Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Mixed-phenotype acute leukemia: historical overview and a new definition


Acute leukemia with a mixed phenotype is a rare disease and comprises 2–5% of all acute leukemias. These disorders have been known historically by a variety of names, such as mixed lineage leukemia, bilineal leukemia and biphenotypic leukemia, and the criteria for diagnosis have often been arbitrary. The scoring criteria proposed by the European Group for the Immunological Characterization of Leukemias represented a major attempt to define this disorder. However, the relative weight given to some markers and the lack of lineage specificity of most markers have raised questions regarding the significance of this approach. In 2008, the World Health Organization classification of hematopoietic and lymphoid tumors proposed a simpler diagnostic algorithm, which relies on fewer and more lineage-specific markers to define mixed-phenotype acute leukemia (MPAL). MPAL with t(9;22) and MLL rearrangement have been separated. Several studies have suggested that patients with acute leukemia of mixed phenotype have a worse clinical outcome when compared with matched controls with acute myeloid leukemia or acute lymphoblastic leukemia. Further studies are needed to confirm the significance of MPAL as currently defined, to determine a standardized treatment approach and to better understand the biological and clinical aspects of this disease.


The diagnosis and classification of acute leukemia relies on a multidisciplinary approach including morphology, immunophenotyping, karyotype analysis and more specific molecular genetic analysis. Using this approach, most acute leukemias can be unequivocally assigned to myeloid, B- or T-lymphoid lineage. However, even after extensive immunophenotyping, a small and heterogeneous subset of leukemias cannot be readily classified. Leukemias falling into this category have been given many different names, including acute mixed lineage leukemias, biphenotypic leukemias, hybrid leukemias, undifferentiated leukemias and, most recently, leukemia of ambiguous lineage.1 Among the latter group are mixed-phenotype acute leukemia (MPAL) and acute undifferentiated leukemia. Some acute leukemias with a mixed phenotype contain two morphologically and immunophenotypically distinct populations of blasts and these have been referred to as acute bilineal (or bilineage) leukemias.2 However, these cases are often included with cases showing a single blast cell population expressing mixed phenotypic markers (so-called biphenotypic acute leukemias) and the relationship between the bilineal and biphenotypic leukemia is not clear.2 Some acute leukemias with recurring cytogenetic abnormalities characteristically demonstrate a mixed phenotype, but the cytogenetic findings apparently define a more distinct disease entity when compared with the immunophenotype. These include acute myeloid leukemia (AML) with t(8;21)(q22;q22), AML with inv(16)(p13.1q22) or t(16;16)(p13.1;q22), as well as acute lymphoblastic leukemia (ALL) with t(9;22)(q34;q11.2) and ALL with 11q23 translocations.3, 4, 5, 6, 7, 8 As we learn more about the significance of these biological acute leukemia types, we may place even less relevance on immunophenotypes in comparison with genetic changes.

Historical overview of definitions used for biphenotypic leukemia

The first published reports on biphenotypic acute leukemia occurred in the 1980s, when monoclonal antibodies were first being used to characterize leukemic cells.9 One of these early publications noted co-expression of myeloperoxidase and terminal deoxynucleotidyl transferase (TdT) in AML.10 This and other reports suggested the possible existence of stem cell leukemia capable of differentiating into myeloid and lymphoid cells.11 One of the earliest large series of acute mixed lineage leukemia was done by Mirro et al.,12 who looked at the frequency and significance of acute leukemia displaying both lymphoid and myeloid characteristics in 123 children. In this study, the definition of acute mixed lineage leukemia included having individual blasts with more than one lineage.12 This was determined using lymphoid-associated antibodies such as anti-CALLA (CD10), T-11 (CD2) or T101 (CD5), and myeloid-associated antibodies were composed of MY-1 (CD15), MCS.2 (CD13) and Mo1 (CD11b); however, none of these markers are now considered to be lineage specific. On the basis of these markers, acute mixed lineage leukemia comprised 20% of the total number of 123 cases.12 From the available clinical data, the authors noted that most patients with ALL and myeloid markers entered complete remission, whereas in those with AML and lymphoid markers the clinical response was more heterogeneous.12

As more antibodies became available, it became clear that a significant percentage of AML and ALL cases demonstrated aberrant immunophenotypes and specific criteria were needed to diagnose a true mixed phenotype leukemia.6, 7 A defined scoring system for biphenotypic acute leukemia was proposed by Catovsky et al.13 in 1991, further suggesting that these cases represent a distinct type of acute leukemia with a unique biology and prognosis. Catovsky's scoring criteria were based on the number and different weights given to diagnostic markers according to their generally accepted specificity at that time (Table 1). The most specific and heavily weighted markers included cytoplasmic CD3, cytoplasmic CD22 and myeloperoxidase (MPO).13 Two or more points from two separate lineages were needed to classify a case as biphenotypic acute leukemia.

