Printer Friendly

Acute erythroid leukemia.


Acute erythroid leukemia (AEL) is an uncommon type of acute myeloid leukemia (AML) and the criteria for establishing the diagnosis have evolved substantially over time. The first case report of AEL appears to have been contributed in 1912 by Copelli, (1) who described a hematologic disorder, erythromatosis. It was Giovanni Di Guglielmo, however, who more fully developed the understanding of this disease in a series of articles (2,3) beginning in 1917. Di Guglielmo described 2 forms of what is now known as AEL. The more common variant, referred to as erythremic myelosis, and subsequently as erythroleukemia and erythroid/myeloid leukemia by others, is composed of immature erythroid and myeloid elements. Over time the myeloblasts progressively increase, and the disease evolves into AML. (4) Di Guglielmo described a second variant characterized by a pure normoblastic proliferation, acute erythremia, subsequently referred to by others as true erythroleukemia, minimally differentiated erythroleukemia, and pure erythroid leukemia. Later, Dameshek and others on advocated the terms Di Guglielmo syndrome and Di Guglielmo disease in honor of Di Guglielmo's contribution to the understanding of this disease. (5)

In 1976, the French-American-British (FAB) Cooperative Group designated AEL as AML-M6, defined as the presence of 30% or more blasts, with the denominator being that all nucleated bone marrow cells and 10% or more erythroid precursors show dyserythropoiesis. (6) In the subsequent revision of the FAB classification in 1985, (7) the criteria for the diagnosis of AML-M6 were refined to include the following: (1) erythroblasts that represent 50% or more of all nucleated bone marrow cells; (2) prominent dyserythropoiesis; (3) and myeloblasts that represent 30% or more of the nonerythroid cells in the bone marrow. The entity currently known as pure erythroid leukemia did not meet the requirements of the FAB classification for the diagnosis of acute leukemia and was therefore most often classified as a myelodysplastic syndrome (MDS) with erythroid predominance, usually in the category of refractory anemia with excess of blasts in transformation (RAEB-T).

Although the FAB classification has not been formally updated since its 1985 revision, other investigators modified the FAB classification informally by expanding the M6 category to include M6a and M6b subtypes. M6a was equivalent to the initial designation of M6 in the FAB scheme. M6b was more akin to the pure erythroid leukemia subtype of AEL, although different investigators had different definitions. Mazzella and colleagues (8) and Kowal-Vern and colleagues, (9) for example, defined M6b and proposed a third category, M6c. As with M6a, in both M6b and M6c the erythroid precursors represented 50% or more of the nucleated cells. In M6b, 30% or more pronormoblasts were present, with an insignificant myeloblastic component. The M6c category was defined by the presence of 30% or more myeloblasts and 30% or more pronormoblasts. Mazzella and colleagues suggested that the M6a, M6b, and M6c types most likely represented different phases of the same disease.

The 2001 World Health Organization (WHO) classification (10) lowered the blast percentage threshold to at least 20% for establishing the diagnosis of all types of AML, thereby eliminating the RAEB-T category. This classification also used the term acute erythroid leukemia and recognized 2 subtypes. The erythroid/myeloid subtype was defined by a neoplastic proliferation of myeloblasts (20% or more of the nonerythroid cells in bone marrow) in a background of erythroid hyperplasia (50% or more of erythroid precursors). The pure erythroid leukemia subtype was defined by the presence of immature cells committed exclusively to the erythroid lineage ([greater than or equal to] 80% of bone marrow cells), with no evidence of a significant myeloblastic component. The so-called M6c category was not specifically recognized in the WHO classification.

The 2008 WHO classification (11) closely follows the previous version, but makes the criteria for its diagnosis more rigorous. Cases with 20% or more blasts and dysplasia involving at least 50% of cells in 2 or more lineages are now moved to the category of AML with myelodysplasia-related changes, even in the face of erythroid predominance. The 2008 WHO classification recognizes that therapy, particularly erythropoietin, can cause erythroid predominance that may result in a morphologic picture that mimics AEL. Both the 2001 and 2008 versions of the WHO classification also recognized the category of therapy-related AML, another type of AML that can be associated with numerous erythroid precursors. A comparison of these classification schemes is listed in Table 1. One of the consequences of these refinements in criteria for the diagnosis of AEL is that this disease, in part, is a diagnosis of exclusion. Another consequence of these refinements is that the frequency of AEL has been reduced.

In the remainder of this review we discuss the clinical, morphologic, immunophenotypic, and cytogenetic findings in AEL. We also review the molecular data available, which is relatively scant. Because of the extreme rarity of the pure erythroid leukemia subtype, this review is heavily biased toward the erythroleukemia subtype of AEL.


