Myelodisplasias

Improvement of criteria for refractory cytopenia with multilineage dysplasia according to the WHO classification based on prognostic significance of morphological features in patients with refractory anemia according to the FAB classification

Article metrics

Abstract

In the criteria of refractory cytopenia with multilineage dysplasia (RCMD) according to the WHO (World Health Organization) classification, the frequency threshold concerning dysplasia of each lineage was defined as 10%. To predict overall survival (OS) and leukemia-free survival (LFS) for patients with refractory anemia (RA) according to the French-American-British (FAB) classification, we investigated prognostic factors based on the morphological features of 100 Japanese and 87 German FAB-RA patients, excluding 5q-syndrome. In the univariate analysis of all patients, pseudo-Pelger–Huet anomalies 10% (Pelger+), micromegakaryocytes 10% (mMgk+), dysgranulopoiesis (dys G) 10% and dysmegakaryopoiesis (dys Mgk) 40% were unfavorable prognostic factors for OS and LFS (OS; P<0.001, LFS; P<0.001). The prognostic effects of the morphological features were similar in both Japanese and German patients. However, dys Mgk 10% was not correlated with OS and LFS. In the multivariate analysis, mMgk+ and dys Mgk40% were adverse prognostic factors for OS for all patients, and dys G 10% and dys Mgk40% were adverse prognostic factors for LFS for all patients. On the basis of the present analysis, we propose the following modified morphological criteria for RCMD. Modified RCMD should be defined as FAB-RA, excluding 5q-syndrome with dys G 10%, dys Mgk40% or mMgk+.

Introduction

Myelodysplastic syndromes (MDSs) are acquired clonal stem cell disorders characterized by ineffective hematopoiesis with myelodysplasia,1 and are associated with a high risk of progression to acute leukemias.2 MDSs are very heterogeneous in terms of their morphology, clinical features and survival.3 Refractory anemia (RA) according to the French-American-British (FAB) classification is generally classified as a low-risk group.4 The International Prognostic Scoring System (IPSS) was reported to be useful for assessing prognosis in MDS patients according to the FAB classification.5 According to the WHO (World Health Organization) classification,6 most FAB-RA patients are re-classified into refractory cytopenia with multilineage dysplasia (RCMD) or WHO-RA. It was reported that RCMD patients showed a more unfavorable prognosis than WHO-RA patients.7, 8, 9 The criteria for RCMD include a uniform threshold of 10% for dysplasia in each lineage. However, the impact of this threshold on prognosis has not been fully assessed. Concerning the individual forms of dysplasia, we have previously reported that pseudo-Pelger–Huet anomalies (Pelger) and micromegakaryocytes (mMgk) were significantly correlated with overall survival (OS) and leukemia-free survival (LFS) in FAB-RA patients.10, 11 Here, we report the impact of the threshold for dysplasia in each lineage and the individual dysplasias on the prognosis of FAB-RA patients.

Patients and methods

Patients

A total of 200 patients (Japan, 100 cases; Germany, 100 cases) with a diagnosis of primary FAB-RA were selected randomly. Patients who had previously been treated with antineoplastic drugs or ionizing radiation were excluded from the analysis. Japanese patients were diagnosed at the Saitama Medical University Hospital, Nagasaki University Hospital or affiliated hospitals in Japan between April 1976 and January 2002. German patients were diagnosed at the Department of Hematology, Oncology and Clinical Immunology of the Heinrich-Heine University in Germany between January 1973 and December 2002. Thirteen FAB-RA patients with isolated del(5q) (5q-syndrome; all were German patients) were excluded from the analysis. This retrospective analysis was performed in 100 Japanese and 87 German FAB-RA patients. Age, sex and cytogenetic findings of the patients at the diagnosis are summarized in Table 1.

Table 1 Results of morphological analysis and univariate analysis of OS and LFS in patients with FAB-RA, excluding 5q-syndrome

Morphological study

Microscopical examinations were performed using standard methods (bone marrow (BM) Wright-Giemsa (WG) or May-Giemsa (MG), Prussian blue and periodic acid-Schiff stained films and peripheral blood (PB) WG or MG stained films). PB and BM differential counts were performed on 100 and 500 cells, respectively. Evaluations of the BM cellularity and number of megakaryocytes were performed using the specimens of BM trephine biopsy and/or clot section.