Table 1 Catovsky et al scoring system for biphenotypic acute leukemia

The European Group for the Immunological Characterization of Leukemias (EGIL) later proposed an immunological classification and characterization of acute leukemias, which also included a definition for biphenotypic acute leukemia.14 This definition was largely based on the proposal made by Catovsky14 (Table 2). Some of the differences included the addition of newer markers thought to have a high degree of lineage specificity, such as CD79a.14 Biphenotypic acute leukemia was defined by the EGIL when a score over 2 points was achieved for the myeloid as well as one of the lymphoid lineages (Table 2). At least 20% of cells staining with a monoclonal antibody was chosen as a cut-off point to consider a marker as positive.14 An exception was made for MPO, CD3 and CD79a markers, as well as for TdT, owing to their high degree of specificity. For these specific markers, a minimum cut-off of 10% of cells staining with an antibody was considered adequate. EGIL also listed acceptable techniques to look for marker expression and these included flow cytometry, indirect immunofluorescence or immunocytochemical techniques.14 In 1998, the EGIL group15 slightly revised their scoring system after concluding that CD117 had a high specificity for myeloid lineage. In addition to providing a definition for biphenotypic acute leukemia, the EGIL group also proposed a recommended panel of markers needed to characterize acute leukemias (Table 3).

Table 2 Revised EGIL scoring system for biphenotypic acute leukemia
Table 3 EGIL-recommended panel of markers to characterize acute leukemias

The 2001 World Health Organization (WHO) classification of hematopoietic and lymphoid neoplasms included a category of acute leukemias of ambiguous lineage, wherein the authors alluded to the scoring system proposed by the EGIL.16 This category included undifferentiated acute leukemias as well as both bilineal and biphenotypic acute leukemias. Although bilineal acute leukemia was recognized as having a dual population of blasts of distinct lineage, it was also acknowledged that these cases may evolve into biphenotypic acute leukemia and the relationship between the two was not clear.16 In addition, cases with immunophenotypic evidence of lineage switch were mentioned in this category, which likely represented expansion of a pre-existing minor population of blasts with a different immunophenotype. Emphasis was placed on distinction from myeloid antigen-positive ALL or lymphoid antigen-positive AML by suggesting that multiple antigens associated with more than one lineage are required and co-expression of one or two cross-lineage antigens is not sufficient to diagnose biphenotypic leukemia. In the 2001 WHO classification listing of the scoring system proposed by EGIL, a typographical error was made and the second printed version of this work stated that a score of 2 was needed for the myeloid and one of the lymphoid lineages instead of >210 as the EGIL had proposed.16 This error led to the misconception of loosening of the criteria for biphenotypic leukemia and added to the confusion that already surrounded the diagnostic criteria for these disorders. For example, in a series of biphenotypic cases published by Aribi et al.17 using 2 criteria mentioned in the 2001 WHO classification (rather than the >2 criteria of the EGIL), the number of biphenotypic leukemias tripled from 10 using EGIL criteria to 31 using this modified criteria.

Even when the correct EGIL criteria are applied, strict application of the point system can lead to an inaccurate classification. For instance, classical AML cases with t(8;21)(q22;q22) or t(15;17)(q22;q12) could be classified as a biphenotypic leukemia,18 because B-cell antigen expression (most commonly CD19 and PAX5, but less commonly TdT, CD79a and CD20) is common in AML with t(8;21) and T-cell antigen expression (especially CD2) is frequent in acute promyelocytic leukemia.19, 20 In addition, the relative weight given to some of the markers in the EGIL criteria has been challenged. For example, cytoplasmic CD79a, considered to be a highly specific marker for B-lymphoid lineage in the EGIL scoring system, is positive in a significant percentage of T-ALL21, 22 and has been reported in some myeloid leukemias.23, 24 Similarly, MPO positivity has been found in up to 23% of adult B-ALL cases when a polyclonal antibody is used on paraffin sections.25 Furthermore, the EGIL scoring system did not reflect the intensity of antigen expression, which has been reported to be important in determining lineage specificity.26 For these reasons, the 2008 WHO has replaced the EGIL scoring system with a simpler diagnostic algorithm that relies on fewer markers to define mixed phenotype acute leukemia.27

MPAL as defined by the 2008 WHO classification

The 2008 WHO classification system grouped bilineal and biphenotypic acute leukemias together under a new heading, MPAL. The diagnosis of MPAL is assigned to a case of acute leukemia that shows expression of a combination of antigens of different lineages, so that it is not possible to assign a single lineage. A specific reference is now made to exclude cases that can be classified in another category, either by genetic or by clinical features.27 For instance, AML with t(8;21), t(15;17) and inv(16) can express lymphoid-associated markers but should be classified as AML with recurrent genetic abnormalities. This applies to all cases of AML with recurrent cytogenetic abnormalities. Provisional entities such as AML with mutated NPM1 and CEBPA do not trump the diagnosis of biphenotypic leukemia. Cases of chronic myelogenous leukemia (CML) in blast crisis, AML with myelodysplasia-related changes and therapy-related AML should be classified as their respective entities even if they happen to have a mixed phenotype.27 Presence of a complex karyotype or any other MDS-related cytogenetic changes without a history of MDS should be considered as AML with myelodysplasia-related changes. Similarly, presence of multilineage dysplasia classifies a case as AML with myelodysplasia-related changes using the 2008 WHO classification.27 Although cases of CML in blast crisis can show lineage infidelity,28 these cases often have a background of CML, including leukocytosis and basophilia or a clinical history of CML.