Acute erythroid leukemia is a rare type of AML, representing less than 5% of all cases, characterized by a predominant ([greater than or equal to] 50%) erythroid population in the bone marrow. Most cases of AEL develop de novo, accounting for approximately 1% of all de novo AML, and the disease is not associated with any identifiable risk factors. Rare cases of de novo familial erythroleukemia, being autosomal dominant with variable penetrance, have been described. (12) Cases of AEL evolving from other antecedent diseases have been reported in the literature, and these cases are often referred to as so-called secondary AEL. Common antecedent diseases or factors include MDS (for example, refractory anemia with excess of blasts or refractory cytopenia with multilineage dysplasia), myeloproliferative neoplasms (for example, chronic myelogenous leukemia with erythroblastic crisis), (13) and exposure to toxins such as benzene. Secondary AEL also has been reported in patients with a history of other types of cancer treated with chemotherapy, immunosuppressive treatment, or ionizing radiation. Other than the de novo and familial categories of disease, in our view, many of the cases of secondary AEL (usually designated as erythroleukemia or AML-M6 in the literature) should be viewed with skepticism. With the current WHO classification, it seems highly likely that many of these cases would not be classified as AEL, and would fit better either as AML with myelodysplasia-related changes or therapy-related AML. Nevertheless, here we attempt to review the literature as it is, with the caveat that at least some cases in the literature designated as erythroleukemia or AML-M6 do not meet the current WHO classification criteria for AEL.


Acute erythroid leukemia is a disease of adults that primarily affects people older than 50 years. Some studies (9) have shown a bimodal age distribution, with a smaller peak below the age of 20 and a more definitive and broader peak in the seventh decade of life. The rare familial cases usually manifest in the sixth decade of life. (12) There is a slight male predominance. The clinical features are often relatively nonspecific, and include pallor, fever, and hepatosplenomegaly. Extramedullary presentation as a myeloid sarcoma is unusual in patients with AEL, but extramedullary sites can be involved and the diagnosis of AEL has been established by lymph node biopsy. (14) Anemia is often severe in patients with AEL, with a mean hemoglobin level of 7.5 g/dL reported in 1 study. (15) Thrombocytopenia and leukopenia, to varying degrees, are also common. Up to one-third of the patients can have hemorrhage, hepatomegaly, or splenomegaly. (14) Severe hemolysis can sometimes occur. (16) Patients can be symptomatic for an interval of time before diagnosis, but usually the diagnosis is established in 1 to 3 months. It is rare for patients with AEL to have symptoms that last for longer than 6 months before an initial diagnosis. We also have observed patients with AEL who, although symptomatic, were highly functional at time of diagnosis, leading clinicians to initially doubt the diagnosis and consider trial therapy for nutritional deficiency. In these few patients, the disease clinically declared itself more fully within a short follow-up interval.


The diagnosis of AEL requires morphologic examination of the bone marrow. Although a striking erythroblastemia can sometimes be identified in affected patients, examination of peripheral blood smears can be misleading because blood smears are often devoid of blasts and show only cytopenias (Figure 1). Nonspecific red blood cell abnormalities can be present, such as anisocytosis, poikilocytosis, anisochromia, basophilic stippling, schistocytes, and erythrocytes that are poorly hemoglobinized. Normoblasts may or may not be observed in the blood smear. Other nonspecific findings that can be observed in the blood smear include mild to marked neutropenia and thrombocytopenia, and pseudo-Pelger-Huet neutrophils.

In bone marrow aspirate smears and touch imprints, by definition, erythroid precursors predominate ([greater than or equal to] 50%) in AEL. In the erythroleukemia subtype, maturation of erythroid precursors is often left shifted and dysplasia is identified in all maturation stages. Findings of dysplasia include 1 or more of the following: abundant megaloblastoid forms, nuclear budding, bizarre nuclear shapes, multinucleation, foamy cytoplasmic vacuoles, or cytoplasmic pseudopods (Figure 2, A and B). Prominent multilineage dysplasia is common, but variable, involving granulocytes (Figure 2, C) or megakaryocytes (Figure 2, D) or both. Auer rods are unusual but can be seen in myeloblasts. The natural history of the erythroleukemia subtype is for the disease to accrue blasts of myeloid or myelomonocytic lineage with variable differentiation. Thus, over time the morphologic picture of the erythroleukemia subtype can progress to AML with minimal differentiation, AML without maturation, AML with maturation, or acute myelomonocytic leukemia.

In the pure erythroid leukemia subtype of AEL, most ([greater than or equal to] 80%) of the cells in the aspirate smears are erythroid precursors (Figure 3). Erythroid maturation is left shifted with increased pronormoblasts. Pronormoblasts are intermediate in size to large, with round nuclei, fine chromatin, often prominent nucleoli, and deeply basophilic and agranular cytoplasm. Cytoplasmic vacuoles are variably present and can be prominent in pronormoblasts (Figure 3, A). In occasional cases erythroblasts can be smaller, in the size range of lymphoblasts or lymphoma cells. Dysplasia is common in the erythroid elements but is generally a minor feature in the other lineages. Myeloblasts are very few in the pure erythroid leukemia subtype of AEL.

The bone marrow aspirate clot and biopsy specimens are usually hypercellular in both subtypes of AEL. Dysplasia of megakaryocytes often can be appreciated in the erythroleukemia subtype, whereas the neoplasm can appear completely undifferentiated in cases of the pure erythroid leukemia subtype (Figure 3, B). Markedly increased erythroid precursors can be appreciated in sections of the clot and biopsy specimens of AEL cases, but it is difficult to exactly count the various cell types or assess dysplasia in tissue sections of the clot and biopsy specimens, particularly in routinely processed, paraffin-embedded specimens. Perhaps more data can be derived from clot and biopsy specimens that are embedded in plastic, allowing the preparation of very thin tissue sections, but we have no personal experience with these methods in the study of AEL.

