You are here
Refined Medullary Blast and White Blood Cell Count Based Classification of Chronic Myelomonocytic Leukemias
- we showed that proliferative CMML types do worse as compared to dysplastic types.
- progression rates of CMML with <5% blasts are lower as compared to CMML I + II.
- erythroid insufficiency of all CMML types is not as severe as in RCMD, RAEB I and II.
- separate CMML with less than 5% medullary blast count from CMML I.
Since 2001, chronic myelomonocytic leukemia (CMML) is classified by the WHO as myeloproliferative/myelodysplastic neoplasm. Herein we tried to better describe CMML patients with regard to hematological characteristics and prognosis using data of the Duesseldorf registry. We created 6 CMML subgroups, by dividing dysplastic and proliferative CMML at the cut-off of white blood cell count of 13,000/μl and splitting these two groups into 3 subgroups: CMML 0 with <5% blasts (n = 101), CMML I with 5%-9% blasts (n = 204) and CMML II with 10%-19% blasts (n = 81). For comparison we included patients with RCMD, RAEB I and II. The newly created CMML 0 group had better prognosis than CMML I and II, median survival times were 31 months (ms), 19 ms and 13 ms, respectively (p < 0.001). Median survival times between the corresponding dysplastic and proliferative subgroups 0 and 1 differed significantly: CMML 0 dysplastic 48 ms and CMML 0 proliferative 17 ms (p = 0.03), CMML I dysplastic 29 ms and CMML I proliferative 15 ms (p = 0.008), CMML II dysplastic 17 ms and CMML II proliferative 10 ms (p = 0.09). Outcome of CMML patients worsens with increasing medullary blasts and when presenting as proliferative type. Therefore it is justified to separate CMML with <5% medullary blasts.
Keywords: CMML, MDS, MDS/MPN, Overlap, Prognosis, CPSS.
Since 2001 chronic myelomonocytic leukemias (CMML) are classified as mixed myeloproliferative/myelodysplastic neoplasms. Other entities in this group are BCR/ABL negative chronic myeloid leukemia, juvenile myelomonocytic leukemia and finally myeloproliferative/myelodysplastic neoplasms, unclassified  . The revised version of the WHO classification in 2008 still classifies the CMML as an overlap syndrome. During the validity of the French-American-Britain (FAB) leukemia cooperative Group classification starting in 1982 the CMML was considered as myelodysplastic syndrome  . The FAB group stated that CMML usually presents with medullary blast counts fewer than 5%. In 1994 the FAB group proposed to split CMML into a dysplastic subtype with a white blood cell count (WBC) less than 13000/μl and a proliferative type with WBC of more than 13000/μl  (summary of diagnostic criteria see Table 1 ). Cytogenetic abnormalities are present in 20-40% of patients and , there have been efforts on creating cytogenetic risk categories  . Frequently molecular alterations are found in CMML as well, which however are not specific for this entity. Mutations of ASXL1, TET2, SRSF2, RUNX1, NRAS, CBL are frequently found in CMML, , and . Some of them are associated with prognostic relevance, i.e. ASXL1  and TET2  Prognostic scores have been developed for CMML, , , and ,lately the CMML prognostic scoring system (CPSS), which consists of WHO-subtype, FAB-subtype, CMML specific cytogenetic risk classification and red blood cell transfusion dependency  . The Duesseldorf-Score which was already established in 1992 consists of the following parameters: bone marrow blasts greater than or equal to 5%, LDH greater than 200 U/I (upper normal range), hemoglobin less than or equal to 9 g/dl, and platelets less than or equal to 100.000/μI. As a function of their total score, patients were divided into three risk groups  Since the classification of CMML as MPN/MDS, CMML is often excluded from clinical trials. To draw attention back to CMML we tried to better describe patients with regard to their relationship to MDS and MPN.