The lineage-specific markers have been changed in this edition of the WHO and are applicable mostly to cases wherein a single population of blasts is present (Table 4). In cases with two separate populations of blasts with distinct and lineage-specific pattern of differentiation, one population of blasts should meet the immunophenotypic criteria for AML. Although a defining criterion for AML is the presence of 20% myeloblasts in peripheral blood or bone marrow, in these cases <20% myeloblasts is acceptable when the total number of leukemic blasts, including non-myeloblasts, is 20%. If there is only a single population of blasts present that otherwise meets the criteria for B-ALL or T-ALL, myeloid lineage is defined by the presence of MPO positivity using flow cytometry, immunohistochemistry or cytochemistry.27 Although no specific threshold for MPO positivity was set, it has been noted that in cases of limited MPO positivity, it is necessary to demonstrate that MPO is present on leukemic cells and not on residual normal blasts.1 If MPO is studied by flow cytometry, its expression should be demonstrated on an aberrant population. If cytochemistry is used to study MPO, more than 3% of blasts should be positive.29 In the absence of MPO, either diffuse positivity for non-specific esterase or the presence of more than one monocytic marker such as CD11c, CD14, CD64 and lysozyme is needed to support myeloid differentiation. The authors suggest that CD13, CD33 and CD117 are not specific enough to allow identification of MPAL.27

Table 4 2008 WHO classification: acute leukemias of ambiguous lineage

T lineage is defined by the presence of cytoplasmic CD3 on either the entire blast population or a distinct subpopulation of blasts.27 Cytoplasmic CD3 expression should be as bright or nearly as bright as on normal residual T cells present in the sample. Surface CD3 is also sufficient, but is acknowledged to be rare in these cases. Cytoplasmic CD3 should be identified by flow cytometry with an antibody to a CD3 epsilon chain. Although immunohistochemical stains performed on paraffin-embedded bone marrow biopsies using CD3 are acceptable, it is emphasized that this method can also react with the zeta chain of T-cell receptor, which is also present in the cytoplasm of natural killer cells and is therefore not T lineage specific.27

Unlike T lineage, no single specific marker for B-cell differentiation is identified in the 2008 WHO definition.27 In this new definition of MPAL, B-cell differentiation requires strong CD19 expression together with strong expression of one of the following antigens: CD79a, cytoplasmic CD22 or CD10. Alternatively, if weak CD19 expression is present, two of the following antigens must be strongly expressed: CD79a, cytoplasmic CD22 or CD10. A case may even be assigned to B-lineage without expression of CD19; however, it is noted that both CD79a and CD10 lack specificity.27 These definitions for myeloid, T and B lineage apply mostly to cases with a single population of blasts. In those cases wherein two or more distinct populations of blasts are seen, one of the blast populations needs to meet the criteria for AML. Caution is advised not to overinterpret AML with increased hematogones or ALL in a marrow with increased myeloid precursors. Hematogones have phenotypic stability with three stages of maturation, which leads to characteristic patterns on flow-cytometry dot plots.30, 31 Hematogones have low side scatter and reproducible maturation pattern, whereas blasts in B-ALL tend to have higher side scatter and phenotypic abnormalities.31 In addition, hematogones tend to be dispersed throughout the marrow, whereas blasts form small clusters.32 These differences help to distinguish hematogones that can be present in AML from bilineal MPAL.

Two genetic lesions have been reported frequently enough in MPAL to now be considered as separate entities.27 The first is MPAL with t(9;22)(q34;q11.2) or BCR-ABL1 rearrangement. The clinical features associated with this translocation are similar to those of other patients with MPAL, but it is important not to make this diagnosis in patients with CML. Such a history would place a case in blast transformation of CML rather than MPAL. Philadelphia chromosome-positive leukemias are generally more frequent in older patients. Although most studies found the frequency of MPAL with t(9;22) to be 28–35%, pediatric studies report it to be much lower at 3%.33 Many of these cases show a dimorphic population of blasts, with most showing B and myeloid lineage. This translocation in leukemias with a mixed phenotype was associated with a worse prognosis in some studies.34 The second most frequent genetic lesion in MPAL are translocations involving MLL gene, with the most common partner gene being AF4 on chromosome 4 band q21.35 This tends to occur more commonly in children and is more frequent in infancy.36 These cases also tend to present with a dimorphic blast population and the lymphoblasts have a CD19-positive, CD10-negative, B-precursor immunophenotype and are frequently positive for CD15. The prognosis of MPAL patients with MLL rearrangement is poor.35

If the blasts lack the above-mentioned genetic abnormalities, MPAL cases are categorized by the lineage of the blasts. Thus, there are three other less-specific categories, which include B/myeloid, not otherwise specified, T/myeloid, not otherwise specified, and other rare types of MPAL listed under MPAL, not otherwise specified. It is acknowledged that cases with a combination of B and T lineage and trilineage cases are very rare. For instance, in cases of T-cell leukemia, CD79a and CD10 alone should not be considered as evidence of B-cell differentiation as these markers are relatively common in T-ALL.27

Clinical features

Presenting clinical symptoms in acute leukemia with a mixed phenotype are usually due to bone marrow failure, similar to other acute leukemias, and include fatigue, infections and bleeding disorders.1 The white blood cell count is high and most cases have a high number of circulating blasts.1 Early on, using variable diagnostic criteria, the reported prevalence of acute leukemia with a mixed phenotype ranged from 2 to 20% of all acute leukemias; however, more recent studies using EGIL criteria show a frequency of 2–5%.35 Using a panel proposed by EGIL, a recent study that evaluated the immunophenotype of 325 adult leukemias found that 1.8% were biphenotypic.37 The frequency of biphenotypic acute leukemia seems to be similar in all age groups. Owaidah et al.35 found that out of 676 cases of both adult and pediatric acute leukemia patients, 23 cases (3.4%) were diagnosed as biphenotypic by EGIL criteria. In all, 11 patients were aged 14 years or less and the other 12 cases were classified as adults.35 Killick et al.34 studied 20 de novo cases of biphenotypic leukemia using EGIL criteria; 8 patients were under the age of 15, whereas 12 were over 15 years of age.