Many cytochemical stains have been used to characterize the cells of AEL. In both the erythroleukemia and pure erythroid leukemia subtypes, erythroblasts are often highlighted by the periodic acid-Schiff (PAS) reaction in either a globular or diffuse pattern, with the diffuse pattern occurring in more mature erythroblasts (Figure 3, C). Periodic acid-Schiff positivity in AEL is thought to reflect a cytoplasmic maturation defect. An iron stain usually shows increased iron stores and may show ring sideroblasts (Figure 2, B [inset]). Myeloblasts (but not erythroblasts) are usually positive for myeloperoxidase (MPO), Sudan Black B, and, often, chloroacetate esterase. A nonspecific esterase reaction will highlight a monocytic component that can predominate in a subset of cases. In the pure erythroid leukemia subtype, PAS staining is commonly positive, often in cytoplasmic vacuoles in a blocklike pattern (Figure 3, C). Ring sideroblasts are uncommon. The blasts are negative for MPO (Figure 3, D), Sudan Black B, and chloroacetate esterase. Erythroid precursors also can be positive for the [alpha]-naphthyl acetate esterase and [alpha]-naphthyl butyrate esterase.



Immunophenotyping of AEL is best performed by flow cytometry because a wider array of antibodies is available than with immunohistochemistry. Erythroblasts in both subtypes of AEL express erythrocyte-associated antigens, and expression correlates with the degree of differentiation or maturation of the neoplastic cells. Erythroblasts usually express CD71 (transferrin receptor), but occasionally CD71 expression can be aberrantly dim. Erythroblasts variably express hemoglobin A, glycophorin A, spectrin, ABH blood group antigens, and HLA-DR, and are negative for myeloid-associated markers such as MPO. The Gerbich blood group (Gero) antibody, carbonic anhydrase 1, and CD36 can be expressed by more immature erythroblasts. The myeloblasts of the erythroleukemia subtype express myeloid-associated markers including CD13, CD33, CD117 (KIT), and MPO, with or without expression of CD34 and HLA-DR. Figure 4 shows an example, using the CD45 and side scatter gating strategy (Figure 4, A), to illustrate myeloblasts in one case of AEL that expressed CD33, CD34 (Figure 4, B), CD13 (Figure 4, C), and MPO (Figure 4, D).

Others (17) have reported that cases of AEL, particularly the pure erythroid leukemia subtype, can express the megakaryocyte-associated antigens CD41 and CD61. Expression is often of low intensity (dim) or partial. Whether these cases are truly AEL or a mixed lineage neoplasm with both erythroid and megakaryocytic differentiation is unclear, but our bias is to keep the category of AEL pure.

Immunohistochemistry has a role in the workup of cases of AEL, particularly in cases in which the aspirate material is scant. Here we mention markers or applications not already addressed in the flow cytometry section above. The anti-CD61 antibody can be useful for identifying micromegakaryocytes. Blasts in the erythroleukemia type of AEL can be positive for lysozyme or CD68. Results with MPO are typically negative. In our experience, the blasts in the pure erythroid type of AEL are often negative for all erythroid-associated antibodies assessed by immunohistochemistry because many of the target antigens require some degree of erythroid maturation to be expressed and very dim expression may not be detectable by immunohistochemistry (Figure 3, E). Most results with myeloid-associated antigens are negative with the exception of CD117, which can be weakly expressed in a subset of cases of the pure erythroid subtype (similar to normal pronormoblasts). Staining for terminal deoxynucleotidyl transferase is usually negative in AEL.




Most cytogenetic data available for AEL cases are derived from conventional cytogenetic analysis of the most common subtype, erythroleukemia. However, as stated above, a subset of cases in the literature that were classified by using older terms that appear equivalent to AEL probably do not fulfill the criteria for this disease in the current WHO classification. This subset of cases is difficult to tease out of the various individual publications and we therefore present the data as it is, as not all relevant information is provided.

To date, no specific chromosome abnormalities have been described in AEL. Depending on the study, 50% to 80% of patients have an abnormal karyotype. (8,18-20) Complex karyotypes with multiple structural abnormalities are common. The most frequent abnormalities include monosomy 5 or del(5q), monosomy 7 or del(7q) and trisomy 8. (21) Complex karyotypes (3 or more cytogenetic abnormalities) also have been reported in AEL cases. (22,23) Rare cases of erythroleukemia with der(1;7)(q10;p10) have been reported, especially in East Asian individuals. (24,25) Additionally, t(8;16)(p11.2;p13.3) in AEL has been reported to be associated with erythrophagocytosis and coagulopathy. (18) Cytogenetic features in AEL can be used to stratify patients into prognostic groups. Patients with -5/del(5q), -7/del(7q), inv(3q), +8, 11q abnormalities, 17p abnormalities, del(20q), +13, and complex karyotypes are generally associated with unfavorable outcomes.