|WHO 2008||FAB 1982||FAB 1998|
|Classified as MDS/MPD||Classified as MDS||Classified as|
|blood monocytosis >1×109/l||blood monocytosis >1×109/l||blood monocytosis >1×109/l|
|CMMLI<10% blasts in BM||< 20% blasts in BM, < 5%||< 20% blasts in BM,|
|CMMLII10% -19% blasts in BM, 5% -19% blasts in PB, or||blasts in PB||< 5% blasts in PB|
|auer rods present irrespective of blast count|
|No Philadelphia chromosome or|
|BCR/Abl fusion gene|
|No rearrangement of PDGFA or PDGFB|
|Dysplasia in one or more myeloid lineages. If|
|myelodysplasia is absent or minimal, the diagnosis of CMML|
|may still be made if the other requirements are met and|
|an acquired, clonol cytogenetic or molecular genetic|
|abnormality is present|
|monocytosis has persisted for at least 3 month|
|all other causes of monocytosis have been excluded|
BM=bone marrow, PB= peripheral blood
Our retrospective study included 2507 patients from the Duesseldorf MDS registry, diagnosed between 1975 and 2012, who received best supportive care or disease specific therapy except for allogeneic transplantation. The diagnostic procedures were the same as reported earlier and . Diagnoses were made according to the proposals of the WHO 2008 classification  . Cytogenetic analyses at the time of diagnosis were performed at the Institute of Human Genetics, Heinrich-Heine University, Duesseldorf. The clinical data were gathered from the original patients’ charts. Follow-up data were obtained from our outpatient department or by contacting the primary care physician. Local ethics approval for this study and informed consent were obtained. Patients were censored at the date of last follow-up. Overall survival was calculated using the Kaplan-Meier method, to compare the parameter LDH, hemoglobin etc. the Mann-Whitney-U test was used.
We included 386 patients with CMML into our analysis. In a first step, we reanalyzed the prognostic meaning of medullary blast count in CMML and found that not only a separation of CMML I from CMML II using the 10% blast cut-off, but also a separation using a blast cut-off of 5% is of importance. Patients with less than 5% of medullary blast count had a better survival time than CMML I and CMML II patients, corresponding median survival times were 31 months (ms), 19 ms and 13 ms, respectively (p < 0.0001) ( figure 1 ). As a consequence, we newly defined a group of CMML patients with less than 5% blast, called CMML 0.
Then we analyzed, if dysplastic and proliferative CMML types at the cut-off of a white blood cell count of 13,000/μl differ with regard to hematological characteristics and prognosis. According to the above mentioned grouping system we identified 204 dysplastic CMML patients, subdivided into 61 dysplastic CMML 0-, 105 dysplastic CMML I- and 38 dysplastic CMML II-patients and further 182 proliferative CMML patients subdivided into 40 proliferative CMML 0-, 99 proliferative CMML I- and 43 proliferative CMML II-patients. Obviously, all CMML types were equally distributed when comparing dysplastic and proliferative types. For comparison with patients with MDS we included 1293 RCMD patients, 387 RAEB I patients and 441 RAEB II patients. Patients’ characteristics are summarized in Table 2 .
