24-Mar-2014|
B Jamoh Yusuf
Department of Medicine, Ahmadu Bello University Teaching Hospital, Zaria, Nigeria
Source of Support: The authors did not receive any form of support from any
source., Conflict of Interest: The authors declare that they have no conflict
of interest in this study.
Background: Myelodysplastic syndromes (MDS) are the most common of myeloid malignancies, yet the morphological diagnosis is usually not straightforward especially in the low-grade forms. Immunophenotyping by Flow cytometry (FCM) is considered essential in the WHO's co-criteria for diagnosis of MDS. The Existing FCM Protocol utilizes a two-tube, two-colour approach to identify lineage specific cluster of differentiation (CD) markers, which is labour-intensive and time-consuming. A new FCM was recently developed and validated among Japanese cohorts. It utilizes a three-tube, five-colour approach and generates more information in the form of 'cardinal parameters'. The aim of this study is to determine the diagnostic utility of the new protocol by comparing it with the existing protocol, in the diagnosis of low-grade MDS in our study population. Materials and Methods: We analyzed bone marrow samples of 30 subjects at King's, London. They comprised of 27 patients who had a tentative diagnosis of MDS and three healthy bone marrow donors as controls. Immunophenotyping by FCM was performed using the Existing and New Protocols and the data obtained by the two different methods were compared. 'Cardinal parameters' were generated in the new protocol, which are not applicable to the existing protocol. Results: There was no statistical difference between the data generated by the two protocols in the diagnosis of MDS. The sensitivity and specificity of the 'cardinal parameters' of the new protocol appeared to be outstanding. Conclusion: It has been shown that the new multiplex FCM protocol for the diagnosis of MDS is relatively easy, cost effective and not inferior, compared to the Existing Protocol. However, small sample size has been identified as a limitation to the study and therefore a larger, multicenter study is recommended to assess this validation exercise.
Keywords: Diagnosis, Low grade, Myelodysplastic syndromes, New flow cytometry protocol, Validity
|How to cite this article:|
Yusuf B J, Goggolidou P, Milne T, Abba A A, Bakari A G, Sa'idu A, Dutse A I, Gwaram B A, Ashfaq K, Abubakar S A. Validity of new flow cytometric protocol in diagnosis of low-grade myelodysplastic syndromes. Sub-Saharan Afr J Med 2014;1:40-7
|How to cite this URL:|
Yusuf B J, Goggolidou P, Milne T, Abba A A, Bakari A G, Sa'idu A, Dutse A I, Gwaram B A, Ashfaq K, Abubakar S A. Validity of new flow cytometric protocol in diagnosis of low-grade myelodysplastic syndromes. Sub-Saharan Afr J Med [serial online] 2014 [cited 2020 Aug 11];1:40-7. Available from: http://www.ssajm.org/text.asp?2014/1/1/40/129314
| Introduction|| |
Haematological malignancies are vast in spectrum and together, they rank 5 th in the list of common malignancies in the UK. 
Myelodysplastic syndromes (MDS) are defined as "a heterogeneous group of clonal haematopoietic stem cell diseases characterized by peripheral blood cytopenia(s), morphological dysplasia in one or more of the major myeloid cell lines, ineffective haematopoiesis, progressive marrow failure and increased risk of development of Acute Myeloid Leukaemia (AML)". , They are the most common of myeloid malignancies,  and therefore they carry a significant medical and economic importance.
Morphological examination of bone marrow specimens has been the most widely employed laboratory tool in the diagnosis of MDS.  However, it is not always straightforward to diagnose MDS by this modality because not all clinically suspected cases are true MDS  due to the fact that a number of differentials may mimic the disease. Commonly, patients may present with characteristic bone marrow disorders typical for MDS, including peripheral cytopenias, ineffective haematopoiesis, morphological dysplasia and even recurrent cytogenetic abnormalities.  In some patients there is lack of specific diagnostic markers, while in other instances differentiating hypocellular MDS from Aplastic Anaemia is not possible due to striking clinical and morphological similarities.  Attention is currently being focused on the case of Idiopathic Cytopenia of Undetermined Significance (ICUS), which is sometimes indistinguishable from MDS when one uses the clinical presentation or traditional diagnostic techniques. 
