Case 2 -

Refractory cytopenia with multilineage dysplasia (RCMD) Sa Wang, UT MD Anderson Cancer Ctr, Houston, TX

Click on slide thumbnail images for an enlarged view.

If you have any difficulties viewing these slides, email the webmaster.

Introduction: The classification of bone marrow neoplasms manifesting with erythroid predominance has been fraught with controversy since their original description. For myeloid neoplasms with ≥50% marrow erythroid elements that are not therapy-related or AML with recurrent cytogenetic abnormalities, the blast count plays a critical role in further classification. The WHO recommendation is to first enumerate myeloblasts as a proportion of total cells: if they are 20% or more, most of these cases will be classified as AML with myelodysplasia-related changes (AML-MRC) either due to a prior history of MDS, high-risk karyotype, or significant morphologic dysplasia (as discussed above). If myeloblasts are less than 20% of total cells, the recommendation is to count the myeloblasts as a proportion of the non-erythroid elements: if 20% or more, the diagnosis is AEL (acute erythoid leukemia) (a subtype of AML-NOS), and if less than 20%, the case is classified as MDS. The classification algorithm is illustrated in Figure 1. Clinical History: 84 year old woman, with a history of an artificial heart valve, noticed significant fatigue and paleness around May 2009. Upon evaluation she was found to be pancytopenic. She received red cell transfusion, and bone marrow (BM) biopsy and aspirate was performed: CBC at the time of BM biopsy: WBC: 2.3 with 70% neutrophils; 28% lymphocytes and 2% monocytes; Hb: 6.8; Hct 23.2%; MCV 92; Platelet: 32 Bone Marrow: 80% cellularity; trilineage dyspoiesis; 1% blasts; 63% erythroid; 12% lymphocytes/plasma cells; and 25% all other myeloid series. Chromosomal analysis: 46, XX[20] Case 2 - Figure 1 BM biopsy 50x Case 2 - Figure 2 BM biopsy 100x Case 2 - Figure 3 BM crush Case 2 - Figure 4 BM smear Case 2 - Figure 5 CD34 Case 2 - Figure 6 E-cadherin-2 Case 2 - Figure 7 glycophorin Case 2 - Figure 8 iron Case 2 - Figure 9 Diagnosis: Refractory cytopenia with multilineage dysplasia (RCMD); Interim history: She received 5-azacitidine from 09/28/09 to 07/21/10 with an initially significant improvement in transfusion dependency. In August 2011, her counts deteriorated. The patient is now referred back here for an assessment of the situation and treatment recommendations. CBC at this visit: WBC 0.6 with 12.5% neutrophils; 82.5% lymphocytes; 5% eosinophils; Hb 7.1, Hct 24.1%; MCV 94; Platelet 19 Bone Marrow: Cellularity: 70%; Trilineage dyspoiesis; Blasts: 8%; Erythroid precursors: 72%; other maturing myeloid elements: 9%. Immunohistochemistry: CD34: 10% positive cells E-Cadherin: many erythroid precursors Chromosomal Analysis: 61-58,XX,+1,+2,+6,+7,+9,+16,+16,-21[cp2]]/ 46, XX[18]; FISH for chromosome 8, trisomy 8: 4%, tetrasomy 8: 2% Case Discussion RAEB with erythroid hyperplasia (≥50%) versus AEL share very similar morphological features. Erythroid lineage often demonstrates variable degree of dysplasia, such as nuclear irregularity, binucleation, nuclear fragmentation, vacuolization of cytoplasm, and coarse basophilic stippling of the cytoplasm. Ring sideroblasts are a frequent finding in RAEB-E and AEL, but appear to have no clinical or prognostic significance. Myeloblasts can be distinguished from pronormoblasts by their smaller size, more scant cytoplasm that is pale and far less basophilic than that of pronormoblasts, and more irregularly dispersed nuclear chromatin. Myeloblasts should always be counted separately from pronormoblasts; the latter are usually counted together with all erythroid elements and a separate enumeration of pronormoblasts is only required for a diagnosis of pure erythroid leukemia (PEL). A high-quality bone marrow aspirate (or well-prepared touch imprints) must be examined to assess dysplasia and quantify blasts and erythroid elements. A 500-cell count of the bone marrow aspirate should be performed in all myeloid neoplasms with erythroid predominance to ensure an accurate count, particularly since small differences in the percentages of blasts or erythroids may profoundly affect the diagnosis. The bone marrow biopsy is also an essential component of the workup of myeloid neoplasms with erythroid proliferation, particularly in cases where the aspirate may be compromised or hemodilute due to marrow fibrosis. Myeloblasts are usually CD34+ and CD117+ and thus can be identified and their numbers estimated in the biopsy (Figure 2E), although this should not substitute for a blast count on the aspirate or touch preparation. Early erythroid elements in both benign conditions and in myeloid neoplasms may variably express CD117, but are CD34 negative. Glycophorin can help to highlight erythroid elements; E-cadherin is the most sensitive and specific marker of benign and neoplastic pronormoblasts and effectively stains PEL cases. In addition to the count of myeloblasts, other subtle morphological differences are observed between AEL and RAEB-E. In AEL, the maturing myeloid elements are often markedly decreased, and the BM cellular component is mainly of erythrocytes with maturation; myeloblasts and some lymphocytes. In contrast, RAEB-E BM often contains more maturing myeloid elements. The border between a high-grade MDS with erythroid predominance versus AEL is rather blurred, and it can be particularly challenging in cases of AEL arising from MDS. A slight difference in blast counting may render two different diagnoses, and may subject patients to different clinical management: MDS or AML. To identify more subjective criteria, and clinicopatholoic features that help us to correctly make diagnosis are crucial. The 2008 WHO classification has emphasized the incorporation of clinical presentation, cytogenetic and/or molecular genetic findings into myeloid neoplasm classification. However, the relative contribution of each of these parameters to the final diagnosis varies depending on the particular disease entity, and a multiparameter approach is required. In myeloid neoplasms with a prominent erythroid proliferation, cytogenetic changes, rather than blast counts have been found to be more critical in predicting outcomes of patients. AEL and RAEB-E may form a continuum analogous to therapy-related myeloid neoplasms, in which clinical outcomes are generally similar irrespective of blast count, but are strongly influenced by the karyotype risk group.