|n||m in %||f in%||age median and range||median survival time in month||median and range WBC/μl||median and range Hemoglobin g/dl||median and range platelet count/μl||median and range LDH U/l||median and range monocyte count/μl||karyotype risk group according to CPSS(%)||CPSS risk groups (%)||Duesseldorf Score risk groups (%)||Cumulative risk of AM L after 2 years %|
|CMML0||101||65||35||73 [48-90]||31||10300[200-138000]||10.9 [4.8-15.9]||143 [8-978]||210 [103-1435]||2275 [1000-16370]||66||24||10||40||23||36||0||23||73||4||7|
|CMML0dysplastic||61||67||33||73[52-88]||48||6400 [2000-12100]||10.8 [5.6-15.9]||139 [8-863]||208 [103-1200]||1587 [1000-4708]||75||17||8||66||22||11||0||26||70||4||4|
|CMML0 proliferative||40||60||40||74 [48-90]||17||20650 [13400-138000]||11.6 [4.8-15.3]||145 [21-978]||242 [152-1435]||3815 [1098-16370]||53||35||12||0||25||75||0||19||78||3||12|
|CMMLI||204||64||36||73 [30-96]||19||12150 [1500-147800]||10.7 [2.5-16.3]||104 [1-979]||219 [75-1176]||2484 [1000-44340]||67||19||14||30||32||32||5||0||62||38||18|
|CMMLI dysplastic||105||65||35||72 [43-96]||29||6100 [1500-12600]||10.5 [4.7-16.3]||99 [6-744]||200 [95-1176]||1365 [1000-12600]||71||19||10||60||18||22||0||0||64||36||13|
|CMMLI proliferative||99||63||37||73 [30-91]||15||25600 [13000-147800]||10.8 [2.5-15.5]||112 [1-979]||242 [75-796]||5480 [1000-44340]||62||19||19||0||50||38||12||0||60||40||21|
|CMML II||81||63||37||71 [32-91]||13||15300 [2700-145000]||9.9 [3.9-15.7]||84 [2-1069]||238 [98-1350]||3119 [1000-88450]||65||21||14||0||10||90||0||0||44||56||36|
|CMML II dysplastic||38||58||42||69 [32-88]||17||6100 [2700-12600]||10.3 [4.7-15.7]||89 [2-1069]||199 [98-1350]||1640 [1000-7378]||64||22||14||0||18||82||0||0||58||42||35|
|CMML II proliferative||43||67||33||74 [44-91]||10||30100 [13800-145000]||9.7 [3.9-15.1]||80 [3-907]||262 [144-904]||8161 [1896-88450]||67||20||13||0||0||75||25||0||32||68||37|
|RCMD||1293||57||43||72 [19-94]||34||3900 [4-46000]||9.4 [2.2-16.9]||132 [1-1540]||195 [2-2500]||200 [0-4774]||10|
|RAEBI||387||58||42||70 [24-105]||18||3200 [4-57600]||9.3 [2.6-17.5]||94 [0-1408]||208 [0-3800]||151 [0-5640]||24|
|RAEBII||441||58||42||71 [18-93]||11||2800 [4-139000]||9.1 [3.4-15.9]||80 [3-809]||222 [0-4680]||99 [0-4488]||44|
WBC= white blood cell count, LDH= laktatdehydrogenase, CPSS=CMML Prognostic Scorring System, m=male, f=female, n= number of patients
3.1. Hematological features
CMML 0 patients had significant higher hemoglobin (Hb)-levels in comparison to RCMD patients (10.9 g/dl vs 9.4 g/dl, p < 0.001), CMML I patients had higher Hb-levels in comparison to their dysplastic counterpart RAEB I as well (10.7 g/dl vs 9.3 g/dl), as had CMML II (Hb 9.9 g/dl vs. 9.1 g/dl). Median platelet counts did not differ. Between the corresponding dysplastic and proliferative CMML subtypes hemoglobin levels and platelet counts did not differ. LDH levels were significantly higher in CMML I and II proliferative than in the CMML I and II dysplastic (p < 0.0001) (see Table 2 ). In CMML 0 the difference did not reach the significance level (p = 0.067).
The CPSS was available for 107 CMML patients. The patients were assigned to four risk groups low, intermediate I, intermediate II and high risk, median survival times were as follows: 63ms, 36ms, 15ms and 16ms (p = 0.024). 29 patients were in the low risk group, 12 CMML 0 dysplastic- and 17 CMML I dysplastic- patients, 27 patients were in the intermediate I group, 4 CMML 0 dysplastic-, 3 CMML 0 proliferative-, 5 CMML I dysplastic-, 13 CMML I proliferative-, 2 CMML II dysplastic–patients, 45 patients were in the intermediate II group 2 CMML 0 dysplastic-, 9 CMML 0 proliferative-, 6 CMML I dysplastic-, 10 CMML I proliferative-, 9 CMML II dysplastic- and 9 CMML II proliferative and 6 patients were in the high risk group 3 CMML I proliferative- and 3 CMML II proliferative patients.