The diagnosis of MDS requires minimal diagnostic criteria, developed during the International Working conference in Vienna. The consensus recommends the presence of two prerequisite-type criteria (A-criteria), at least one out of three decisive criteria (B-criteria) and several co-criteria (C-criteria) for diagnosis of the disease. 
The C-type criteria include the flow cytometric data analysis regarding the expression of specific immunophenotypic markers by the cells, which is the basis of this research.
Immunophenotyping by flow cytometry may determine the size and immunotype of the blast population and may also determine the maturation pattern of the myeloid series.  Furthermore, pathological populations of CD34 or CD117 cells may suggest disease evolution.  In advanced stages of MDS and AML-transformed MDS, immaturity markers such as CD7 and CD117 are frequently expressed. 
The new FCM protocol appears to be relatively easier and more informative, since it generates more information in the form of 'cardinal parameters'. It may also conserve time taken to analyze samples. Saving time in any laboratory means increased efficiency and fewer requirements for staff, which is monumentally important to both patients and the management of the laboratory. Comparing the two protocols may provide justification to maintain the existing protocol or to replace it with the new one.
This study was carried out based on the hypothesis that the New FCM protocol gives reliable information similar, and hence not inferior, to that obtained using the Existing Protocol, for the diagnosis of MDS.
| Materials and Methods|| |
Patients Characteristics and Sample Collection
The study population comprised of patients who presented to the haemato-oncology department of King's college hospital, assessed by clinicians who made clinical impression of MDS. They were recruited consecutively as they presented, by convenience sampling method. Control group was pooled from healthy subjects who consented for bone marrow donation for transplant. Specimens under study were bone marrow samples, obtained by conventional bone marrow aspiration technique.
We analyzed bone marrow samples of 30 subjects. They comprised of 27 patients who had a tentative diagnosis of MDS at presentation. Three controls were pooled from healthy bone marrow donors. Immunophenotyping by FCM was performed using the Existing and New Protocols, with a view to comparing the data generated by the 2 different methods. As part of the New Protocol, four (Cardinal) parameters were generated from which we calculated the sensitivity and specificity of the New Protocol.
All specimens were analyzed for cellular morphology and blast counts as well.
While carrying out the new protocol, antibody staining was performed by adding, to the test tubes, 5μl of each antibody: Fluorescien Isothiocyanate (FITC), Peridin chlorophyll (PerCP-Cy5.5), Allophycocyanin (APC) or Phycoerythrin (PE). FITC-labeled antibodies were used against CD45, CD15 and CD7; PE against CD56, CD13, CD117 and CD19; APC against CD 11b and CD33; and Per CP were used against CD34.
Next, 2 × 10 5 nucleated cells (bone marrow sample) were added to each of the 3 test tubes. Isotype-matched negative control raised against a non-human protein, IgG2, was added to the 4 th , negative control tube. All antibodies were obtained from Becton Dickinson Biosciences, UK and Beckman Coulter LTD.
The mixtures were incubated for 15 min at room temperatures (18-25 o C), then spun for 2 min at 2,500 revolution per minute, after which (2ml of) 1:10 FACS lyse (BD) was added to lyse the red blood cells.
After 15 min incubation, the mixtures were spun again to tip off the lysing agent and then washed with phosphate buffered saline (PBS). Procedures were completed within the recommended time of 48 h after sample collection.
Data were acquired using the machine, FACS Canto II flowcytometer (BD). At least 80,000 cell events were acquired for most samples. Compensation maneuver was done using single-labeled cells in cases of inappropriate fluorescence emission overlap.