We held two meetings on BM morphology at the Heinrich-Heine University, as reported previously.12 At the first joint review, we mainly discussed the evaluation of dysplasia and diagnosis using the training slides. After the first joint review, the Japanese and German groups evaluated the detailed morphological analysis separately in each country. After this separate review, the second joint review meeting for morphological consensus was performed. The observers were blinded to the clinical and laboratory data, including cytogenetics, until finishing this joint review meeting for morphological consensus.

In the present study, we limited dysplasias to only dysplasias described in the WHO classification6 as follows. Dysplasias of the nucleus in erythroid lineage cells were defined as budding, bridging, internuclear, karyorrhexis, multinuclearity or megaloblastoid change. Dysplasias of the cytoplasm in erythroid lineage cells were defined as ringed sideroblasts, vacuolization or PAS positivity (diffuse or granular). Concerning granulocytes, dysplasias were defined as small size, nuclear hypo-segmented mature neutrophils, hypersegmentation, hypogranularity or pseudo-Chediak-Higashi granules. Dysplasias of megakaryocytes were defined as micromegagakaryocytes, non-lobulated nuclei or multiple widely separated nuclei. A morphological study was performed in detail. A minimum of 25 megakaryocytes, 200 erythroblasts and 200 neutrophils in BM were examined in each patient. The cutoff levels for dyserythropoiesis (dys E) and dysgranulopoiesis (dys G) were defined as 10% according to the WHO classification. Dysmegakaryopoiesis (dys Mgk) was evaluated with two cutoff levels, 10% according to the WHO classification or 40% according to data previously reported from the German group.7, 13 Patients with decreased megakaryocytes were excluded from the evaluation of dys Mgk. Two distinct dysplastic changes, Pelger and mMgk, were also evaluated. We defined hypo-segmented mature neutrophils with strikingly clumpy chromatin as ‘Pelger’, and mono- or binucleated megakaryocytes with a size equal to or smaller than promyelocytes as ‘mMgk’. Positivity for Pelger (Pelger+) was defined as the presence of 10% or more Pelger among 200 mature neutrophils. Positivity for mMgk (mMgk+) was defined as the presence of 10% or more mMgk among 25 or more megakaryocytes. Patients with decreased megakaryocytes were judged to be negative for mMgk (mMgk−). The final morphological evaluation was based on the consensus among the Japanese and German groups by joint review.

Cytogenetics

Cytogenetic analysis was performed with a trypsin-Giemsa banding technique on BM cells from aspirates. Ordinarily, 20–30 metaphases were examined. Cytogenetic aberrations were grouped according to the IPSS publication.

Statistical analysis

Patients were followed from the date of diagnosis until June 2004 for the Japanese and July 2003 for the German patients. Prognosis was evaluated by OS and LFS. OS was measured from the date of diagnosis until death owing to any cause, until the date of stem cell transplantation or until the last patient contact. LFS was measured from the date of diagnosis until the date of diagnosis of acute leukemia. Univariate analysis of sex, age category, each morphological parameter and cytogenetic subgroups according to IPSS on prognosis was evaluated with cumulative probabilities using the Kaplan–Meier method and compared using a log-rank test. Multivariate analysis was performed with several significant parameters from univariate analysis. The interaction between parameters was also examined. The effects of parameters were evaluated as hazard ratios and their 95% confidence intervals. Continuous data were compared using the nonparametric Mann–Whitney test, and proportions were compared using the χ2 test. A two-sided P-value of <0.05 was considered to be statistically significant. All statistical analyses were performed with the use of StatView (version 5.0, SAS Institute, Cary, NC, USA) or SAS software (version 8.2, SAS Institute).

Results

Morphological analysis

In the 187 cases reviewed, we evaluated suitable marrow preparations for the detailed assessments of myelodysplasia. The results of morphological analysis are summarized in Table 1. All patients showed dys E10%. Some marrow preparations could not be examined in detail. Especially concerning the frequency of dys G, 24 cases could not be evaluated, because the observation of granules of neutrophils was difficult owing to the poor staining condition of the films. Most of the patients with mMgk+ had dys Mgk40%. Of the patients with dys Mgk<40%, only one patient had mMgk+. However, even with this patient it was judged that the frequency of dys Mgk was 36%.