The frequency of MPAL using the 2008 WHO criteria is not yet clear. Al-Seraihy et al.33 retrospectively evaluated 633 children under 14 years of age with acute leukemia and found that 24 (3.8%) of these cases were classified as biphenotypic acute leukemia using EGIL criteria, while 11 (1.7%) were categorized as MPAL using the new 2008 WHO criteria. A retrospective meta-analysis of major recent studies of biphenotypic acute leukemia using the 2008 WHO criteria is summarized in Table 4. Although not all markers outlined in the 2008 WHO criteria are reported in all of these studies and expression of markers is not well defined, the number of patients encompassed in this new MPAL category is significantly lower (Table 5).

Table 5 Analysis of the frequency of EGIL biphenotypic AL and MPAL using WHO criteria in prior studies

Immunophenotypic findings

Most cases of MPAL express CD45 and the early hematopoietic markers CD34, CD38, TdT and HLA-DR.1 An example of MPAL with strong expression of MPO and CD79a, CD10 and CD19 is shown in Figures 1a–d and an image of a bilineal case is shown in Figure 2. Oiwadah et al.35 found that most of their biphenotypic cases (65%) co-expressed myeloid and B-lymphoid antigens, while 26% were myeloid and T lineage. In a recent study of pediatric cases of acute mixed lineage leukemia defined by EGIL criteria, Rubnitz et al.38 found that the majority had blasts with T and myeloid markers (57%), unlike in adults in whom B and myeloid markers seem to predominate. In a series of 19 acute bilineal leukemias, Weir et al.2 found that all cases had a myeloid component, while 10 also had a T-lymphoid component and 9 had a B-lymphoid component. Lee et al.39 looked at the common markers present in 43 biphenotypic acute leukemias and found that myeloid markers MPO, CD13 and CD33 were most frequently present in cases that contained both myeloid and B-lymphoid antigens as well as myeloid and T-lymphoid antigens. The most common lymphoid markers included CD10 in myeloid and B-lymphoid cases and TdT in all biphenotypic cases.39

Figure 1

MPAL with a single population of blasts that show expression of MPO and CD79a (a), CD19 and CD10 (b), TdT (c) and CD34 (d).

Figure 2

MPAL with two separate populations of blasts (also known as acute bilineal leukemia).

Genetic findings

There is no single chromosomal aberrancy that is uniquely associated with MPAL. In their study of 23 biphenotypic (EGIL) cases, Owaidah et al.35 found that 68% had a clonal abnormality whereas 32% had a normal karyotype. The most common abnormalities included rearrangement of 11q23, the site of MLL, followed by the Philadelphia chromosome t(9;22)(q34;q11.2). Additional abnormalities included deletion of 6q, 5q and 12p.35 Similar genetic findings were noted by Legrand et al.40 in their study of 23 biphenotypic acute leukemias. These authors also looked at MDR1 gene expression and found that it was higher in patients with biphenotypic acute leukemia than AML using RT-PCR and flow cytometry. Lee et al.39 studied 43 patients with biphenotypic acute leukemia and found that the Philadelphia chromosome t(9;22) was the most common abnormality, present in 14 patients (32%). All of the patients with the Philadelphia chromosome were characterized by myeloid and B-lymphoid immunophenotype.39 However, the incidence of the Philadelphia chromosome was noted to be significantly lower in pediatric biphenotypic cases, estimated at 3% in one study.33 Rubnitz et al.38 found that most of their pediatric cases had abnormal karyotypes (29/33 in total), with the most common being abnormalities of chromosomes 5 or 7 followed by 12p abnormalities. In addition, these authors also used a gene-expression database to compare biphenotypic acute leukemia with AML and ALL. They observed that 8 of their 13 mixed lineage cases formed a distinct cluster with only few similarities to other leukemias, whereas five cases clustered with known AML cases.38 Wouters et al.41 analyzed AML gene-expression profiling data and found that the presence of CEBPA mutations correlated with two separate clusters. A supervised analysis revealed a set of genes that were highly expressed in these CEBPA-silenced leukemias and included those related to T-lymphoid development, such as CD7, CD3D, LCK and T-cell receptor delta locus (TRD). This could be related to a broad lineage-determining impact of CEBPA.