Analysis of data was done using BD FACS Diva 6.1.3 software, by investigators blinded for the patients' clinical and other laboratory details, until after the completion of analyses.
The "Cardinal parameters" generated were:
- CD34+ myeloid progenitors expressed as percent total nucleated cells.
- CD34+ B cell progenitors expressed as percent of total CD34+ progenitors.
- Lymphocyte- and myeloblasts CD45 expression expressed as a ratio.
- Granulocyte- and lymphocyte side scatter expressed as a ratio.
Gating for Cardinal Parameters
On the SSC vs FSC plot [Figure 1], population of all nucleated cells was defined as P1. These cells were then displayed on CD34 vs CD45 plot to identify the CD34+ cells with intermediate CD45 expression. When all cells were plotted on CD45 vs SSC display, lymphocytes and myeloblasts were defined. Finally, channels for granulocytes and lymphocytes SSC were defined.
|Figure 1: Gating for the cardinal parameters. Panel (a) SSC vs FSC was plotted to generate all nucleated cells. Panel (b) P1 clusters (arrow) displayed on CD45 vs CD34 plot to generate granulocytes and lymphoblasts. In panels (c) and (d), SSC vs CD45 was plotted to generate granulocytes and lymphocytes. Panels (e) and (f) display channels of lymphoblasts and myeloblasts expressing CD45 respectively|
Click here to view
Differences between continuous variables (mean CD expression) in the New and Existing Protocols were analyzed using Student's t-test, as paired samples with different variance, using Microsoft Excel © 2010 statistical software. The significant p value taken for all parameters was <0.05.
| Results|| |
We had a total of 30 subjects recruited for the study and were subjected to peripheral blood and bone marrow examination. Six patients were confirmed cases of MDS based on clinical presentation, morphological dysplasia, blast count and cytogenetics and who were already receiving conventional management. They are therefore referred to as "Known MDS cases" in this write-up. Another group of 3 recruits were normal subjects and who are healthy bone marrow donors, referred to as "Normal subjects". Seven patients did not have clear-cut diagnosis at the time of recruitment and are referred to as "Suspected MDS cases". The rest of the recruits (14 in number) had morphological blast counts of either 20% and above or less than 3%. At the end of the study, 8 and 1 patients (in this last group) were diagnosed with AML and Myelofibrosis respectively.
Samples from all subjects were analyzed for cell markers by FCM technique using the "New Protocol" and the "Existing Protocol" (the raw data is given in appendix I), and the mean and standard deviation for each of the parameters were calculated for the different Protocols. Statistical tests of significance were performed to test variables of the expressed CD between the new and current protocols.
Expression of Individual and Dual CD Markers
From the data obtained for all CD markers (Appendix I), values of individual CDs were extracted for all subjects and grouped into "New Protocol" and "Existing protocol". To summarize the indices, Bar Charts were plotted to compare the means of the individual CDs in the two groups [Figure 2]. In order to accommodate all the information within the limited space in the plot, letter "A" was used to denote the New Protocol and letter "B" was used to denote the existing Protocol. Note that the box plots on each bar highlights the standard error of the mean, which varies from one bar to another. Color-coding system was used to differentiate one CD from another.
|Figure 2: Percentage expression of individual CD numbers obtained from New Protocol (A) and current protocol (B). The box plots represent the standard error of the mean. Larger error is seen in CD13 using the|
existing protocol. Note that in most cases, values obtained from the existing protocol are slightly higher than the current protocol but the overall picture reveals that they are similar
Click here to view
In an expression pattern similar to the individual CD markers, results of co-expressed dual CDs were extracted from the crude data and the mean of each value was used to plot a Bar Chart [Figure 3] to compare the two groups, also represented by "A" and "B" along with their standard errors of the mean. In the two figures (3.1 and 3.2) it is immediately apparent that a close resemblance exists between the two groups.
