Mixed phenotype acute leukemia: 13 patients and a proposed management algorithm

Highlights

• •Thirteen patients with mixed phenotype acute leukemia managed over 10 years.
• •A B/myeloid phenotype predominated; 11 of 13 cases met the stricter WHO criteria.
• •Treatment-related mortality was high (53.8%) and mostly infectious.
• •Median overall survival was 19.2 months in this cohort.
• •A pragmatic, literature-based management algorithm is proposed.

Abstract

Mixed phenotype acute leukemia (MPAL) is a rare, heterogeneous group of acute leukemias of ambiguous lineage (2–5% of cases) with no consensus treatment. We retrospectively reviewed all patients with MPAL managed in a single Tunisian hematology department between January 2014 and December 2023. Diagnosis relied on multiparameter flow-cytometric immunophenotyping using the EGIL score and the WHO classifications; survival was estimated by the Kaplan–Meier method. Thirteen patients were included (median age 23 years, range 2–46; sex ratio ≈ 1.1). Twelve had a B/myeloid and one a T/myeloid phenotype; on cross-classification, 11 of the 13 EGIL-defined cases also fulfilled the stricter WHO criteria. Cytogenetic abnormalities were found in 10 patients, including t(9;22) in two and a complex karyotype in three. Eleven patients received an ALL-type induction, with a tyrosine kinase inhibitor in BCR-ABL1-positive disease; complete remission was obtained in 9 of 11 evaluable patients, and two underwent allogeneic stem-cell transplantation. Treatment-related mortality was high (53.8%), mainly infectious, and median overall survival was 19.2 months (1- and 5-year overall survival 54% and 30.8%). This series, one of the first dedicated cohorts from Tunisia, confirms the rarity and severity of MPAL and supports precise diagnosis, individualized multidisciplinary management, and multicenter collaboration.

1. Introduction

Acute leukemias are conventionally divided into acute myeloid leukemia (AML) and B- or T-lineage acute lymphoblastic leukemia (ALL) [1]. In 2–5% of cases, however, the blasts co-express markers of more than one lineage, defining mixed phenotype acute leukemia (MPAL) [2,3]. The European Group for the Immunological Characterization of Acute Leukemias (EGIL) proposed a scoring system in 1995 [3], subsequently refined by the WHO classifications, which rely on a restricted set of highly specific markers and now recognize genetically defined subtypes (MPAL with BCR-ABL1 and with KMT2A rearrangement) [1,4].

Most cases express a B/myeloid or T/myeloid phenotype, while the B/T subtype is exceptional [2]. Because of its rarity, biological heterogeneity and the absence of randomized trials, treatment remains non-consensual and is generally derived from ALL or AML protocols, with allogeneic hematopoietic stem-cell transplantation (allo-HSCT) frequently favored in eligible patients [5]. We describe the characteristics and outcomes of MPAL in a Tunisian center and propose an adapted management algorithm.

2. Materials and methods

We conducted a retrospective, single-center study of patients with MPAL followed in the Department of Clinical Hematology of Farhat Hached University Hospital between January 2014 and December 2023. Patients aged 2–65 years diagnosed on bone marrow aspiration with flow-cytometric immunophenotyping were included; pure myeloid or lymphoid leukemias and patients not receiving curative treatment were excluded. Cases were primarily identified using the EGIL score (>2 in ≥2 lineages) and were cross-classified against the WHO criteria (2008/2016 and the 5th edition, 2022) [3,4,6]. The work-up included blood counts, hemostasis and biochemistry panels, bone marrow morphology (FAB), immunophenotyping, conventional karyotyping and molecular screening for the BCR-ABL1 transcript; KMT2A testing was unavailable locally. Treatment was decided by a multidisciplinary team: most patients received a high-risk ALL-type regimen (EORTC 58,951), with a tyrosine kinase inhibitor (TKI) added in BCR-ABL1-positive disease; allo-HSCT was offered to patients with an HLA-identical family donor. Complete remission (CR) was defined as <5% marrow blasts with count recovery, and measurable residual disease (MRD) was assessed by flow cytometry. Overall survival (OS), event-free survival (EFS) and relapse-free survival (RFS) were estimated by the Kaplan–Meier method. The study complied with the Declaration of Helsinki.