Although we have no doubt that cytogenetic abnormalities correlate with prognosis, as has been shown in many studies of AML in general, the spectrum and types of the reported cytogenetic abnormalities in cases classified previously as erythroleukemia suggest that this "entity" has been highly heterogeneous. The high frequency of -5/del(5q) and -7/del(7q) supports the concept that a subset of reported cases of erythroleukemia are closely related to MDS, and that some reported cases may be better designated as AML with myelodysplasia-related changes. This was predicted in the 2008 WHO classification. Cases of so-called erythroleukemia with t(9;22)/BCR-ABL1, as reported in the literature, would most likely be designated as chronic myelogenous leukemia currently. Similarly, cases reported as erythroleukemia that carry chromosomal translocations associated with specific types of AML, such as t(8;21) (q22;q22)/RUNX1-RUNX1T1, would currently be classified according to the chromosomal abnormality. For example, Virchis et al (26) described a case of erythroleukemia associated with t(15;17)(PML-RARA). This case would be classified as acute promyelocytic leukemia with current criteria. Virtually all cases of erythroleukemia in patients with a history of cancer that were treated with chemotherapy or radiation therapy would be classified as therapy-related AML with erythroid predominance with the current WHO classification.


Because of the rarity of AEL, relatively few cases have been assessed at the molecular level, and the cases assessed have been primarily of the erythroleukemia subtype. In addition, the few cases of AEL assessed by molecular methods are found within large studies of AML including all types, without a focus on AEL. As a result, less than a complete picture is available, and one should certainly use some caution when interpreting some of these results.

The nucleophosmin gene (NPM1) is commonly mutated in AML, and found in up to 50% to 60% of cases, especially in patients with normal karyotype. In the absence of coexisting fms-related tyrosine kinase 3 gene (FLT3) mutations, NPM1 mutations are thought to be a favorable prognostic marker for patients with AML. (27) In one study, (28) NPM1 mutation was found in 1 of 5 cases (20%) of AEL. Although a very small number of cases, this apparent frequency of 20% in AEL was second only to AMLs with monocytic differentiation (acute myelomonocytic or monocytic leukemia). (28) FLT3 is also commonly mutated in AML, and seen in up to 30% of adult cases. In AEL, however, FLT3 mutations are rare. (29) In our own limited experience, FLT3 mutations occur in approximately 10% of cases of AEL (Z.Z., L.J.M., unpublished data, October 2008). Nevertheless, FLT3 expression levels in AEL are reported to be among the highest of all AML types, suggesting that FLT3 may be involved in pathogenesis. (30) NRAS mutations are rare in AEL (Z.Z., L.J.M., unpublished data, October 2008) as compared to the 10% to 20% overall frequency in all types of AML. Mutation of Janus kinase 2 (JAK2), commonly found in myeloproliferative neoplasms, was identified in 1 of 12 cases (8.3%) of AEL by 1 group. (31) Auewarakul and colleagues (32) reported mutations of the core binding factor gene, RUNX1 (also known as AML1), which normally is involved in myeloid differentiation, in 2 of 7 (28.6%) AEL cases. The TP53 tumor suppressor gene was in the wild-type form in all 5 cases of AEL assessed in 1 study, (33) unlike the greater than 10% frequency of p53 mutations in AML overall. Taken together, this molecular profile suggests that AEL may arise by pathogenetic mechanisms that are distinct from those of other types of AML. This concept needs to be confirmed in studies with larger cohorts of patients with AEL.



There is a subset of cases of AEL, erythroleukemia subtype, in which the overall blast count can be low. Specifically, the overall blast count is in the range of 5% to 10%, far below the WHO classification blast cutoff for AML of at least 20%, but these cases have numerous erythroid precursors (70% to 90%). As a result, the blasts can represent 20% or more of the nonerythroid cells, thereby allowing the diagnosis of AEL. Selby and colleagues (34) have questioned whether these cases should be designated as AEL, as this diagnosis may lead to overly aggressive therapy for these patients. In our experience, a subset of patients with "low blast count" AEL do not have associated cytogenetic abnormalities, perhaps supporting the view that these cases are more akin to MDS than to AEL. (20) Follow-up data on patients such as these are needed to address this issue.


The differential diagnosis for AEL includes both neoplastic and benign diseases. A suggested diagnostic algorithm is illustrated in Figure 5. For the erythroleukemia subtype, the most important entities in the differential diagnosis include MDS with erythroid predominance, AML with myelodysplasia-related changes, and other types of AML with increased erythroid precursors. The most important distinguishing features for this differential diagnosis are blast percentage and the presence of multilineage dysplasia (2 or 3 lineages) in at least 50% of cells of a given lineage. In the WHO classification, this degree of dysplasia is considered a surrogate for cytogenetic abnormalities associated with MDS. (11)

Patients with MDS usually have anemia, and the bone marrow can show erythroid predominance, reflecting an attempt to compensate for the anemia. (35) Morphologically, erythroid dysplasia and ring sideroblasts are commonly observed in MDS as well as in the erythroleukemia subtype of AEL. The key distinguishing factor is the bone marrow blast percentage. If the blast count is less than 20% of all nucleated cells and also less than 20% of the nonerythroid elements, a diagnosis of MDS is most appropriate.

Acute myeloid leukemia with myelodysplasia-related changes is a recently proposed entity in the 2008 WHO classification. These cases are characterized by 20% or more blasts and multilineage dysplasia in at least 50% of cells in 2 or more lineages. Cases of AML with myelodysplasia-related changes can be erythroid rich, with 50% or more erythroid precursors, but these cases are not classified as AEL, erythroleukemia subtype with current WHO criteria. Moreover, cases with at least 20% peripheral blood or bone marrow blasts and MDS-related cytogenetic abnormalities or a prior history of MDS, even when associated with a population of erythroid precursors in the bone marrow of 50% or more, should be considered as AML with myelodysplasia-related changes rather than AEL.