|Figure 3: Percentage co-expression of dual CDs obtained from the New Protocol (A) and the Existing Protocol (B). The trend is similar to figure 2 (above)|
Click here to view
Blast Counts and Expression of Myeloblasts Markers
The result of morphological blast count obtained directly by from the blood film and compared to the corresponding blast counts obtained by Immunophenotyping using the two protocols [Figure 4]. Myeloblasts were defined by CD34/CD13, CD34/CD45 and CD45/CD117 co-expression in both New and Existing Protocols. This particular comparison used the 10 patients (A to J) who had blast counts in the range diagnostic for MDS (i.e. 3-19%).
|Figure 4: Percentage of myeloblasts (CD34/117 or CD45/117 coexpression) in patients A through J, whose blast counts fall within the MDS diagnostic criteria (2 to 19%), obtained by the two protocols plotted against blast count by smear morphology|
Click here to view
Note the following observations in fig 3.3:
- Five patients (A, B, C, D and I) had similar blast counts in all the 3 modalities.
- Eight patients (A, B, C, D, E, F, G, I and J) had similar counts in the New and Existing protocols.
- Seven patients (A, B, C, D, H, I and J) had similar counts in the New protocol and Morphology, and
- Five patients (A, B, C, D and I) had similar counts in Existing protocol and morphology.
Errors in enumerating the blasts on blood film has been identified as a potential source of error and to minimize that, confirmation of the results was sought for from the head of the Laboratory.
The Cardinal parameters were analyzed using the data generated from the New Protocol [Table 1]. For each of the parameters Reference range values were: CD34+Myeloblasts (>1.2%), CD34+ B-cell (>3.5%), Ly/Mbl CD45 (<7.5) and Gra/ly SSC (>4.5).
|Table 1: Values of the Cardinal parameters for the Normal subjects (A' to C'), Known MDS cases (D' to G') and Suspected MDS case (J' to P'). Figures highlighted in red are considered significant for MDS diagnosis|
Click here to view
Next, a scoring system for the cardinal parameters was devised, where 1 point was assigned to any parameter outside the RR. Scores of 0 and 1 were considered as normal, while scores of 2 and above (out of 4) were considered significant for the diagnosis of MDS [Table 2].
|Table 2: Diagnostic power of the new protocol, with respect to the Cardinal parameters. Scores of 2 and above (shown in red) are considered significant in Normal controls, Known MDS and suspected MDS cases. Specificity and Sensitivity in the Suspected MDS group were not calculated|
Click here to view
There were three "Normal" subjects and the result revealed that all of them had normal scores. Similarly, all of the 6 "Known MDS" cases had significant scores (2 or 3 out of 4). Of the 7 "Suspected MDS" cases five had normal scores (0 or 1) while two had significant scores (2 out 4 each).Specificity and sensitivity of the New Protocol were calculated for the "Normal subject" and "Known MDS" groups as 100% respectively.
Comparative Analysis of Existing and New Protocols
Following the acquisition of data on all the CD markers obtained from the two protocols, statistical test of significance was performed on these variables using a two - way Student's t-test, as means of two samples with different variables, for each of the corresponding parameters in the two different groups. The significant level taken was at P < 0.05 for all the analyses [Table 3]. The result of the analysis revealed that there was no statistically significant in values of all the parameters (both individual CD expression and co-expression of dual CDs) between the Existing Protocol and the New Protocol.
This final result illustrates that in terms of diagnostic utility, the New Protocol and Existing Protocols yield similar information in diagnosis of MDS.