Univariate analysis of the effects of each parameter on OS and LFS

Follow-up periods ranged from 1 to 292 months (median 43 months). During the follow-up period, 79 patients died and 24 patients transformed to acute leukemia. Japanese FAB-RA patients aged 60 years or less had a more favorable prognosis than German FAB-RA patients aged 60 years or less in OS (P=0.001). Table 1 shows the univariate analysis of the effects of each parameter on OS and LFS. In the analysis of all 187 patients, Pelger+, mMgk+, dys G 10% and dys Mgk40% were significant adverse prognostic factors for OS and LFS. In a separate analysis for each country, there was no prognostic difference regarding Pelger+, mMgk+, dys G 10% or dys Mgk 40% between the Japanese and German patients. When cases with dys G 10% were divided into dys G 10% with or without Pelger+, the prognosis of dys G 10% with Pelger+ was not significantly different from that of dys G 10% without Pelger+ on OS and LFS in all patients and in the patients of each separate country. When cases with dys Mgk 40% were divided into dys Mgk 40% with or without mMgk+, the effect on the prognosis of dys Mgk 40% with mMgk+ was greater than that of dys Mgk 40% without mMgk+ on OS and LFS in all patients. In a separate analysis for each country, the prognosis of patients showing dys Mgk 40% with mMgk+ was worse than that of patients with dys Mgk 40% without mMgk+ regarding OS in patients of both countries and LFS in Japanese patients. Cytogenetic subgroups according to IPSS significantly affected OS and LFS in all patients. In a separate analysis for each country, they significantly affected OS in German patients and LFS in patients of both countries. Age >60 years significantly affected OS in all patients. In a separate analysis for each country, age >60 years significantly affected OS and LFS in only Japanese patients.

Multivariate analysis of the effects of each parameter on OS and LFS

As a next step, we performed a multivariate analysis based on our results obtained in univariate analysis. Table 2 shows the multivariate analysis of the effects of each parameter on OS and LFS. We analyzed parameters in all patients and in patients of each country separately. Model A included the country, age category, sex, dys G category and dys Mgk category. Model B included the country, age category, sex, dys G category, Pelger, dys Mgk category and mMgk. Model C included parameters of model B and cytogenetic findings. In model A, dys G 10% and dys Mgk40% were significantly associated with an adverse prognosis regarding OS and LFS in all patients. In Japanese patients, dys G 10% was a significant adverse prognostic factor for OS and LFS, and Mgk 40% was a significant adverse prognostic factor for LFS. In German patients, dys G 10% was a significant adverse prognostic factor for LFS, and Mgk 40% was a significant adverse prognostic factor for OS. In model B, dys Mgk 40% and mMgk+ were significantly associated with an adverse prognosis regarding OS, and dys G 10% and dys Mgk 40% were significant adverse prognostic factors for LFS in all patients. In Japanese patients, dys G 10% was a significant adverse prognostic factor for OS and LFS. In German patients, dys G 10% and dys Mgk 40% were significant adverse prognostic factors for LFS. In model C, mMgk+ and cytogenetic subgroups were significantly associated with an adverse prognosis regarding OS, and dys G 10%, dys Mgk 40% and cytogenetic subgroups were significant adverse prognostic factors for LFS in all patients. In Japanese patients, dys G 10% was a significant adverse prognostic factor for OS, and dys G 10% and cytogenetic subgroups were significant adverse prognostic factors for LFS. In German patients, cytogenetic subgroups were a significant adverse prognostic factor for OS, and dys G 10% was a significant adverse prognostic factor for LFS. Age >60 years was a significant adverse prognostic factor for OS and LFS of all models in Japanese patients. In contrast, age >60 years was not a significant adverse prognostic factor in German patients.

Table 2 Multivariate Cox hazard analysis of parameters for overall and leukemia-free survival in patients with FAB-RA, excluding 5q-syndrome