3. Results

Thirteen patients were collected over the 10-year period. The median age was 23 years (range 2–46), with a slight male predominance (7 men; sex ratio ≈ 1.1). Presentation was dominated by an anemic syndrome (9 patients); a tumoral syndrome was present in six, fever in three, hemorrhage in three and bone pain in three. A leukocyte count above 100,000/mm³ was found in three patients; all had circulating blasts and anemia, and 12 had thrombocytopenia. On immunophenotyping, 12 patients had a B/myeloid and one a T/myeloid phenotype. All 13 fulfilled the EGIL criteria; on cross-classification, 11 also met the more restrictive WHO criteria, whereas two cases lacked the WHO myeloid-lineage requirement (definitive myeloperoxidase expression or monocytic differentiation) and would be reclassified as myeloid-antigen-positive acute leukemia (Table 1).

Table 1.

Main features of the cohort (N = 13). WBC, white blood cell count.

Characteristic	Value
Median age, years (range)	23 (2–46)
Male, n (sex ratio)	7 (≈1.1)
WBC > 100,000/mm³, n	3
B/myeloid phenotype, n	12
T/myeloid phenotype, n	1
Cytogenetic abnormality, n	10
Complex karyotype, n	3
BCR-ABL1 positive, n	2

Conventional karyotyping detected abnormalities in 10 patients, including a complex karyotype in three, t(9;22) in two, t(8;21) with loss of Y in one and t(9;11) in one; the BCR-ABL1 transcript was detected in two patients. Individual data are summarized in Table 2. Eleven patients were treated according to the high-risk EORTC 58,951 protocol, one with the Saint-Jude protocol and one with an AML-type regimen; BCR-ABL1-positive patients also received a TKI. Two patients (15.3%) died during induction from toxicity; CR was obtained in 9 of 11 evaluable patients (81.8%). Two patients underwent allo-HSCT, one dying from pulmonary graft-versus-host disease. Febrile neutropenia occurred in all patients, bacterial infection in nine and fungal infection in six. Seven patients (53.8%) relapsed during maintenance, predominantly B/myeloid forms.

Table 2.

Individual characteristics, treatment and outcome of the 13 patients. CR, complete remission; allo-HSCT, allogeneic hematopoietic stem-cell transplantation; OS, overall survival; EFS, event-free survival; neg., negative; pos., positive; n.a., not available.

No.	Age/Sex	Phenotype	Cytogenetics; BCR-ABL1	Protocol	Outcome (OS/EFS, mo)
1	5/M	T/myeloid	Normal; neg.	EORTC VHR	Early relapse, died (21.1/7.1)
2	34/M	B/myeloid	pH+; pos.	EORTC VHR + dasatinib	Septic shock, died (2.7/2.7)
3	29/F	B/myeloid	Hyperdiploid; neg.	EORTC VHR	CR, referred for allo-HSCT; alive (54.1/54.1)
4	21/M	B/myeloid	t(9;22), −7; pos.	EORTC VHR + imatinib	Relapses, T315I, died (29/7.5)
5	46/M	B/myeloid	Complex; neg.	EORTC VHR	Death during induction (2/2)
6	7/M	B/myeloid	t(8;21), −Y; neg.	EORTC VHR → AML	Relapse, septic shock, died (6.5/5.7)
7	3/F	B/myeloid	del(12p); neg.	EORTC VHR	Relapse, palliative, died (9/3.5)
8	23/F	B/myeloid	Trisomy 1q; neg.	EORTC VHR → FLAG-IDA	Died (12/5)
9	25/M	B/myeloid	Normal; neg.	EORTC VHR + allo-HSCT	Pulmonary GVHD, died (19.2/19.2)
10	26/F	B/myeloid	t(9;11); neg.	AML-type	Alive, well (69.3/69.3)
11	23/F	B/myeloid	Normal; neg.	EORTC VHR	Death day 9 post 3+7 (9.9/8.2)
12	2/M	B/myeloid	n.a.; n.a.	Saint-Jude HR	Alive, well (127.5/25.1)
13	13/F	B/myeloid	del(1q); neg.	EORTC VHR + salvage	Alive, good health (29.8/4.3)

The median follow-up was 27 months. Median OS was 19.2 months, with 1- and 5-year OS of 54% and 30.8% (Fig. 1); median EFS was 7.1 months, with 1-year EFS of 31% (Fig. 2). Nine deaths occurred (three from relapse, six from infection or multi-organ failure); four patients were alive at the end of the study. The exploratory univariate analysis showed no significant prognostic association, an absence to be interpreted cautiously given the small sample size.