It is important to note that the Cardinal parameters generated in the New protocol (and not in the Existing Protocol) were found to be of high specificity and sensitivity.
|Table 3. Comparison of mean percentage of CD numbers between the New and Existing Protocols. In all the parameters there is no statistically significant difference between the two groups at the significant level p < 0.05|
Click here to view
| Discussion|| |
In this study, we used a multicolour (multiplex) flowcytometric protocol devised by the laboratory with the view of validating the New FCM Protocol. This validation was achieved by comparing it with the existing protocol. Some of the conceived reasons that inform the going over to the new approach include the following:
The Existing Protocol utilizes a two-tube, two-colour approach while the new technique utilizes a three-tube, five-colour approach, making the former more labour intensive. It then follows that more time is saved with the new technique. The new multiplex approach can be easily automated and the laboratory has now automated the new testing approach.
The Existing Protocol gives only the basic quantitation of progenitor cells while the new technique gives both quantitative and qualitative information. Other additional data given by the new protocol include percentage of CD34+ myeloblasts, percentage of B-cell progenitors and Granulocyte/Lymphocyte side scatter peak channel ratio (see the discussion on Cardinal parameters) and the ability to produce a scoring system of abnormalities.
The storage of data on the existing technique is cumbersome, compared to the new approach, which gives a more streamlined data storage platform.
Although the cost of running the new protocol is marginally higher than that of the existing protocol (see appendix II), the difference is offset by the laboratory time saved in the new approach. The reduction in staff processing time makes the new approach more cost effective.
The three tube (new) approach, as against the traditional two tube approach, has some benefits which include improved CD19 antibody performance, less impact of CD15-FITC on other antigens and gaining an additional Euroflow compliant antibody (CD33APC).
Performance of the Three Tube Multiplex Approach
The New Protocol in which three-tube multiplex approach was used has been found to be a good modality of FCM in MDS diagnosis.
Although many values of individual and dual CD markers in the Existing Protocol were marginally higher than the New Protocol [Figure 2], the inference is that the seeming difference is due to chance variation. The standard error in CD13 analyzed by the Existing protocol appeared to be wider than that by the New Protocol, hence less reliable. Similar pattern is also observed in co-expression of dual markers [Figure 3].
The power of the multiplex approach in approximating morphological blast count appeared to be remarkable, because in analyzing blood samples using the New protocol, 7 out of 10 (70% of) patients who were in MDS range based on peripheral blood picture a strongly positive correlation was observed between the morphological blast count andimmunophenotypic blast count [Figure 4]. This appears better than the Existing Protocol in which only 50% (5 out of 10) of the patients showed a strong correlation with morphological blast count. This finding is in keeping with the fact that blast estimation by immunophenotyping usually yields a lower count than by morphology, because not all blast cell express CD34 immune marker. 
The Cardinal Parameters
At the end of the pilot study, attempt was made to evaluate the power of the new technique by using the Cardinal parameters, determined from the Normal subjects, Known MDS cases and Suspected MDS cases. The definitive Reference Ranges for each of the cardinal parameters are computed using mean ± 2SD, but because of a relatively small sample size in this study these ranges might not be reliable. However, the devised scoring system in which one point is awarded for each parameter outside the estimated reference ranges, and where scores of 0 or one out of four are considered to represent normal bone marrow findings while scores of two and above were considered significant for diagnosis of MDS may be reproducible.
Analysis of the Cardinal Parameters
CD34 + myeloblast population vs total nucleated cell count
In most cases of MDS, the expression of CD34in myeloblasts increases while CD34 + B-cells decrease.  It then follows that analyzing them separately would be more accurate than estimating all CD34 + populations (done in the existing protocol). That is why in this study we chose to identify the population of myeloblasts that express CD34 separately, because the proliferation and arrested maturation of myeloblasts coupled with dysplastic changes are hallmark for MDS. 
In this study, this parameter appeared to be within the reference range in all the normal subjects [Table 2]. However, it appeared to be lower in the Known MDS group than seen by Ogata et al., i.e. remained within the normal range. This may be due to the fact that total nucleated count (the denominator) was relatively higher in this study and therefore bringing the proportion down. This may be the result of different lysing and gating strategies used. Remotely, difference in the patient cohort was thought of as the possible cause of these apparently low values. In a similar study,  22.2% of MDS cases and 1.1% of normal controls had this parameter outside the normal range.