Concerning the cytogenetics 88 patients were assigned into the low risk cytogenetic category, as defined by Such et al.  27 CMML 0, 42 CMML I, and 19 CMML II patients, 28 patients were assigned into intermediate risk cytogenetic category 10 CMML 0, 12 CMML I,and 6 CMML II patients, 17 patients had high risk cytogenetic findings 4 CMML 0, 9 CMML I,and 4 CMML II patients.
3.3. Duesseldorf score
According to the Duesseldorf Score CMML patients are assigned to 3 risk groups low intermediate and high risk; median survival times were: 87ms (19 patients, 13 CMML 0 dysplastic- and 6 CMML 0 proliferative- patients) vs. 28ms (200 patients, 35 CMML 0 dysplastic-, 25 CMML 0 proliferative-, 56 CMML I dysplastic-, 53 CMML I proliferative-, 19 CMML II dysplastic- and 12 CMML II proliferative- patients) vs 11 ms (110 patients, 2 CMML 0 dysplastic-, 1 CMML 0 proliferative-, 32 CMML I dysplastic-, 35 CMML I proliferative-, 14 CMML II dysplastic- and 26 CMML II proliferative- patients) (p < 0.001).
3.4. Median survival
First we compared the median survival times between the newly created subgroups of CMML, CMML 0, I and II which differed significantly from each other (31 ms, 19 ms, 13 ms respectively) (p < 0.0001) (see figure 1 ).
Median survival times between the corresponding dysplastic and proliferative subgroups with the same blast counts differed: CMML 0 dysplastic 48 ms and CMML 0 proliferative 17 ms (p = 0.03), CMML I dysplastic 29 ms and CMML I proliferative 15 ms (p = 0.008) and CMML II dysplastic 17 ms and CMML II proliferative 10 ms (p = 0.09) (seeFigure 2, Figure 3, and Figure 4).
In the next step we compared the median survival times of the CMML subtypes with their dysplastic MDS counterparts. There was no difference between the corresponding pairs in median survival times in the dysplastic types: CMML 0 48 ms and RCMD 34 ms and CMML II 19 ms and RAEB II 18 ms (p values n.s). The dysplastic CMML I patients had significantly longer median survival time of 29 ms in comparison to RAEB I 18 ms (p= 0.004). CMML 0 proliferative had a worse survival than RCMD (17 vs 34 ms, p = 0.01). The median survival times between CMML I proliferative and RAEB I and CMML II proliferative and RAEB II did not differ (15 ms vs 18 ms and 10 ms vs 11ms, respectively).
3.5. AML evolution
AML rates at 2 years after diagnosis differed between CMML 0, I and II with 7%, 18% and 36% of patients, respectively (p < 0.0001) (see figure 5 ). AML rates between the corresponding dysplastic and proliferative subgroups did not differ as well as corresponding dysplastic CMML and MDS subgroups (see Table 2 ).
Based on data of 386 patients with CMML we were the first, who could show that in parallel to the myelodysplastic syndromes a group of patients with CMML and less than 5% blasts in the bone marrow has a higher median survival than those with 5 to 9% blasts. Besides, we showed that proliferative CMML types do worse as compared to dysplastic types, even after separating 3 CMML groups according to their medullary blast counts. Progression rates of the CMML with less than 5% medullary blast count are significantly lower as compared to CMML I as well as CMML II types. In addition we could show that erythroid hematopoietic insufficiency of all CMML types is not as severe as in RCMD, RAEB I, and RAEB II. Such et al. utilized the prognostic impact of proliferation and increased medullary blast counts, together with cytogenetic categories and hematopoietic insufficiency ending up in 4 different risk groups that nicely identifies patients with different prognosis  . We also showed that the Duesseldorf Score works well in CMML and we were able to validate the Duesseldorf score for CMML 0 as well. The group of CMML 0 patients with an available CPSS was too small to be significantly different according to CPSS-risk groups. This should be validated with a larger and independent data set.