Buccheri et al.42 showed that a substantial number of biphenotypic acute leukemia cases also have rearrangements or deletions of the immunoglobulin heavy chain or T-cell receptor genes. Of the 12 biphenotypic acute leukemias that were included in their study, 7 cases showed immunoglobulin gene rearrangements or deletions and 5 contained T-cell receptor gene rearrangements or deletions. Interestingly, Buccheri et al.42 found that biphenotypic acute leukemia with higher scores using Catovsky criteria tended to have more frequent gene rearrangements. Of note, T-cell receptor and IGH@ rearrangements can occur in non-biphenotypic AML and these ‘promiscuous’ gene rearrangements are not limited to MPAL.43

Clinical outcome

There are no set treatment protocols for patients with MPAL, which makes evaluation of outcome difficult in studies of this entity. A few studies have compared the outcome of patients with MPAL with that of matched controls with AML or ALL. Killick et al.34 showed that in their adult group, the survival of patients with biphenotypic leukemia was worse when compared with matched controls with AML or ALL. Similarly, Xue et al.36 found that adult patients with biphenotypic leukemia had a worse overall survival and disease-free survival than those with either AML or ALL. Legrand et al.40 also found that adult patients with biphenotypic acute leukemia had a worse prognosis in terms of achieving complete remission and 4-year overall survival when compared with AML or ALL patients. Pediatric cases of biphenotypic acute leukemia have been found to have better prognosis when compared with adult biphenotypic cases.34 However, although some found that there was no difference in survival between pediatric MPAL patients and those with AML or ALL,34 others found that the outcome of children with biphenotypic leukemia was inferior to that of children with ALL.38 This same study showed that there was no difference in outcome between pediatric biphenotypic patients as compared with those with AML.38

A few studies have attempted to correlate immunophenotype with outcome in MPAL patients. In their study of pediatric biphenotypic acute leukemia cases, Rubnitz et al.38 found no difference in survival between those with B/myeloid and those with T/myeloid leukemia. However, when evaluating survival by immunophenotype in an adult population, Lee et al.39 found that cases with myeloid and T-lymphoid phenotype had a worse overall survival as compared with the rest of the biphenotypic cases in a multivariate analysis. Expression of CD19 and CD14 was associated with unfavorable outcome in this study. Zheng et al.44 showed that MPO was not a prognostic indicator in this disease.

Correlation of genetic abnormalities with clinical outcome in 43 patients with biphenotypic acute leukemia by Lee et al.39 showed that presence of the Philadelphia chromosome did not correlate with survival. However, Killick et al.34 found that 86% of their patients with t(9;22) were dead at follow-up as compared with 40% without this translocation, which suggests that the presence of this translocation correlates with a worse outcome (P=0.03). Similarly, Xue et al.36 found that Philadelphia-positive biphenotypic leukemia had a worse prognosis when compared with other cases. In their study, Killick et al.34 found that there was no apparent difference in survival between Philadelphia-positive biphenotypic leukemia and Philadelphia-positive ALL. In addition, the outcome of patients with biphenotypic leukemia and absence of t(9;22) was similar to that of ALL patients without t(9;22).34

Evaluation of clinical outcome by classification schemes showed that pediatric patients with MPAL diagnosed using the 2008 WHO criteria had a better outcome as compared with the rest of biphenotypic cases classified using EGIL criteria.33 Specifically, these 11 MPAL patients were alive and did not have relapse of their disease and only a subset (six patients) were transplanted after achieving remission.33 Those patients who were initially diagnosed with biphenotypic leukemia using EGIL criteria but did not conform to the 2008 WHO criteria for MPAL did poorly, with an overall survival of 54%.33 Although fewer patients will be classified as MPAL using the 2008 WHO criteria (Table 4), the clinical outcome of these patients has not been evaluated in larger studies.


As there is no set therapy for mixed-phenotype leukemia patients, different therapeutic approaches have been reported. There is no agreement as to whether induction therapy should be with lymphoid and/or myeloid drugs and whether this should be followed by bone marrow or peripheral blood stem cell transplantation.34 For instance, Al-Seraihy et al.,33 in their study of childhood MPAL, mentioned that the leukemia team at their institution elected to treat these patients using a strategy based on the St Jude Total XIII-B high-risk protocol.45 Although this protocol was originally developed for the treatment of patients with high-risk ALL, it incorporates several agents that are effective in the treatment of myeloid leukemias. Specifically, this protocol includes induction chemotherapy composed of six agents effective in both lymphoid and myeloid leukemias, followed by consolidation therapy with two doses of high-dose methotrexate. After consolidation, patients received weekly continuation therapy consisting of non-cross-resistant drug pairs for 120 weeks.33 Killick et al.34 reported that induction treatment with combined AML/ALL drugs led to a high rate of early death, although most of these patients were adults with t(9;22). Thus, their recommendation is that induction should include either AML or ALL drugs and specific regimens should be determined in larger trials. In a study of pediatric patients with MPAL, Rubnitz et al.38 found that complete remission rates were higher in pediatric patients undergoing induction with ALL-type therapy than AML-type induction therapy (83% versus 52%). Furthermore, 8 of 10 patients who failed to achieve complete remission with AML-type therapy, subsequently achieved it with ALL-type induction therapy.38 Al-Seraihy et al. found similar results in their pediatric cases, with better response to ALL-type of induction therapy.33

Summary and continued controversies

Overall, acute leukemias with mixed phenotypes are uncommon and comprise 2–5% of all acute leukemias. Most studies in the last decade have used the EGIL scoring system to diagnose biphenotypic acute leukemia. The 2008 WHO classification system has proposed a more simple algorithm that relies on fewer markers, and cases diagnosed with this system are referred to as MPAL. Although the diagnosis of MPAL is simpler with the 2008 WHO system, it is not yet clear whether the new criteria actually refine this diagnosis in comparison with the EGIL criteria. The 2008 WHO definition clarifies how to handle some case types that might otherwise be misdiagnosed, such as AML with t(8;21), which is a biologically distinct entity that should not be placed in the MPAL category even if it fulfills the criteria for MPAL. Could a similar argument be made for ALL?