CD34 + B cell precursors
This parameter appeared, in performance, to obey the predictions by Ogata et al11.: none of the normal control subjects had this parameter outside the reference range, while all the known MDS cases had the parameter outside the range. It is, therefore, one of the parameters that will have reliable positive predictive values. It is expected that the normal subjects would have CD34+ B-cells at high levels (>3.5%) [Table 2], while the known MDS cases would have lower levels (where it is supposedly expressed at higher levels by the myeloblasts instead). This point obviates the fact that here are two major populations of CD34+ cells in the bone marrow, which increase or decrease in a reciprocal manner.  However, this parameter is not specific, because a decrease in CD34+ B-ell precursors is also seen in other haematological malignancies.  In order to circumvent pit falls associated with hemodilution in the enumeration of progenitor B-cells in these 'other' conditions, it has been recommended that these cells are expressed as a fraction of all (CD34+) blast cells. 
Lymphocyte CD45 density vs. Myeloblast CD45 density
Two of the four (50% of the) known MDS patients gave an abnormal value for this parameter [[Figure 1] and [Table 2]. According to Satoh et al.13 , "the quantification of myeloblasts CD45 expression can increase or decrease…we did not find any difference in the ly/Mbl ratio in low grade MDS". It is important to note that the laboratory's software package used mean fluorescence to calculate this parameter whereas Ogata et al., used Mean Fluorescence Intensity (by using another software package). Generation of a different reference range using known normal patients is hence essential.
Granulocyte vs. lymphocyte Side Scatter
The median Side Scatter value was used in this study, as opposed to peak Side Scatter value. The reference range we develop will differ to that of other users because our software is FACSDiva, while Ogata et al., used CellQuest Pro. It is encouraging to note that all the known MDS patients had values for this parameter outside the reference range while only one normal subject had a borderline value [Table 2]. The inclusion of CD10 in the calculation of this parameter was discounted in Ogata's paper however this is currently being investigated as a future development.
So far, the sensitivity of 100% and specificity of 100% were seen in Normal cases and Known MDS cases [Table 3]. These values are slightly higher than those reported by Ogata and co-workers, where sensitivity and specificity were 75% and 93% respectively. It is thought that these marginal differences are due to inter-laboratory variations and probably differences in patient cohorts. More importantly, the small sample size in our study may be another factor, which was brought about by time constraint as the limiting factor.
Following the statistical analysis of the mean values of all the markers expressed by the cells in the samples under study, no significant difference was seen between the Existing Protocol and the New Protocol in any of the parameters that could be defined by both methods [Table 2]. This finding is an interesting one because it means that regarding the diagnostic importance, the Existing and Current Protocols yield very similar information in the evaluation Myelodysplastic syndrome.
| Conclusion|| |
It has been shown that the new multiplex FCM protocol for the diagnosis of MDS is relatively easy, cost effective and straight forward, with no significant difference in diagnostic yield, compared to the Existing Protocol. The promise shown by this protocol is not only in identifying actual cases of MDS, but also in excluding the differential diagnoses. It is also worthy of mentioning that ability of the protocol to define blast cells appears to be remarkable.
The fact that an analysis of the cardinal parameters is fairly reproducible reveals that an inter-laboratory variation in data acquisition is not much pronounced. Although some individual parameters (notably CD34 expression on myeloblasts) show less ability to pick up the disease, the overall sensitivity and specificity of the parameters taken together appear to be outstanding. Allowing for the limitations of this study, one could see that the New Protocol is valid as co-criteria for the diagnosis of MDS.However, the sample size in this study was too small to offer much confidence to the protocol, and for it to be proved specifically reliable in its positive and negative predictive values, a larger or multicenter study is necessary.
