Oncofusion

start / graph / disease-mpal

Diseasemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q-patient-stories

Mixed phenotype acute leukaemia (MPAL) and acute leukaemia of ambiguous lineage

also called MPAL · biphenotypic acute leukaemia · bilineal acute leukaemia · acute leukaemia of ambiguous lineage · ALAL

In a few children the leukaemia cells do not commit to being either the lymphoid kind or the myeloid kind, and carry markers of both - sometimes as two separate populations in the same marrow. It matters practically, because which of the two standard treatment programmes the child is put on changes survival more than almost any other decision.

The same thing, for a scientist

MPAL and the wider category of acute leukaemia of ambiguous lineage (ALAL) cover leukaemias whose blasts cannot be assigned to a single lineage, either as one population co-expressing defining markers of two lineages (biphenotypic) or as two distinct populations (bilineal). It accounts for roughly 2-5% of childhood leukaemia. The largest paediatric cohort, the retrospective international iBFM-AMBI2012 study of 233 children, found that primary ALL-type therapy gave a 5-year event-free survival of 80% +/- 4%, against 36% +/- 7.2% for AML-type and 50% +/- 12% for combined-type primary therapy; for CD19-positive leukaemia with ALL- and AML-specific gene fusions excluded, the contrast was 83% +/- 5.3% on ALL-type versus 0% +/- 0% on AML-type therapy. No overall benefit of transplantation was documented. KMT2A rearrangement is over-represented: the same cohort contained 15 children with a KMT2A::AFF1 rearrangement and 11 with another KMT2A rearrangement, and these aggregated in the poor-outcome hybrid-treatment group. [E1][E2] This is a minimal entry created so that patient-story edges resolve; it is not a full treatment of the disease.

Go deeper
Why this entry exists and what it does not cover

This is a minimal, fully cited placeholder created by the patient-stories queen so that an edge from a patient entry with B/myeloid MPAL resolves to a disease node with the canonical id disease-mpal. It covers the definition, the frequency, the treatment-strategy finding and the KMT2A over-representation, and nothing else. A full entry should add: the WHO and ICC definitional changes, the genomic subgroups (ZNF384, BCL11B), MRD methodology in a bilineal disease, and the adult literature.

The treatment-assignment finding

The single most consequential result in paediatric ALAL is that the choice of primary protocol dominates outcome. In 233 children, 5-year EFS was 80% +/- 4% with ALL-type primary therapy against 36% +/- 7.2% with AML-type and 50% +/- 12% with combined-type. Restricting to CD19-positive disease with ALL- and AML-specific fusions excluded sharpened it further: 83% +/- 5.3% versus 0% +/- 0% on AML-type primary treatment. The authors recommend ALL-type protocols for the majority, reserve AML-type treatment for a minority that is CD19-negative with no other lymphoid feature, found no overall benefit from transplantation, and advise changing treatment type only when 5% or more blasts persist after induction 1.

The relevance to this graph is that KMT2A-rearranged cases sit disproportionately in the group that did badly. The cohort contained 26 children with a KMT2A rearrangement, and they aggregated among patients given hybrid lymphoid-and-myeloid treatment 2.

Facts and figures
transplantno overall benefit documented in iBFM-AMBI2012
noteminimal placeholder entry created by the patient-stories queen so that patient-story edges resolve; not a full disease entry
kmt2a in cohort
other_kmt2a11.0other_kmt2a: 11kmt2a_aff115.0kmt2a_aff1: 15total26.0total: 26cohort_n233.0cohort_n: 233
frequency childhood leukaemia PCT
[
  2,
  5
]
subtypes
[
  "biphenotypic (single population co-expressing two lineages)",
  "bilineal (two distinct populations)"
]
efs 5y by primary therapy
{
  "ALL_type_pct": 80,
  "ALL_type_se": 4,
  "AML_type_pct": 36,
  "AML_type_se": 7.2,
  "combined_type_pct": 50,
  "combined_type_se": 12,
  "cohort": "iBFM-AMBI2012, n=233 children"
}
kmt2a in cohort — the numbers behind the chart
{
  "kmt2a_aff1": 15,
  "other_kmt2a": 11,
  "total": 26,
  "cohort_n": 233
}
How this could bite us back
  • A lineage-agnostic target such as the menin-KMT2A interface is in principle the right answer to a lineage-ambiguous disease, but it inherits the same problem in reverse: any escape mechanism that is not about lineage, such as a MEN1 binding-site mutation, is unaffected by which clone is dominant.
Still open
  1. Does the ALL-type-therapy advantage in ambiguous-lineage leukaemia hold within the KMT2A-rearranged subgroup specifically, where lineage plasticity is greatest? The iBFM-AMBI2012 cohort has 26 such children and the question is answerable from existing data.
  2. In a bilineal leukaemia, what should MRD be measured on - the lymphoid clone, the myeloid clone, or the shared genomic breakpoint? A breakpoint-based assay tracks the clone regardless of phenotype and is the only measure that survives a lineage switch.
Every source behind this entry (2)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In the retrospective international iBFM-AMBI2012 study of 233 children with acute leukaemia of ambiguous lineage, 5-year event-free survival was 80% +/- 4% with ALL-type primary therapy versus 36% +/- 7.2% with AML-type and 50% +/- 12% with combined-type therapy; for CD19-positive leukaemia excluding ALL- and AML-specific gene fusions it was 83% +/- 5.3% on ALL-type versus 0% +/- 0% on AML-type primary treatment; no overall benefit of transplantation was documented, and a change of treatment type was advised only for patients with 5% or more blasts after remission induction.

    PMID 29720486DOI
    What this evidence cannot show

    Abstract read, not full text. A retrospective multinational study, so treatment assignment was not randomised and the ALL-type versus AML-type comparison is open to confounding by the immunophenotypic features that drove the assignment in the first place - the authors present it as a basis for a prospective trial rather than as a trial result. Definitions of ambiguous lineage changed over the study period and the cohort spans 2002-2015. The 0% +/- 0% figure rests on a small subgroup.

    Evidence that argues against this (1)
    • None found that contests the direction of the treatment-strategy finding; the standing limitation is that it is retrospective and has not yet been tested prospectively, which the authors state. source 2018
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    MPAL accounts for 2-5% of leukaemia in children, with event-free survival under 50%, and lineage switch or persistence of only one of the two clones is frequently observed in biphenotypic and bilineal cases; in the iBFM-AMBI2012 cohort of 233 children, 15 had a KMT2A::AFF1 rearrangement and 11 another KMT2A rearrangement, and these most often aggregated in the group given hybrid lymphoid-and-myeloid treatment, whose 5-year event-free survival was 28%.

    PMID 33718232DOI
    What this evidence cannot show

    Full text read (PMC7953899). These figures are reported in the introduction and discussion of a case report citing other studies, so they are second-hand for everything except the case itself; the KMT2A subgroup counts and the 28% hybrid-treatment EFS are attributed to the iBFM-AMBI2012 study and should ideally be read from that paper directly. Note that 28% is a subgroup figure and does not match the 50% +/- 12% combined-type figure in the AMBI2012 abstract, which covers a different denominator; both are recorded here rather than reconciled.

    Evidence that argues against this (1)
    • MPAL frequency depends on the immunophenotypic criteria in force, which the WHO has revised repeatedly, so "2-5% of leukaemia in children" is a statement about classification practice as much as about disease. Lineage switch being "frequently observed" in biphenotypic and bilineal cases is also observed in cases selected for having ambiguous phenotype to begin with. Blood Cells, Molecules and Diseases 2010
Connected to