The 2008 WHO classification creates distinct categories for two of the most common cytogenetic abnormalities described in leukemias of mixed phenotype: t(9;22) and translocations involving 11q23. Although leukemias with mixed phenotypes with these genetic abnormalities have been found in some studies to correlate with a worse prognosis, these cytogenetic abnormalities are also common and are also associated with a worse prognosis in ALL. One could argue that these cases would be best placed in the 2008 WHO ALL categories of ALL with t(9;22) and ALL with 11q23 translocations as they are probably more biologically related to ALL cases with these genetic abnormalities. It is well known that these ALL types typically have aberrant myeloid antigen expression, so it is not surprising that some of them would meet the criteria for MPAL. Further studies that directly compare the outcome of these two MPAL types with their specific ALL counterparts are needed to further address whether the current diagnostic separation is warranted. Gene expression patterns reported in pediatric patients,38 mostly without t(9;22) or 11q23 translocations, suggest that these other types of acute mixed lineage leukemias are a unique biological entity. Further genetic studies are needed on this topic to determine whether adult and pediatric acute leukemias with mixed phenotypes have a different behavior and whether these in fact represent different biological diseases. Finally, the prevalence and clinical significance of MPAL using the new 2008 WHO definition remains to be determined. Clinical treatment for MPAL is not well established and further understanding of the genetic basis of this disease is needed to determine whether this represents a distinct biological and clinical entity. As with most acute leukemias, wherein the cytogenetic and molecular genetic changes have emerged to be of greatest biological importance, we may find that this is also the case in MPAL. If that is true, the controversy on how to diagnose and treat these disorders will shift away from the immunophenotype and towards more specific molecular genetic markers.


  1. 1

    Weir EG, Borowitz MJ . Acute leukemias of ambiguous lineage. In: Jaffe ES, Harris NL, Vardiman JW, Campo E, Arber DA (eds). Hematopathology. Elsevier: Philadelphia, 2010.

    Google Scholar 

  2. 2

    Weir EG, Ali Ansari-Lari M, Batista DA, Griffin CA, Fuller S, Smith BD et al. Acute bilineal leukemia: a rare disease with poor outcome. Leukemia 2007; 21: 2264–2270.

    CAS  Article  Google Scholar 

  3. 3

    Kita K, Nakase K, Miwa H, Masuya M, Nishii K, Morita N et al. Phenotypical characteristics of acute myelocytic leukemia associated with the t(8;21)(q22;q22) chromosomal abnormality: frequent expression of immature B-cell antigen CD19 together with stem cell antigen CD34. Blood 1992; 80: 470–477.

    CAS  PubMed  PubMed Central  Google Scholar 

  4. 4

    Hurwitz CA, Raimondi SC, Head D, Krance R, Mirro Jr J, Kalwinsky DK et al. Distinctive immunophenotypic features of t(8;21)(q22;q22) acute myeloblastic leukemia in children. Blood 1992; 80: 3182–3188.

    CAS  PubMed  PubMed Central  Google Scholar 

  5. 5

    Adriaansen HJ, te Boekhorst PA, Hagemeijer AM, van der Schoot CE, Delwel HR, van Dongen JJ . Acute myeloid leukemia M4 with bone marrow eosinophilia (M4Eo) and inv(16)(p13q22) exhibits a specific immunophenotype with CD2 expression. Blood 1993; 81: 3043–3051.

    CAS  PubMed  PubMed Central  Google Scholar 

  6. 6

    Khalidi HS, Medeiros LJ, Chang KL, Brynes RK, Slovak ML, Arber DA . The immunophenotype of adult acute myeloid leukemia: high frequency of lymphoid antigen expression and comparison of immunophenotype, French-American-British classification, and karyotypic abnormalities. Am J Clin Pathol 1998; 109: 211–220.

    CAS  Article  Google Scholar 

  7. 7

    Khalidi HS, Chang KL, Medeiros LJ, Brynes RK, Slovak ML, Murata-Collins JL et al. Acute lymphoblastic leukemia. Survey of immunophenotype, French-American-British classification, frequency of myeloid antigen expression, and karyotypic abnormalities in 210 pediatric and adult cases. Am J Clin Pathol 1999; 111: 467–476.

    CAS  Article  Google Scholar 

  8. 8

    Childs CC, Hirsch-Ginsberg C, Walters RS, Andersson BS, Reuben J, Trujillo JM et al. Myeloid surface antigen-positive acute lymphoblastic leukemia (My+ ALL): immunophenotypic, ultrastructural, cytogenetic, and molecular characteristics. Leukemia 1989; 3: 777–783.

    CAS  PubMed  Google Scholar 

  9. 9

    Bene MC . Biphenotypic, bilineal, ambiguous or mixed lineage: strange leukemias!. Hematologica 2009; 94: 891–893.

    Article  Google Scholar 

  10. 10

    McGraw TP, Folds JD, Bollum FJ, Strass SA . Terminal deoxynucleotidyl transferase-positive acute myeloblastic leukemia. Am J Hematol 1981; 10: 251–258.