| References|| |
|1.||Jemal A, Siegel R, Xu J. Cancer Statistics 2010. CA Cancer J Clin 2010; 60: 277-300. |
|2.||Valent P. and Horny HP. Minimal diagnostic criteria for Myelodysplastic Syndrome and separation from ICUS and IDUS: Update and open questions. Eur J Invest. 2009;39:548-553. |
|3.||Mufti GJ, Bennett JM, Goasguen J, Bain BJ, Baumann I, Brunning R, et al.; International Working Group on Morphology of Myelodysplastic Syndrome. Diagnosis and classification of myelodysplastic syndrome: International Working Group on Morphology of myelodysplastic syndrome (IWGM-MDS) consensus proposals for the definition and enumeration of myeloblasts and ring sideroblasts. Haematologica 2008;93:1712-7. |
|4.||Hamblin TJ. Epidemiology of MDS. In: Bennett JM,editor. MDS: Pathobiology and Clinical Management. New York:Marcel Dekker Inc.; 2002.p. 15-28. |
|5.||Brunning RD, Orazi A, Germing U, le Beau MM, Porwit A, Baumann I, et al. Myelodysplastic Syndromes/Neoplasms, overview. In: WHO classification of tumours of haematopoietic and lymphoid tissues. © WHO, 2008. |
|6.||List AF, Sandberg AA, Doll DC.Myelodysplastic Syndromes.In:Greer P, Foerster J, Lukens J, et al.,editors.Wintrobe's Clinical Hematology.11 th ed. Philadelphia: Lippincott Williams and Wilkins;2004.p.1793-808. |
|7.||Truong F, Smith BR, Stachurski D, Cerny J, Medeiros LJ, Woda BA, et al. The utility of flow cytometricimmunophenotyping in cytopenic patients with a non-diagnostic bone marrow: A prospective study. Leuk Res 2009;33:1039-46. |
|8.||Ogata K, Della Porta MG, Malcovati L, Picone C, Yokose N, Matsuda A, et al. Diagnostic utility of flow cytometry in low-grade myelodysplastic syndromes: A prospective validation study. Haematologica 2009;94:1066-74. |
|9.||Loken MR, Van De Loosdrecht A, Ogata K, Orfao A, Wells DA. Flow cytometry in myelodysplastic syndromes: Report from a working conference. Leuk Res 2008;32:5-17. |
|10.||Valent P, Horny HP, Bennett JM, Fonatsch C, Germing U, Greenberg P, et al. Definitions and standards in the diagnosis and treatment of the myelodysplastic syndromes: Consensus statements and report from a working conference. Leuk Res 2007;31:727-36. |
|11.||Ogata K, Kishikawa Y, Satoh C, Tamura H, Dan K, Hayashi A.Diagnostic application of flow cytometric characteristics of CD34+ cells in low-grade myelodysplastic syndromes. Blood. 2006;108:1037-44. |
|12.||Valent P, Samorapoompichit P, Sperr WR, Horny HP, Lechner K. Myelomastocytic leukaemia: Myeloid neoplasm characterized by partial differentiation of mast cell-lineage cells. Hematol J. 2002;3:90-94. |
|13.||Satoh C, Dan K, Yamashita T, Jo R, Tamura H,Ogata K. Flow cytometric parameters with little interexaminaer variability for diagnosing low-grade myelodysplastic syndromes. Leuk Res 2008;32:699-707. |
|14.||McKenna RW, Washington LT, Aquino DB, Picker LJ, Kroft SH. Immunophenotypic analysis of hematogones (B-lymphocyte precursors) in 662 consecutive bone marrow specimens by 4-color flow cytometry. Blood 2001;98:2498-507. |
|15.||Arjan A, van de LoosdrechtA, Alhan C, Bene MC, Della Porta MG, Drager AM, et al. Standardization of Flow cytometry in Myelodysplastic syndromes: Report from the first European LeukaemiaNet working conference on Flow cytometry in Myelodysplastic syndromes. Haematologica 2009;94:1124-34. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3]