Other types of AML also can be associated with increased numbers of erythroid precursors when they are initially diagnosed. In all of these cases, blasts represent 20% or more of all nucleated bone marrow cells. When erythroid precursors represent 50% or more of nucleated cells, in our experience, the most common explanation is prior erythropoietin therapy, which can increase erythroid precursors (including immature erythroblasts) and cause dyserythropoiesis. In the face of erythropoietin therapy it is difficult to reliably establish the diagnosis of AML, unless at time of relapse in a patient with history of this disease. We recently reviewed a large number of cases in which erythroid precursors were predominant and the diagnosis of AEL (or older, equivalent terms) had been made or suggested for a number of years at The University of Texas M.D. Anderson Cancer Center (Houston). (36) According to the criteria of the 2008 WHO classification, an appreciable subset of these cases had a history of erythropoietin therapy, known mostly in retrospect, thereby calling the diagnosis of AEL into question. (36) Prior chemotherapy also can result in a rebound erythrocytosis and can complicate the analysis of follow-up specimens after therapy. As is illustrated in this discussion of the differential diagnosis, the diagnosis of erythroleukemia is, in part, one of exclusion.

The differential diagnosis of the pure erythroid leukemia subtype of AEL includes other types of AML, acute lymphoblastic leukemia (ALL), lymphomas, plasma cell myeloma, and several nonneoplastic disorders that are characterized by increased erythroid precursors with dyserythropoiesis. (37,38) Cases of the pure erythroid leukemia subtype in which there is clear evidence of erythroid maturation are relatively easier to recognize. However, many cases of the pure erythroid leukemia subtype of AEL show minimal or no erythroid maturation and are composed of numerous immature blasts. Although the blast cell cytoplasm is deeply basophilic, this feature is not specific. For these cases, cytochemistry and immunophenotypic analysis are necessary. The blasts of AML undifferentiated type (FAB-M0), for example, are negative for MPO by cytochemistry, similarly to the blasts of pure erythroid leukemia. However, the blasts of undifferentiated AML often express a variety of myeloid-associated antigens, as shown by flow cytometry or immunohistochemistry, and therefore this neoplasm can be reliably distinguished from pure erythroid leukemia. Acute megakaryoblastic leukemia can be more difficult to distinguish from pure erythroid leukemia. With cytochemistry, MPO staining is negative in both diseases and PAS can show cytoplasmic positivity in both. Immunophenotypic analysis is essential in this case. Cases of pure erythroid leukemia express 1 or more erythroid-associated antigens (eg, hemoglobin A, glycophorin, or others) and cases of acute megakaryoblastic leukemia express megakaryocyte-associated antigens such as CD41, CD61, and factor VIII. However, in highly immature neoplasms, many of these antigens may be absent. In addition, cases of pure erythroid leukemia are reported with partial expression of either CD41 or CD61. Whether these cases are better classified as acute mixed erythroid-megakaryoblastic leukemia or not is uncertain. Electron microscopy to show platelet peroxidase can be helpful to diagnose acute megakaryoblastic leukemia. The t(1;22)(p13;q13) anomaly and abnormalities involving chromosome 3q26 also support the diagnosis of acute megakaryoblastic leukemia. In some cases of pure erythroid leukemia, the blasts may be intermediate in size or small, falling into the size range of ALL and malignant lymphoma. Immunophenotyping will be helpful to distinguish these entities, as ALL and lymphomas express either B- or T-cell antigens, and ALL is usually positive for terminal deoxynucleotidyl transferase. The basophilic cytoplasm of erythroblasts and neoplastic plasma cells also can be confused morphologically. (37) Immunophenotypic analysis is useful for this differential diagnosis, as plasma cells express CD38, CD138, and cytoplasmic immunoglobulin, unlike the pure erythroid leukemia subtype of AEL.

A number of nonneoplastic diseases can cause erythroid predominance in the bone marrow and therefore must be excluded before the diagnosis of AEL can be established (Table 2). The most common nonneoplastic disorders that need to be excluded before the diagnosis of AEL is established include megaloblastic anemia due to nutritional deficiency (either vitamin B12 or folate), heavy metal intoxication (eg, arsenic), drug effects (such as antineoplastic agents or chloramphenicol), and congenital dyserythropoiesis. (38) Prior chemotherapy or erythropoietin therapy also can induce erythrocytosis and be associated with dyserythropoiesis in the bone marrow. A complete history and necessary laboratory tests can often help to rule out these possibilities. For some patients, a trial with vitamin B12 or folate may be helpful.