Fig. 1.

Kaplan–Meier estimate of overall survival (N = 13). Ticks indicate censored patients.

Fig. 2.

Kaplan–Meier estimate of event-free survival (N = 13). Ticks indicate censored patients.

4. Discussion

MPAL accounts for fewer than 5% of acute leukemias [7]. Our cohort of 13 patients over a decade, with a median age of 23 years, a slight male predominance and a marked B/myeloid predominance, is consistent with the literature [2,8]. The discordance we observed between the EGIL score and the WHO criteria (13 versus 11 cases) is expected and well described: EGIL scores numerous markers, whereas the WHO classification relies on a few highly specific lineage-defining markers—in particular myeloperoxidase for myeloid assignment—and is therefore more restrictive [6,3]. The cytogenetic profile, with t(9;22)/BCR-ABL1 in two patients and complex karyotypes in three, reflects the heterogeneous and often unfavorable genomic landscape of this entity [9].

Therapeutically, the predominant use of an ALL-type backbone with addition of a TKI in BCR-ABL1-positive cases mirrors current practice, ALL-directed regimens generally yielding higher remission rates than AML-directed ones [5]. Nonetheless, our outcomes were poor, with high treatment-related mortality (53.8%), frequent relapse and short survival, in keeping with reported series [8,10]. The predominance of infectious deaths highlights the importance of intensive supportive care, and allo-HSCT remains the cornerstone of consolidation in eligible patients [5].

On the basis of our experience and the literature, we propose a pragmatic algorithm (Fig. 3): an ALL-type induction (± TKI for BCR-ABL1-positive disease); assessment of corticosensitivity and MRD; allo-HSCT in first remission for patients with a donor or high-risk features, otherwise ALL-type consolidation; a switch to an AML-type regimen (e.g. FLAG-IDA) for induction failure or persistent MRD; and, at relapse, blinatumomab for CD19-positive disease, hypomethylating agents plus venetoclax for myeloid phenotypes, and TKI switching for persistent BCR-ABL1.

Fig. 3.

Proposed decision algorithm for the management of MPAL. TKI, tyrosine kinase inhibitor; MRD, measurable residual disease; CR1/CR2, first/second complete remission; allo-HSCT, allogeneic hematopoietic stem-cell transplantation; HMA, hypomethylating agent.

The strengths of this work include a homogeneous single-center cohort and its contribution to the scarce regional literature. Its limitations are the small sample size, the retrospective design, and the absence of advanced molecular characterization (no KMT2A testing or next-generation sequencing), related to local constraints.

5. Conclusion

MPAL is a rare and aggressive entity with diagnostic and therapeutic challenges. Our series confirms its severity and the high rate of early, mainly infectious, events, and underscores the need for rigorous immunophenotypic and cytogenetic diagnosis, individualized multidisciplinary management and, when feasible, allogeneic stem-cell transplantation. Multicenter collaboration is needed to refine therapeutic strategies for this orphan disease.

Ethics

The study was conducted in accordance with the Declaration of Helsinki. Ethical review and approval were not required under local legislation and institutional requirements, as this was a retrospective analysis of anonymized, routinely collected clinical data, for which individual informed consent was waived.

Declaration of generative AI in scientific writing

During the preparation of this work, the authors used a generative AI assistant (Anthropic Claude) in order to assist with drafting and language editing of the text, formatting, and rendering of the figures from the authors’ own data and content. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Statement for studies in humans/animals

This study was conducted in accordance with the Declaration of Helsinki. As a retrospective analysis of anonymized, routinely collected clinical data, ethical review and approval were not required under local legislation and institutional requirements, and the requirement for individual informed consent was waived.

No animals were involved in this study.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Wafa Chenbah: Conceptualization, Data curation, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing. Mohamed Ben Abdelkrim: Conceptualization, Formal analysis, Methodology, Writing – original draft. Manel Kasdallah: Data curation, Validation, Visualization. Nadia Sassi: Visualization, Writing – original draft. Monia Guermazi: Methodology, Writing – original draft. Yosra Ben Youssef: Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.