These results fit into the theory of a correlation of hematopoietic insufficiency and proportion of a malignant cell clone in the marrow, which is well described in the myelodysplastic syndromes - with increasing medullary blast count the risk of progression rises and median survival drops. Up to now, no study addressed this topic in detail, as most reports only separated patients with 5-9% versus 10-19%   . Based on our results, it is justified to separate a CMML “0” presenting with less than 5% medullary blasts, as this group, both in its dysplastic as well as in its proliferative shape has a better prognosis and a lower risk of progression to AML than CMML I and II.
Initially, our intention was to analyze if dysplastic CMML types could be reclassified within the MDS types, i.e. RCMD, RAEB I and RAEB II. However, we could show that with regard to hematopoietic characteristics as well as cytogenetic features, dysplastic CMML is closer to proliferative CMML as to RCMD, RAEB I, and RAEB II. Erythroid hematopoietic insufficiency is less severe in CMML, and cytogenetic aberrations are less frequent in CMML as compared to MDS types. That is properly why CMML I in our cohort has a longer median survival time than RAEB I, median hemoglobin levels are 9.3 g/dl in RAEB I and 10.5 g/dl in CMML I, further 21% of RAEB I patients have high or very high risk cytogenetics according to IPSS-R whereas only 10% of CMML I have high risk cytogenetics according to CPSS.
Earlier critics on the WHO-classification of CMML should be extenuated.
In summary, we propose to separate CMML with less than 5% medullary blast count from the CMML I group and to separate dysplastic from proliferative types within CMML 0, CMML I, and CMML II. This refined categorization potentially has impact on treatment decision making.
Schuler E. analysed the data and wrote the paper, Schroeder M. performed the research, Neukirchen J. contributed essential data, Strupp C. contributed essential data, Xicoy B. analysed the data, Kündgen A. analysed the data, Hildebrandt B. contributed essential data, Haas R, analysed the data, Gattermann N. designed the research study and analysed the data, Germing U designed the research study, analysed the data and wrote the paper.
The authors state that they have no conflicts of interest.
-  Brunning RD OA, Germing U, Le Beau MM, Porwit A, Bauman I, Vardiman JW, Hellström-Lindberg E. Myelodysplastic Syndromes. In: Swerdlow SH CE, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman JW., ed. WHO classification of tumours of haematopoetic and lymphoid tissues: IARC Lyon 2008.
-  J.M. Bennett, D. Catovsky, M.T. Daniel, G. Flandrin, D.A. Galton, H.R. Gralnick, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol.. 1982 Jun;51(2):189-199
-  Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick H, et al. The chronic myeloid leukaemias: guidelines for distinguishing chronic granulocytic, atypical chronic myeloid, and chronic myelomonocytic leukaemia. Proposals by the French-American-British Cooperative Leukaemia Group. Br J Haematol. 1994 Aug;87(4):746-54.