    CAS  Article  Google Scholar 

  11. 11

    Scamurra DO, Davey FR, Nelson DA, Kurec AS, Goldberg J . Acute leukemia presenting with myeloid and lymphoid cell markers. Ann Clin Lab Sci 198; 13: 396–502.

    Google Scholar 

  12. 12

    Mirro J, Zipf TF, Pui CH, Kitchingman G, Williams D, Melvin S et al. Acute mixed lineage leukemia: clinicopathologic correlation and prognostic significance. Blood 1985; 66: 1115–1123.

    CAS  PubMed  Google Scholar 

  13. 13

    Catovsky D, Matutes E, Buccheri V, Shetty V, Hanslip J, Yoshida N et al. A classification of acute leukemia for the 1990s. Ann Hematol 1991; 62: 16–21.

    CAS  Article  Google Scholar 

  14. 14

    Bene MC, Castoldi G, Knapp W, Ludwig WD, Matutes E, Orfao A et al. Proposals for the immunological classification of acute leukemias. European Group for the Immunological Characterization of Leukemias (EGIL). Leukemia 1995; 9: 1783–1786.

    CAS  PubMed  Google Scholar 

  15. 15

    Bene MC, Bernier M, Casasnovas RO, Castoldi G, Knapp W, Lanza F et al. The reliability and specificity of c-kit for the diagnosis of acute myeloid leukemias and undifferentiated leukemias. Blood 1998; 92: 596–599.

    CAS  PubMed  PubMed Central  Google Scholar 

  16. 16

    Jaffe E, Harris N, Stein HVJ, Vardiman JW (eds). Pathology and Genetics of Tumors of Hematopoietic and Lymphoid Tissues, 2nd printing, World Health Organization Classification of Tumours. IARC Press: Lyon, France, 2001.

    Google Scholar 

  17. 17

    Aribi A, Bueso-Ramos C, Estey E, Estrov Z, O’Brien S, Giles F et al. Biphenotypic acute leukaemia: a case series. Br J Haematol 2007; 138: 213–216.

    Article  Google Scholar 

  18. 18

    Bagg A . Lineage ambiguity, infidelity, and promiscuity in immunophenotypically complex acute leukemias: genetic and morphologic correlates. Am J Clin Pathol 2007; 128: 545–548.

    Article  Google Scholar 

  19. 19

    Valbuena JR, Medeiros LJ, Rassidakis GZ, Hao S, Wu CD, Chen L et al. Expression of B cell-specific activator protein/PAX5 in acute myeloid leukemia with t(8;21)(q22;q22). Am J Clin Pathol 2006; 126: 235–240.

    Article  Google Scholar 

  20. 20

    Chapiro E, Delabesse E, Asnafi V, Millien C, Davi F, Nugent E et al. Expression of T-lineage-affiliated transcripts and TCR rearrangements in acute promyelocytic leukemia: implications for the cellular target of t(15;17). Blood 2006; 108: 3484–3493.

    CAS  Article  Google Scholar 

  21. 21

    Hashimoto M, Yamashita Y, Mori N . Immunohistochemical detection of CD79a expression in precursor T cell lymphoblastic lymphoma/leukemias. J Pathol 2002; 197: 341–347.

    CAS  Article  Google Scholar 

  22. 22

    Pilozzi E, Pulford K, Jones M, Müller-Hermelink HK, Falini B, Ralfkiaer E et al. Co-expression of CD79a (JCB117) and CD3 by lymphoblastic lymphoma. J Pathol 1998; 186: 140–143.

    CAS  Article  Google Scholar 

  23. 23

    Arber DA, Jenkins KA, Slovak ML . CD79 alpha expression in acute myeloid leukemia. High frequency of expression in acute promyelocytic leukemia. Am J Pathol 1996; 149: 1105–1110.

    CAS  PubMed  PubMed Central  Google Scholar 

  24. 24

    Bhargava P, Kallakury BV, Ross JS, Azumi N, Bagg A . CD79a is heterogeneously expressed in neoplastic and normal myeloid precursors and megakaryocytes in an antibody clone-dependent manner. Am J Clin Pathol 2007; 128: 306–313.

    Article  Google Scholar 

  25. 25

    Arber DA, Snyder DS, Fine M, Dagis A, Niland J, Slovak ML . Myeloperoxidase immunoreactivity in adult acute lymphoblastic leukemia. Am J Clin Pathol 2001; 116: 25–33.

    CAS  Article  Google Scholar 

  26. 26

    Paietta E, Racevskis J, Bennett JM, Wiernik PH . Differential expression of terminal transferase (TdT) in acute lymphocytic leukaemia expressing myeloid antigens and TdT positive acute myeloid leukaemia as compared to myeloid antigen negative acute lymphocytic leukaemia. Br J Haematol 1993; 84: 416–422.

    CAS  Article  Google Scholar 

  27. 27

    Borowitz MJ, Bene MC, Harris NL, Porwit A, Matutes E . Acute leukemias of ambiguous lineage. In: Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman JW (eds). WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC Press: Lyon, 2008, pp 150–155.

    Google Scholar 

  28. 28

    Khalidi HS, Brynes RK, Medeiros LJ, Chang KL, Slovak ML, Snyder DS et al. The immunophenotype of blast transformation of chronic myelogenous leukemia: a high frequency of mixed lineage phenotype in “lymphoid” blasts and a comparison of morphologic, immunophenotypic, and molecular findings. Mod Pathol 1998; 11: 1211–1221.