Proposal for morphological criteria for RCMD

Regarding OS, RCMD patients who were diagnosed using a uniform threshold of 10% for dys G and dys Mgk according to the original WHO classification did not show a worse prognosis than WHO-RA patients (P=0.111) (Figure 1). This finding indicates that the morphological criteria for RCMD of the original WHO classification may be insufficient for assessing the prognosis. Dys Mgk 10% was not correlated with OS and LFS. However, the frequency of dys Mgk was correlated with prognosis in FAB-RA patients, excluding 5q-syndrome. Patients with dys Mgk 70% or dys Mgk of 40–70% showed a more unfavorable prognosis than patients with dys Mgk of 10–40% or dys Mgk<10% (OS, P<0.001; LFS, P<0.001). Patients with dys Mgk 70% had a more unfavorable prognosis than patients with dys Mgk of 40–70% (OS, P=0.003; LFS, P=0.114). However, there was no prognostic difference between the patients with dys Mgk of 10–40% and patients with dys Mgk<10% (OS, P=0.277; LFS, P=0.881). (Figure 2) Most of the patients with mMgk+ had dys Mgk40%. Of the patients with dys Mgk<40%, only one patient had mMgk+. The prognostic effect of mMgk+ might relate to the prognostic difference between patients with dys Mgk 40% and patients with dys Mgk<40%. To clarify this point, we compared the OS and LFS between patients showing dys Mgk40% without mMgk+ (dys Mgk40%/mMgk−) and patients with dys Mgk<40%. However, patients with dys Mgk40%/mMgk− had a more unfavorable prognosis than patients with dys Mgk<40% (median survival: dys Mgk40%/mMgk−, 76 months; dys Mgk<40%, 217 months; P=0.001, 10%; LFS: dys Mgk40%/mMgk−, 20 months; dys Mgk<40%, not reached; 25% LFS: dys Mgk40%/mMgk−, 38 months; P<0.001). In addition, dys Mgk40% and mMgk+ were independent adverse prognostic factors for OS in the model B of multivariate analysis.

Figure 1
figure1

Cumulative overall survival and leukemia-free survival of WHO-RA and RCMD patients according to the original WHO classification. In overall survival, the RCMD patients, according to original WHO classification, did not show a more unfavorable prognosis than the original WHO-RA patients (P=0.111).

Figure 2
figure2

Cumulative overall and leukemia-free survival and frequency of dysmegakaryopoiesis in patients with FAB-RA, excluding 5q-syndrome. (a) Overall survival (OS). (b) Leukemia-free survival (LFS). (a) Patients with dys Mgk 70% or dys Mgk of 40–70% showed a more unfavorable OS than patients with dys Mgk of 10–40% or dys Mgk<10% (P<0.001). Patients with dys Mgk 70% had a more unfavorable OS than patients with dys Mgk of 40–70% (P=0.003). There was no prognostic difference between patients with dys Mgk of 10–40% and patients with dys Mgk<10% (P=0.277). (b) Patients with dys Mgk 70% or dys Mgk of 40–70% showed a more unfavorable LFS than patients with dys Mgk of 10–40% or dys Mgk<10% (P<0.001). The LFS of patients with dys Mgk 70% tended to be worse than that of patients with dys Mgk of 40–70% (P=0.114). There was no prognostic difference between patients with dys Mgk of 10–40% and patients with dys Mgk<10% (P=0.881).

We attempted to modify the original WHO criteria for RCMD. Except for 5q-syndrome, the WHO classification for the MDS category does not include cytogenetic findings. Therefore, we excluded the cytogenetic findings from the parameters for RCMD. Based on uni- and multivariate analyses, we propose modified morphological criteria for RCMD, as shown in the following. FAB-RA patients, excluding 5q-syndrome, are re-classified into RCMD or WHO-RA. Category A is defined as dys E10%. Categories B1, B2 and B3 are defined as dys G10%, dys Mgk40% and mMgk+, respectively. RCMD is diagnosed when category A and any other category B are present. WHO-RA is defined as FAB-RA other than RCMD. Of the 173 present patients who were suitable for a detailed assessment of dysplasia, our FAB patients, excluding 5q-syndrome, were re-classified into 89 modified WHO-RA and 84 modified RCMD patients according to our modified morphological criteria. Frequency of the ‘poor risk karyotype’ according to IPSS in the modified WHO-RA (5%) was lower than that in the modified RCMD (20%) (P=0.002). In contrast, the frequency of the ‘good risk karyotype’ in the modified WHO-RA (80%) was higher than that in modified RCMD (61%) (P=0.006). In the OS, modified RCMD patients were significantly more unfavorable than modified WHO-RA patients (Figure 3a). For patients aged 60 years or less, the OS of the modified RCMD patients was significantly more unfavorable than that of the modified WHO-RA patients. And, for those older than 60 years, the modified RCMD patients show a more unfavorable OS than the modified WHO-RA patients (Figure 3b-c). In the LFS, the modified RCMD patients were significantly more unfavorable than the modified WHO-RA patients (Figure 3a). For patients aged 60 years or less, the LFS of modified RCMD patients was significantly more unfavorable than that of the modified WHO-RA patients. For those older than 60 years, the modified RCMD patients show a more unfavorable LFS than the modified WHO-RA patients (Figure 3b-c).