-  U. Germing, N. Gattermann, H. Minning, A. Heyll, C. Aul. Problems in the classification of CMML--dysplastic versus proliferative type. Leuk Res.. 1998 Oct;22(10):871-878 Crossref
-  P. Fenaux, P. Morel, J.L. Lai. Cytogenetics of myelodysplastic syndromes. Semin Hematol.. 1996 Apr;33(2):127-138
-  E. Such, J. Cervera, D. Costa, F. Sole, T. Vallespi, E. Luno, et al. Cytogenetic risk stratification in chronic myelomonocytic leukemia. Haematologica.. 2011 Mar;96(3):375-383 Crossref
-  O. Abdel-Wahab, A. Mullally, C. Hedvat, G. Garcia-Manero, J. Patel, M. Wadleigh, et al. Genetic characterization of TET1, TET2, and TET3 alterations in myeloid malignancies. Blood.. 2009 Jul 2;114(1):144-147 Crossref
-  A. Kohlmann, V. Grossmann, H.U. Klein, S. Schindela, T. Weiss, B. Kazak, et al. Next-generation sequencing technology reveals a characteristic pattern of molecular mutations in 72.8% of chronic myelomonocytic leukemia by detecting frequent alterations in TET2, CBL, RAS, and RUNX1. J Clin Oncol.. 2010 Aug 20;28(24) 3858-65
-  M. Meggendorfer, A. Roller, T. Haferlach, C. Eder, F. Dicker, V. Grossmann, et al. SRSF2 mutations in 275 cases with chronic myelomonocytic leukemia (CMML). Blood.. 2012 Oct 11;120(15):3080-3088 Crossref
-  R. Itzykson, O. Kosmider, A. Renneville, V. Gelsi-Boyer, M. Meggendorfer, M. Morabito, et al. Prognostic score including gene mutations in chronic myelomonocytic leukemia. J Clin Oncol.. 2013 Jul 1;31(19):2428-2436 Crossref
-  O. Kosmider, V. Gelsi-Boyer, M. Ciudad, C. Racoeur, V. Jooste, N. Vey, et al. TET2 gene mutation is a frequent and adverse event in chronic myelomonocytic leukemia. Haematologica.. 2009 Dec;94(12):1676-1681 Crossref
-  F. Onida, H.M. Kantarjian, T.L. Smith, G. Ball, M.J. Keating, E.H. Estey, et al. Prognostic factors and scoring systems in chronic myelomonocytic leukemia: a retrospective analysis of 213 patients. Blood.. 2002 Feb 1;99(3):840-849 Crossref
-  C. Aul, N. Gattermann, A. Heyll, U. Germing, G. Derigs, W. Schneider. Primary myelodysplastic syndromes: analysis of prognostic factors in 235 patients and proposals for an improved scoring system. Leukemia.. 1992 Jan;6(1):52-59
-  A. Worsley, D.G. Oscier, J. Stevens, S. Darlow, A. Figes, G.J. Mufti, et al. Prognostic features of chronic myelomonocytic leukaemia: a modified Bournemouth score gives the best prediction of survival. Br J Haematol.. 1988 Jan;68(1):17-21 Crossref
-  I. Gonzalez-Medina, J. Bueno, A. Torrequebrada, A. Lopez, T. Vallespi, I. Massague. Two groups of chronic myelomonocytic leukaemia: myelodysplastic and myeloproliferative Prognostic implications in a series of a single center. Leuk Res.. 2002 Sep 26;9 821-4
-  E. Such, U. Germing, L. Malcovati, J. Cervera, A. Kuendgen, M.G. Della Porta, et al. Development and validation of a prognostic scoring system for patients with chronic myelomonocytic leukemia. Blood.. 2013 Apr 11;121(15) 3005-15
-  U. Germing, C. Strupp, A. Kundgen, D. Bowen, C. Aul, R. Haas, et al. No increase in age-specific incidence of myelodysplastic syndromes. Haematologica.. 2004 Aug;89(8):905-910
-  U. Germing, C. Strupp, A. Kuendgen, S. Isa, S. Knipp, B. Hildebrandt, et al. Prospective validation of the WHO proposals for the classification of myelodysplastic syndromes. Haematologica.. 2006 Dec;91(12):1596-1604
-  G.F. Sanz, M.A. Sanz, T. Vallespi, M.C. Canizo, M. Torrabadella, S. Garcia, et al. Two regression models and a scoring system for predicting survival and planning treatment in myelodysplastic syndromes: a multivariate analysis of prognostic factors in 370 patients. Blood.. 1989 Jul;74(1):395-408
a Department of Hematology, Oncology and Clinical Immunology, Heinrich-Heine-University, Düsseldorf, Germany
b Department of Human Genetics and Anthropology, Heinrich-Heine-University, Düsseldorf, Germany
© 2014 Published by Elsevier B.V.