Clinically, patients with AEL are usually treated similarly to patients with other types of AML, not otherwise specified. (20) Stem cell transplantation (SCT) is potentially curative, but is associated with procedure-related morbidity and mortality, particularly for allogeneic SCT. Nevertheless, data have shown that SCT can substantially improve the outcome of this disease, with 5-year leukemia-free survival reaching approximately 60% after HLA-identical sibling SCT. (5) No therapeutic agents that target specific pathways or molecules are currently available for this disease, reflecting the lack of understanding of pathogenetic mechanisms. Clinical remission can be achieved for many patients when treated according to the standard chemotherapy protocols for AML, with cytarabine being the most active agent. Contrary to their antiapoptotic and progrowth effects, high-dose erythropoietin and granulocyte colony-stimulating factor have been reported to induce clinical remission in a few elderly patients. (39) If validated in larger, more rigorous clinical studies, this approach could serve as an alternative therapy for patients whose disease is refractory to or for patients who are ineligible for standard chemotherapy. Common issues during treatment of patients with AEL include primary induction failure, relapse, and toxicity of chemotherapeutic agents. Refractoriness or relapse may be explained by overexpression of the multidrug resistance gene product, P-glycoprotein. In one study, (40) AEL of the erythroleukemia subtype had very high levels of P-glycoprotein expression as compared to that of other types of AML. Modulators of multidrug resistance, such as cyclosporin A, quinidine, verapamil, and PSC 833, have been used in a clinical trial in an attempt to overcome the resistance and improve therapeutic response. (41)


In general, AEL has been associated with an aggressive clinical course. A less favorable outcome is observed in elderly patients, in patients who previously had a diagnosis of MDS, or in patients who have been treated with chemotherapy for another neoplasm before developing AML. As has been stated, many of these cases reported in the literature may not fulfill the criteria for AEL when using up-to-date classification criteria. In a recent large study by Santos and colleagues, (20) the pathologic diagnosis of AEL did not impart, by itself, a worse outcome and there was no difference in complete remission rate when compared with other types of AML, not otherwise specified.

Santos et al (42) and others (43) have emphasized that response to chemotherapy and length of survival for patients with AEL are dependent on a number of factors, with the most important being cytogenetic abnormalities. Liu and colleagues (44) showed that the complete remission rate of patients with AEL associated with an aberrant karyotype was significantly lower than that of corresponding patients with a normal karyotype (37% versus 83%). In another study, the complete remission of patients with AEL and 5q or 7q abnormalities was approximately 50%, with a median survival of 16 weeks, compared to 89% and 77 weeks for patients without these abnormalities. (15) MDR1 expression also correlates with unfavorable cytogenetic findings and may explain the poorer response to chemotherapy and shorter survival time. A history of MDS appears to be another unfavorable factor. The complete remission rate of patients with a history of MDS was 42.8%, compared to 85.2% for patients without previous MDS history. (44)


The entity of AEL has been somewhat controversial since its original conception by Di Guglielmo almost a century ago. Overt, qualitative abnormalities of erythroblasts are common in many AML cases, if methods for detecting erythroid differentiation more sensitive than light microscopy are used, which challenges the uniqueness of this neoplasm. (22,23) Perhaps as a response to this, the criteria for the diagnosis of AEL have become more rigorous with each new classification scheme. Furthermore, the pure erythroid leukemia subtype of AEL, not recognized in the original FAB classification, is now accepted as a subtype of AEL in the current WHO classification. Two net effects of the changing criteria have been (1) a decrease in the overall frequency of AEL; and (2) the calling into question of much of the data reported previously in the literature for this entity, as the patient group appears to be highly heterogeneous when using the criteria of older classification systems.

For the present, the diagnosis of AEL is, in large part, a morphologic diagnosis and a diagnosis of exclusion. It also should be a diagnosis established with caution, particularly for patients with an overall low blast count (ie, <10%), even when blasts represent 20% or more of all nonerythroid cells, as the diagnosis of AEL can lead to aggressive therapy for patients with this constellation of findings. In our opinion, it seems unlikely that increased understanding of this disease will come from the application of traditional pathologic methods. Molecular studies, preferably using high-throughput methods, are most likely needed for a better understanding of disease mechanisms and the development of diagnostic and prognostic markers.


(1.) Copelli M. Di una emopatia sistemizzata rappresentata da una imperplasia eritroblastica (eritromatosis). Pathologica. 1912;(4):460.

(2.) Di Guglielmo C. Ricerche di ematologia, I: un caso di eritroleucemia--megacariociti in circolo e loro funzione piastrinopoietica. Folia Med (Pavia). 1917;(13):386.

(3.) Dameshek W, Baldini M. The Di Guglielmo syndrome. Blood. 1958;13(2): 192-194.

(4.) Dameshek W. The DiGuglielmo syndrome revisited. Blood. 1969;34(5): 567-572.

(5.) Fouillard L, Labopin M, Gorin NC, et al. Hematopoietic stem cell transplantation for de novo erythroleukemia: a study of the European Group for Blood and Marrow Transplantation (EBMT). Blood. 2002;100(9):3135-3140.

(6.) Bennett JM, Catovsky D, Daniel MT, etal. Proposals for the classification of the acute leukaemias. French-American-British (FAB) co-operative group. Br J Haematol. 1976;33(4):451-458.

(7.) Bennett JM, Catovsky D, Daniel MT, et al. Proposed revised criteria for the classification of acute myeloid leukemia: a report of the French-American-British Cooperative Group. Ann Intern Med. 1985;103(4):620-625.

(8.) Mazzella FM, Kowal-Vern A, Shrit MA, et al. Acute erythroleukemia: evaluation of 48 cases with reference to classification, cell proliferation, cytogenetics, and prognosis. Am JCl in Pathol. 1998;110(5):590-598.

(9.) Kowal-Vern A, Mazzella FM, Cotelingam JD, Shrit MA, Rector JT, Schumacher HR. Diagnosis and characterization of acute erythroleukemia subsets by determining the percentages of myeloblasts and proerythroblasts in 69 cases. Am J Hematol. 2000;65(1):5-13.