Figure 3
figure3

Cumulative overall survival and leukemia-free survival of WHO-RA and RCMD patients according to the modified WHO classification. (Top) Overall survival (OS). (Bottom) Leukemia-free survival (LFS). (a) Among all patients with FAB-RA, excluding 5q-syndrome, the modified RCMD patients had a more unfavorable prognosis than the modified WHO-RA patients (OS, P<0.001; LFS, P<0.001). (b) In patients aged 60 years or younger, the modified RCMD patients had a more unfavorable prognosis than the modified WHO-RA patients (OS, P<0.001; LFS, P<.0.001). (c) In patients aged older than 60 years, the modified RCMD patients had a more unfavorable prognosis than the modified WHO-RA patients (OS, P=0.005; LFS, P=0.008).

Discussion

The patients in the present study were selected randomly, and the number of patients was smaller than our previous study.12 However, the clinical features of Japanese patients were different from those of German patients as in the previous study. Japanese patients in present study were significantly younger than German patients (median age: Japan, 56 years; Germany, 62 years; P=0.026). Japanese patients had lower absolute neutrophil counts (median: Japan, 1.39 × 109/l; Germany, 1.82 × 109/l; P=0.069), lower hemoglobin concentrations (median: Japan, 8.2 g/dl; Germany, 10.3 g/dl; P<0.001), lower platelet counts (median: Japan, 34 × 109/l; Germany, 108 × 109/l; P<0.001), and a lower frequency of cytogenetic abnormalities (Japan, 27%; Germany, 64%; P<0.001) than German patients.

Previous reports from our Japanese and German MDS study group, as well as other investigators, confirmed that WHO-RA patients had a more favorable prognosis than RCMD patients.7, 8, 9, 12 In our previous report,12 the concordance rate of morphological diagnosis according to the WHO classification between Japanese and German hematologists was 83.8%, and a significant concordance was achieved while using the WHO classification (κ 0.73, P<0.001). Therefore, we believe that the evaluation of the frequency of dysmyelopoiesis is comparable between Japanese and German hematologists. Moreover, the present final evaluations concerning dysplasia were reached by consensus among the Japanese and German groups by a joint review. We believe that the WHO classification based on morphological features is useful, at least, in our Japanese and German groups, if the morphological features reflect the prognosis. However, the previous study12 was performed according to the criteria of a prior report from Germany.7 The threshold of dys Mgk in this report was defined as 40%. In the criteria of RCMD according to the original WHO classification, the threshold of frequency for the degree of dysplasia in each lineage was defined as 10%. Still, the impact of this threshold (10%) in each lineage on prognosis has not been fully assessed. Nosslinger et al.14 reported that WHO-RA patients did not show more favorable prognoses when compared to the RCMD patients. However, the threshold of dysplasia in their report was 50%. This demonstrates that the threshold of dysplasia for RCMD is still controversial. In the present study, all patients showed dys E10%. Therefore, dys E10% did not have a prognostic effect. Dys Mgk10% was not an unfavorable prognostic factor for OS and LFS. On the other hand, dys G 10% and dys Mgk40% were significant adverse prognostic factors correlated with OS and LFS. And, these threshold levels have similar prognostic effects between Japanese and German patients in uni- and multivariate analyses. We reported earlier that Pelger and mMgk were correlated with OS and LFS in Japanese patients.10, 11 However, when mature neutrophils had two lobes, the definition of ‘Pelger’ in this previous report was different from that in the present study. In this previous report, we defined mature neutrophils with the two lobes joined by a thin, hair-like bridge (‘pince-nez type cells’) as ‘Pelger’. In contrast, we defined hypo-segmented mature neutrophils with strikingly clumpy chromatin as ‘Pelger’ in the present study. Because of this difference of the definition, the frequency of Pelger in the present study was higher than that in this previous report. In the present study, Pelger+ and mMgk+ were significant adverse prognostic factors for OS and LFS. Again, these results were similar between the Japanese and German patients in uni- and multivariate analyses. The results of the present study support our previous results not only in Japanese patients, but also in German patients. Of note, the prognosis of dys G 10% with Pelger+ was not different from that of dys G 10% without Pelger+. In contrast, the prognosis of dys Mgk 40% with mMgk+ was worse than that of dys Mgk 40% without mMgk+.