    CAS  PubMed  Google Scholar 

  29. 29

    Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR 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: 620–625.

    CAS  Article  Google Scholar 

  30. 30

    Sevilla DW, Colovai AI, Emmons FN, Bhagat G, Alobeid B . Hematogones: a review and update. Leuk Lymphoma 2010; 51: 10–19.

    Article  Google Scholar 

  31. 31

    Babusíková O, Zelezníková T, Kirschnerová G, Kankuri E . Hematogones in acute leukemia during and after therapy. Leuk Lymphoma 2008; 49: 1935–1944.

    Article  Google Scholar 

  32. 32

    Rimsza LM, Larson RS, Winter SS, Foucar K, Chong YY, Garner KW et al. Benign hematogone-rich lymphoid proliferations can be distinguished from B-lineage acute lymphoblastic leukemia by integration of morphology, immunophenotype, adhesion molecule expression, and architectural features. Am J Clin Pathol 2000; 114: 66–75.

    CAS  Article  Google Scholar 

  33. 33

    Al-Seraihy AS, Owaidah TM, Ayas M, El-Solh H, Al-Mahr M, Al-Ahmari A et al. Clinical characteristics and outcome of children with biphenotypic acute leukemia. Haematologica 2009; 94: 1682–1690.

    Article  Google Scholar 

  34. 34

    Killick S, Matutes E, Powles RL, Hamblin M, Swansbury J, Treleaven JG et al. Outcome of biphenotypic acute leukemia. Haematologica 1999; 84: 699–706.

    CAS  PubMed  Google Scholar 

  35. 35

    Owaidah TM, Al Beihany A, Iqbal MA, Elkum N, Roberts GT . Cytogenetics, molecular and ultrastructural characteristics of biphenotypic acute leukemia identified by the EGIL scoring system. Leukemia 2006; 20: 620–626.

    CAS  Article  Google Scholar 

  36. 36

    Xu XQ, Wang JM, Lü SQ, Chen L, Yang JM, Zhang WP et al. Clinical and biological characteristics of adult biphenotypic acute leukemia in comparison with that of acute myeloid leukemia and acute lymphoblastic leukemia: a case series of a Chinese population. Haematologica 2009; 94: 919–927.

    Article  Google Scholar 

  37. 37

    Thalhammer-Scherrer R, Mitterbauer G, Simonitsch I, Jaeger U, Lechner K, Schneider B et al. The immunophenotype of 325 adult acute leukemias: relationship to morphologic and molecular classification and proposal for a minimal screening program highly predictive for lineage discrimination. Am J Clin Pathol 2002; 117: 380–389.

    Article  Google Scholar 

  38. 38

    Rubnitz JE, Onciu M, Pounds S, Shurtleff S, Cao X, Raimondi SC et al. Acute mixed lineage leukemia in children: the experience of St Jude Children's Research Hospital. Blood 2009; 113: 5083–5089.

    CAS  Article  Google Scholar 

  39. 39

    Lee JH, Min YH, Chung CW, Kim BK, Yoon HJ, Jo DY et al. Prognostic implications of the immunophenotype in biphenotypic acute leukemia. Leuk Lymphoma 2008; 49: 700–709.

    CAS  Article  Google Scholar 

  40. 40

    Legrand O, Perrot JY, Simonin G, Baudard M, Cadiou M, Blanc C et al. Adult biphenotypic acute leukaemia: an entity with poor prognosis which is related to unfavourable cytogenetics and P-glycoprotein over-expression. Br J Haematol 1998; 100: 147–155.

    CAS  Article  Google Scholar 

  41. 41

    Wouters BJ, Koss C, Delwel R . Gene expression profiling for improved dissection of acute leukemia: a recently identified immature myeloid/T-lymphoid subgroup as an example. Blood Cells Mol Dis 2008; 40: 395–400.

    CAS  Article  Google Scholar 

  42. 42

    Buccheri V, Matutes E, Dyer MJS, Catovsky D . Lineage commitment in biphenotypic acute leukemia. Leukemia 1993; 7: 919–927.

    CAS  Google Scholar 

  43. 43

    Dupret C, Asnafi V, Leboeuf D, Millien C, Ben Abdelali R, Preudhomme C et al. IgH/TCR rearrangements are common in MLL translocated adult AML and suggest an early T/myeloid or B/myeloid maturation arrest, which correlates with the MLL partner. Leukemia 2005; 19: 2337–2338.

    CAS  Article  Google Scholar 

  44. 44

    Zheng C, Wu J, Liu X, Ding K, Cai X, Zhu W . What is the optimal treatment for biphenotypic acute leukemia? Haematologica 2009; 94: 1778–1780.

    Article  Google Scholar 

  45. 45

    Pui C-H, Sandlund JT, Pei D, Campana D, Rivera GK, Ribeiro RC et al. Improved outcome for children with acute lymphoblastic leukemia: results of Total Therapy Study XIIIB at St. Jude Children's Research Hospital. Blood 2004; 104: 2690–2696.

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to O K Weinberg.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Weinberg, O., Arber, D. Mixed-phenotype acute leukemia: historical overview and a new definition. Leukemia 24, 1844–1851 (2010).

Download citation


  • mixed-phenotype acute leukemia
  • review
  • biphenotypic leukemia
  • diagnosis

Further reading


Quick links