(10.) Brunning RD, Matutes E, Flandrin G, et al. Acute myeloid leukemia not otherwise categorized. In: Jaffe ES, Harris NL, Stein H, Vardiman JW, eds. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2001:97-99. World Health Organization Classification of Tumours; vol 3.

(11.) Arber DA, Brunning RD, Orazi A, et al. Acute myeloid leukemia, not otherwise specified. In: Swerdlow SH, Campo E, Harris NL, et al, eds. Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2008:134136. World Health Organization Classification of Tumours; vol 2.

(12.) Novik Y, Marino P, Makower DF, Wiernik PH. Familial erythroleukemia: a distinct clinical and genetic type of familial leukemias. Leuk Lymphoma. 1998; 30(3-4):395-401.

(13.) Bennett JM, Catovsky D, Daniel MT, 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;87(4): 746-754.

(14.) Keifer J, Zaino R, Ballard JO. Erythroleukemic infiltration of a lymph node: use of hemoglobin immunohistochemical techniques in diagnosis. Hum Pathol. 1984;15(11):1090-1093.

(15.) Olopade OI, Thangavelu M, Larson RA, et al. Clinical, morphologic, and cytogenetic characteristics of 26 patients with acute erythroblastic leukemia. Blood. 1992;80(11):2873-2882.

(16.) Zebisch A, Sill H. Severe hemolysis as presenting sign of acute erythroleukemia. JCl in Oncol. 2008;26(2):330-331.

(17.) Daniels L, Guerti K, Vermeulen K, et al. Acute myeloid leukaemia of mixed megakaryocyte and erythroid origin: a case report and review of the literature. Acta Clin Belg. 2007;62(5):308-314.

(18.) Lessard M, Struski S, Leymarie V, et al. Cytogenetic study of 75 erythroleukemias. Cancer Genet Cytogenet. 2005;163(2):113-122.

(19.) Sharma P, Dhingra KK, Roy S, Singh T. An acute myeloid leukemia M6b blast crisis with giant proerythroblasts in chronic myeloid leukemia. J Pediatr Hematol Oncol. 2009;31(3):220-221.

(20.) Hasserjian RP, Zuo Z, Garcia C, et al. Acute erythroid leukemia: a reassessment using criteria refined in the 2008 WHO classification [published online ahead of print December 29, 2009]. Blood. 2009;115(10):1985-1992.

(21.) Schoch C, Haferlach T. Cytogenetics in acute myeloid leukemia. Curr Oncol Rep. 2002;4(5):390-397.

(22.) Niu Y, Chen SC, Jiang B, Li DG, Ge CW, Li RS. Erythroleukemia--a subtype of myelodysplastic syndrome? [in Chinese]. Zhongguo Shi Yan Xue Ye Xue Za Zhi. 2007;15(2):219-223.

(23.) Honda Y, Manabe A, Tsuchida M, et al. Clinicopathological characteristics of erythroblast-rich RAEB and AML M6a in children. Int J Hematol. 2008; 88(5):524-529.

(24.) Obara N, Shinagawa A, Komatsu T, Abe T, Ninomiya H. Derivative (1;7)(q10;p10)in apatientwith denovoacuteerythroblasticleukemia(AML-M6). Cancer Genet Cytogenet. 1999;115(1):62-64.

(25.) Park TS, Cheong JW, Song J, et al. Acute erythroleukemia with der(1;7)(q10;p10) as a sole acquired abnormality after treatment with azathioprine. Cancer Genet Cytogenet. 2008;186(1):58-60.

(26.) Virchis A, Massey E, Butler T, et al. Acute myeloblastic leukaemias of FAB types M6 and M4, with cryptic PML/RARalpha fusion gene formation, relapsing as acute promyelocytic leukaemia M3. Br J Haematol. 2001;114(3): 551-556.

(27.) Schlenk RF, Dohner K, Krauter J, et al. Mutations and treatment outcome in cytogenetically normal acutemyeloid leukemia. N Engl J Med. 2008;358(18): 1909-1918.

(28.) Ruan GR, Li JL, Qin YZ, et al. Nucleophosmin mutations in Chinese adults with acute myelogenous leukemia. Ann Hematol. 2009;88(2):159-166.

(29.) Thiede C, Steudel C, Mohr B, et al. Analysis of FLT3-activating mutations in 979 patients with acute myelogenous leukemia: association with FAB subtypes and identification of subgroups with poor prognosis. Blood. 2002;99(12):4326-4335.

(30.) Kuchenbauer F, Kern W, Schoch C, et al. Detailed analysis of FLT3 expression levels in acute myeloid leukemia. Haematologica. 2005;90(12):1617-1625.

(31.) Steensma DP, McClure RF, Karp JE, et al. JAK2 V617F is a rare finding in de novo acute myeloid leukemia, but STAT3 activation is common and remains unexplained. Leukemia. 2006;20(6):971-978.

(32.) Auewarakul CU, Leecharendkeat A, Thongnoppakhun W, Limwongse C, Tocharoentanaphol C. Mutations of AML1 in non-M0 acute myeloid leukemia: six novel mutations and a high incidence of cooperative events in a South-east Asian population. Haematologica. 2006;91(5):675-678.

(33.) Trecca D, Longo L, Biondi A, et al. Analysis of p53 gene mutations in acute myeloid leukemia. Am J Hematol. 1994;46(4):304-309.