We recently compared the clinical features of Japanese and German patients with FAB-RA and found some different prognostic factors, e.g. cytopenias according to IPSS publication were found to be useful for the assessment of prognosis in German FAB-RA patients, but not in Japanese FAB-RA patients.12 In contrast, the prognostic relevance of the morphological features was similar in Japanese and German FAB-RA patients in the study presented here. However, in the multivariate analyses, there were slight differences between Japanese and German FAB-RA patients. For this reason, we speculate that the prognostic effects of the age category of Japanese patients may have an influence. In the present study, Japanese FAB-RA patients aged 60 years or less had a more favorable prognosis than German FAB-RA patients aged 60 years or less in OS (P=0.001) as in our previous study.12 The degree of dysplasias was more severe in Japanese patients aged older than 60 years than those 60 years or younger. The frequency of RCMD according to our modified criteria was higher in Japanese patients aged older than 60 years than those 60 years or younger (48% vs 35%). Therefore, it seems that morphological features may not be significant independent prognostic factors due to the effects of age category in Japanese patients. In contrast, there were no differences in the frequency of RCMD according to our modified criteria between the German patients aged older than 60 years and those 60 years or younger (60% vs 62%). In addition, because the frequency of poor staining of the films was high in German patients, only a small number of cases could be judged in the morphological study (n=58) of German patients. It is expected that the significance as prognostic factors of morphological features becomes certain in multivariate analyses if we can examine more examples even in German patients. Concerning model C including cytogenetics, the number of patients with poor karyotype among 84 RCMD patients according to the modified definition was 17 (20%). In contrast, the number of patients with poor karyotype among 89 WHO-RA patients was only 4 (5%). We thought that the degree of dysplasias was related to the cytogenetic findings. Therefore, in model C, morphological features may not be significant independent prognostic factors.

In univariate analyses, dys G 10% was correlated with OS and LFS in all patients and in patients from each country. However, dys Mgk 10% was not correlated with OS and LFS in all patients or in patients from either country. In the present patients, RCMD patients diagnosed by using a uniform threshold of 10% for dys G and dys Mgk according to the original WHO classification did not show a worse prognosis than WHO-RA patients. We think that it may be necessary to revise the morphological definition of RCMD to improve the WHO classification. Therefore, we propose modified morphological criteria for RCMD.

This morphological analysis in MDS patients has several limitations. We held two meetings on BM morphology and made great efforts to achieve morphological consensus. However, the evaluation of dysplasias might be different among different observers. The number of evaluable cells in the megakaryocytic lineage is smaller than that in other lineages. Therefore, the concordance rate of frequency of dys Mgk among different observers might be lower than that of dys G or dys E. We think that the different morphological interpretation of megakaryocytes among different observers is one of the main causes of the disagreement in the diagnosis of WHO classification. For example, patients of FAB-RA in which three megakaryocytes among 25 megakaryocytes are judged to be dysplastic are classified as RCMD according to the original WHO classification. In contrast, patients of FAB-RA in which only two megakaryocytes among 25 megakaryocytes are judged to be dysplastic are classified as WHO-RA. Therefore, we think that this threshold (10%) of dys Mgk has problems not only for the assessing of the prognosis but also for the diagnosis of RCMD. It was reported that mMgk was specific dysplasias in MDS patients15 and the concordance rates concerning mMgk was sufficient.10 We think that the disagreement rate of morphological diagnosis might be decreased by using our modified criteria combining the frequency of dys Mgk and mMgk. The threshold (40%) of dys Mgk of our modified criteria is different from that (10%) of the original WHO criteria. The German group had already shown the usefulness of the WHO classification in several large-scale studies7, 13 using this threshold (40%). Therefore, we believe that the usefulness of this threshold shown in the present study is certain.

In conclusion, the present results showed that the degree of dysplasia in FAB-RA patients was related to OS and LFS, and the prognostic effect of dysplasia was similar between the Japanese and German FAB-RA patients. However, the thresholds of dysplasia influencing prognosis were different from the original threshold of the WHO classification. We propose to raise the threshold of dys Mgk in the criteria for RCMD from 10 to 40% and add mMgk+.

References

  1. 1

    Goasguen JE, Bennett JM . Classification and morphologic features of the myelodysplastic syndromes. Semin Oncol 1992; 19: 4–13.

  2. 2

    Ganser A, Hoelzer D . Clinical course of myelodysplastic syndromes. Hematol Oncol Clin North Am 1992; 6: 607–617.

  3. 3

    Koeffler HP . Introduction: myelodysplastic syndromes. Semin Hematol 1996; 33: 87–94.

  4. 4

    Hofmann WK, Ottmann OG, Ganser A, Hoelzer D . Myelodysplastic syndromes: clinical features. Semin Hematol 1996; 33: 177–185.

  5. 5

    Greenberg P, Cox C, LeBeau MM, Fenaux P, Morel P, Sanz G et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 1997; 89: 2079–2088.

  6. 6

    Jaffe ES, Harris NL, Stein H, Vardiman JW (eds.) World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. IARC Press: Lyon, France, 2001.

  7. 7

    Germing U, Gattermann N, Strupp C, Aivado M, Aul C . Validation of the WHO proposals for a new classification of primary myelodysplastic syndromes: a retrospective analysis of 1600 patients. Leuk Res 2000; 24: 983–992.

  8. 8

    Rosati S, Mick R, Xu F, Stonys E, Le Beau MM, Larson R et al. Refractory cytopenia with multilineage dysplasia: further characterization of an ‘unclassifiable’ myelodysplastic syndrome. Leukemia 1996; 10: 20–26.

  9. 9

    Lorand-Metze I, Pinheiro MP, Ribeiro E, de Paula EV, Metze K . Factors influencing survival in myelodysplastic syndromes in a Brazilian population: comparison of FAB and WHO classifications. Leuk Res 2004; 28: 587–594.

  10. 10

    Matsuda A, Jinnai I, Yagasaki F, Kusumoto S, Minamihisamatsu S, Honda S et al. Refractory anemia with severe dysplasia: clinical significance of morphological features in refractory anemia. Leukemia 1998; 12: 482–485.

  11. 11

    Matsuda A, Jinnai I, Yagasaki F, Kusumoto S, Murohashi I, Bessho M et al. New system for assessing the prognosis of refractory anemia patients. Leukemia 1999; 13: 1727–1734.

  12. 12

    Matsuda A, Germing U, Jinnai I, Misumi M, Kuendgen A, Knipp S et al. Difference in clinical features between Japanese and German patients with refractory anemia in myelodysplastic syndromes. Blood 2005; 106: 2633–2640.

  13. 13

    Germing U, Gattermann N, Aivado M, Hildebrandt B, Aul C . Two types of acquired idiopathic sideroblastic anaemia (AISA): a time-tested distinction. Br J Haematol 2000; 108: 724–728.

  14. 14

    Nosslinger T, Reisner R, Koller E, Gruner H, Tuchler H, Nowotny H et al. Myelodysplastic syndromes, from French-American-British to World Health Organization: comparison of classifications on 431 unselected patients from a single institution. Blood 2001; 98: 2935–2941.

  15. 15

    Kuriyama K, Tomonaga M, Matsuo T, Ginnai I, Ichimaru M . Diagnostic significance of detecting pseudo-Pelger–Huet anomalies and micro-megakaryocytes in myelodysplastic syndrome. Br J Haematol 1986; 63: 665–669.

Download references

Acknowledgements

Supported in part by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (no. 16639013) (IJ) and Kompetenznetzwerk ‘Akute und Chronische Leukämien’ des Bundesforschungsministeriums.

Author information

Correspondence to A Matsuda.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Matsuda, A., Germing, U., Jinnai, I. et al. Improvement of criteria for refractory cytopenia with multilineage dysplasia according to the WHO classification based on prognostic significance of morphological features in patients with refractory anemia according to the FAB classification. Leukemia 21, 678–686 (2007) doi:10.1038/sj.leu.2404571

Download citation

Keywords

  • myelodysplastic syndromes
  • refractory anemia
  • refractory cytopenia with multilineage dysplasia
  • WHO classification
  • prognosis

Further reading