(34.) Selby DM, Valdez R, Schnitzer B, Ross CW, Finn WG. Diagnostic criteria for acute erythroleukemia. Blood. 2003;101(7):2895-2896.

(35.) Wang SA, Tang G, Fadare O, et al. Erythroid-predominant myelodysplastic syndromes: enumeration of blasts from nonerythroid rather than total marrow cells provides superior risk stratification. Mod Pathol. 2008;21(11):1394-1402.

(36.) Zuo Z, Kasyan A, Chandra P, Medeiros LJ, Koeppen H. Acute erythroid leukemias--histologic, cytogenetic and molecular characterization. Lab Invest. 2009;89(S1):293A.

(37.) Camacho RM, Martin-Noya A. Pure erythroid leukaemia with leukaemic cells mimicking myeloma cells. Br J Haematol. 2008;141(1):2.

(38.) Tso AC, Kumaran TO, Bain BJ. Case 41: a misdiagnosis of erythroleukemia. Leuk Lymphoma. 2009;50(6):1030-1032.

(39.) Camera A, Volpicelli M, Villa MR, Risitano AM, Rossi M, Rotoli B. Complete remission induced by high dose erythropoietin and granulocyte colony stimulating factor in acute erythroleukemia (AML-M6 with maturation). Haematologica. 2002;87(11):1225-1227.

(40.) Mazzella FM, Kowal-Vern A, Shrit MA, Rector JT, Cotelingam JD, Schumacher HR. Effects of multidrug resistance gene expression in acute erythroleukemia. Mod Pathol. 2000;13(4):407-413.

(41.) Sonneveld P. Multidrug resistance in haematological malignancies. J Intern Med. 2000;247(5):521-534.

(42.) Santos FPS, Faderl S, Garcia-Manero G, et al. Adult acute erythroleukemia: an analysis of 91 patients treated at a single institution. Leukemia. 2009;23: 2275-2280.

(43.) Mazzella FM, Smith D, Horn P, et al. Prognostic significance of pronormoblasts in erythrocyte predominant myelodysplastic patients. Am J Hematol. 2006;81(7):484-491.

(44.) Liu J, Lu DP, Zhang Y. The prognosis and cytogenetic characteristics of acute erythroid leukemia: a report of 55 cases [in Chinese]. Zhonghua Nei Ke Za Zhi. 2006;45(9):738-740.

Accepted for publication January 15, 2010.

From the Department of Hematopathology, University of Texas M. D. Anderson Cancer Center, Houston (Drs Zuo, Kasyan, and Medeiros); and the Department of Pathology, University of South Alabama, Mobile (Dr Polski).

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: Zhuang Zuo, MD, PhD, Department of Hematopathology, Unit 0149, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030 (e-mail:
Table 1. Comparison of French-American-British (FAB) and World
Health Organization (WHO) Classifications for Acute
Erythroid Leukemia

 FAB Classification
 1976 1985
 Criteria Criteria

Name AML-M6 AML-M6

BM findings
Erythroid precursors [greater than or [greater than or
 equal to] $30% equal to] $50%
Blast of nonerythroid [greater than or
 cells equal to] $30%
Dyserythropoiesis [greater than or Prominent
 equal to] $10%
Other findings

 WHO Classification
 2001 Criteria
 Acute Erythroid Leukemia
 Erythroleukemia Erythroid
Name (Erythroid/Myeloid) Leukemia

BM findings
Erythroid precursors [greater than or [greater than or
 equal to] $50% equal to] $80%
Blast of nonerythroid [greater than or
 cells equal to] $20%
Dyserythropoiesis Prominent Prominent

Other findings No multilineage involvement

 WHO Classification
 2008 Criteria
 Acute Erythroid Leukemia
Name Erythroleukemia Leukemia

BM findings
Erythroid precursors [greater than or [greater than or
 equal to] $50% equal to] $80%
Blast of nonerythroid [greater than or
 cells equal to] $20%
Dyserythropoiesis Prominent Prominent

Other findings No multilineage involvement; no prior
 erythropoietin therapy

Abbreviations: AML, acute myeloid leukemia; BM, bone marrow.

Table 2. Diseases or Conditions to Be Excluded
Before Establishing the Diagnosis of Acute
Erythroid Leukemia

Bone marrow insult (resulting in granulocytopenia)
 Viral infection
 Exposure to toxic chemicals

Therapy for cytopenias

Blood loss

Medical conditions (including causes of hemolysis)
 Endocrine diseases
 Valvular heart disease
 Intrinsic erythrocyte diseases

Nutritional deficiency
 Pernicious anemia
 Folate deficiency
 Iron deficiency

Abbreviations: G-CSF, granulocyte colony-stimulating factor;GM-CSF,
granulocyte-macrophage colony-stimulating factor.
COPYRIGHT 2010 College of American Pathologists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Zuo, Zhuang; Polski, Jacek M.; Kasyan, Armen; Medeiros, L. Jeffrey
Publication:Archives of Pathology & Laboratory Medicine
Date:Sep 1, 2010
Previous Article:Adenocarcinoma in ectopic prostatic tissue at dome of bladder: a case report of a patient with urothelial carcinoma of the bladder and adenocarcinoma...
Next Article:A new study of intraosseous blood for laboratory analysis.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters