Oncofusion

start / graph / pt-story-infant-all-kmt2a-aff1-hyperleukocytosis

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

A six-month-old with a week of coughing: KMT2A::AFF1 pro-B ALL presenting with extreme hyperleukocytosis

also called infant ALL leukapheresis case · t(4;11) hyperleukocytic infant case

A six-month-old boy was brought in after a week of coughing and wheezing that had been treated as a chest problem; the blood count that was finally sent showed a white cell count hundreds of times too high and a haemoglobin of 49 g/L. Two sessions of a machine that filters white cells out of the blood, then chemotherapy, brought him into remission, and at thirteen months of age he was growing normally.

The same thing, for a scientist

A 6-month-old male infant presented with one week of paroxysmal cough and wheeze; a pre-admission complete blood count showed WBC 1,560 x 10^9/L, Hb 49 g/L, platelets 32 x 10^9/L. On admission the WBC was 1,755 x 10^9/L with 94% blasts. Liver was 4 cm and spleen 5 cm below the costal margin. Flow cytometry showed a pro-B phenotype (CD34+, CD19+, cCD79a+, TdT+, CD9+, HLA-DR+, CD123+, CD38+; CD10-, CD20-, CD22-, MPO-, CD13-, CD33-), and CSF contained 21.16% abnormal CD19+CD38+CD34+ B lymphoblasts. Karyotype was 46,XY,t(4;11)(q21;q23); a 57-fusion leukaemia panel and whole-transcriptome sequencing both detected KMT2A::AFF1, with the reciprocal AFF1::KMT2A also called. Immune-repertoire sequencing found no clonal IGH, IGK or IGL rearrangement. He deteriorated to intubation for hypoxia, underwent two leukaphereses (70 and 100 min) with low-dose dexamethasone cover, and the WBC fell to 55.63 x 10^9/L, a 96% reduction, without tumour lysis or apheresis complications. Treated on CCCG-iALL-2022 with high-dose methotrexate, he cleared MRD and the KMT2A::AFF1 transcript, with a normal repeat karyotype, and was reported well at 1 year 1 month of age. [E1]

Go deeper
How it began

Not with a lump, not with a rash, and not with anything that looks like cancer. The report records one week of paroxysmal coughing without obvious cause and difficulty clearing sputum, then wheezing that worsened with activity the day before admission. He was given nebulised treatment and intramuscular dexamethasone, and the wheeze improved - which is what a viral wheeze does, and is part of why this diagnosis gets missed. What changed the direction of the case was a complete blood count: white cells 1,560 x 10^9/L, haemoglobin 49 g/L, platelets 32 x 10^9/L 1.

That combination - the anaemia (concept-pallor-and-fatigue-anaemia), the thrombocytopenia (concept-bruising-petechiae-thrombocytopenia) and the white count - is the blood-count picture described in concept-blood-film-and-blood-count-leukemia. Here it arrived inside a presentation that a general paediatric service sees many times a week and which almost never turns out to be leukaemia.

On examination the liver was palpable 4 cm and the spleen 5 cm below the costal margin, firm - the hepatosplenomegaly of concept-hepatosplenomegaly-infant-leukemia, present but not what brought him in.

What was found1 deeper

The white count on the day after admission was 1,755 x 10^9/L with 94% blasts. Marrow was hypercellular with small, regular primitive cells showing nuclear indentation and clefting, scanty agranular cytoplasm and negative peroxidase 1.

The immunophenotype was pro-B and CD10-negative: CD34+, CD19+, cCD79a+, TdT+, CD9+, HLA-DR+, CD123+, CD38+, with CD10, CD20, CD22, CD13, CD33, CD117 and MPO all negative. That is the phenotype concept-flow-cytometry-immunophenotype-kmt2a-r describes, and it was present in the CSF too: 21.16% of nucleated cells in the cerebrospinal fluid were CD19+CD38+CD34+ abnormal B lymphoblasts, that is, CNS involvement at diagnosis (concept-cns-involvement-at-diagnosis).

Karyotype was 46,XY,t(4;11)(q21;q23) - the classic finding of concept-cytogenetics-karyotype-kmt2a. A 57-gene fusion panel and whole-transcriptome sequencing independently called KMT2A::AFF1, and the transcriptome also called the reciprocal AFF1::KMT2A at lower confidence (concept-molecular-confirmation-fusion-transcript). Immunoglobulin-repertoire sequencing found no dominant clonal IGH, IGK or IGL rearrangement - consistent with a clone arrested before the point at which the B-cell receptor is assembled 1.

A number to treat with care

The reported white count, 1,755 x 10^9/L (the paper also writes it as 1,755,000/mm3, so the unit is internally consistent), is far above the figures usually quoted in the hyperleukocytosis literature, where the threshold for concern is 100 x 10^9/L and series describe counts in the hundreds (concept-hyperleukocytosis-emergency). This entry reports the figure the paper reports. It does not vouch for it. A single case report is the weakest possible evidence for an extreme value, and extreme values are exactly what gets published.

What was done and what happened

Transcutaneous oxygen saturation fell to 75-78% with sudden dyspnoea and he was intubated and ventilated. He received two leukaphereses, of 70 and 100 minutes, on a continuous-flow separator through a 6.5 F femoral catheter, with low-dose dexamethasone (3.2 mg/m2, then 6 mg/m2) given around the procedures to begin cytoreduction and limit tumour lysis. The white count fell 1,755 to 946.52 to 493.47 to 55.63 x 10^9/L, a 96% reduction, with no tumour lysis syndrome and no apheresis complication recorded. Between the two procedures he had a fever to 38.5 C and one seizure, managed with sedation and osmotherapy 1.

He was weaned from the ventilator on the third post-procedure day and left intensive care on the fifth. He was then treated on the Chinese Children's Cancer Group infant ALL 2022 protocol with high-dose methotrexate for about half a year, and the report states that MRD fell below the assay threshold, the KMT2A::AFF1 transcript became undetectable, and the repeat karyotype was 46,XY with no tumour-associated abnormality. At the time of writing he was 1 year 1 month old, with normal counts and growth and development comparable to his peers 1.

The report describes remission at roughly seven months from diagnosis. It does not describe cure. KMT2A-rearranged infant ALL relapses predominantly in the first two years, so a child in remission at thirteen months of age is at the beginning of the risk period, not past it.

Why this case is in the graph

Seventeen clinical concept entries in this graph describe signs, tests and emergencies in the abstract. None of them can tell you that the sign which brings a baby to hospital may be a cough.

Three things this case teaches that a category entry cannot:

  1. The presenting complaint and the diagnosis need not resemble each other. A week of wheeze, improved by a nebuliser, over a white count of 1,560 x 10^9/L.
  2. CNS disease can be present at diagnosis and silent. A fifth of the CSF nucleated cells were blasts; nothing in the presentation pointed at the CNS.
  3. Hyperleukocytosis is a mechanical emergency before it is an oncological one. What nearly killed him in the first days was viscosity and leukostasis, treated with a centrifuge, not with anything directed at the fusion.

It also instantiates the commonest lesion in the whole graph. Roughly half of infant ALL is KMT2A::AFF1, and this is what one of those children looked like on the day they arrived.

Facts and figures
age at dx6 months
sex as reportedmale
fusionfusion-kmt2a-aff1
karyotype46,XY,t(4;11)(q21;q23)
cns statusCSF contained 21.16% abnormal CD19+CD38+CD34+ B lymphoblasts
courseintubated for hypoxia; two leukaphereses (70 min, 100 min) with low-dose dexamethasone; WBC 1,755 to 946.52 to 493.47 to 55.63 x 10^9/L; one febrile seizure between procedures; extubated day 3 post-procedure; approximately six months of high-dose methotrexate on CCCG-iALL-2022
outcomein remission with undetectable MRD and undetectable KMT2A::AFF1 transcript, normal repeat karyotype, reported well at 1 year 1 month of age; follow-up approximately 7 months from diagnosis; the report does not describe cure
source typecase report
consent notePublished peer-reviewed case report; the paper states the patient's guardian signed an informed consent form and that the apheresis procedure was approved under a named local new-technology application. This entry records only what the publication discloses. The treating institution named in the paper is deliberately omitted from this entry's body text, because combining it with an exact age, sex and date narrows identity further than the clinical teaching requires.
privacy reviewNo detail added, inferred or reconstructed beyond the publication. Ethnicity as stated in the paper is omitted here as adding no clinical information.
labs at presentation
platelets_x10e9_per_L32.0platelets_x10e9_per_L: 32haemoglobin_g_per_L49.0haemoglobin_g_per_L: 49blasts_pct_peripheral94.0blasts_pct_peripheral: 94wbc_preadmission_x10e9_pe…1,560wbc_preadmission_x10e9_per_L: 1560wbc_x10e9_per_L1,755wbc_x10e9_per_L: 1755
presentation
[
  "one week of paroxysmal cough and difficulty clearing sputum",
  "wheeze worse on exertion, improved after nebulisation and intramuscular dexamethasone",
  "hepatomegaly 4 cm below costal margin",
  "splenomegaly 5 cm below costal margin, firm",
  "oxygen saturation 92% on admission, falling to 75-78% with sudden dyspnoea"
]
labs at presentation — the numbers behind the chart
{
  "wbc_x10e9_per_L": 1755,
  "wbc_preadmission_x10e9_per_L": 1560,
  "blasts_pct_peripheral": 94,
  "haemoglobin_g_per_L": 49,
  "platelets_x10e9_per_L": 32
}
immunophenotype positive
[
  "CD34",
  "CD19",
  "cCD79a",
  "TdT",
  "CD9",
  "HLA-DR",
  "CD123",
  "CD38"
]
immunophenotype negative
[
  "CD10",
  "CD20",
  "CD22",
  "CD13",
  "CD33",
  "CD117",
  "MPO",
  "CD7",
  "CD2",
  "CD56"
]
molecular confirmation
[
  "57-fusion leukaemia panel: KMT2A::AFF1",
  "whole-transcriptome sequencing: KMT2A::AFF1 plus reciprocal AFF1::KMT2A",
  "no clonal IGH/IGK/IGL rearrangement by immune-repertoire sequencing"
]
How this could bite us back
  • A fusion-directed agent does nothing about viscosity. Whatever the menin-inhibitor era delivers, the first 48 hours of a presentation like this are still centrifuges, hydration and ventilation, and a child can die in that window with a perfectly drugged target.
  • CSF involvement at diagnosis means the CNS compartment is seeded before any systemic therapy starts, so a drug's CNS penetration is a requirement and not a refinement.
  • Remission at seven months in infant KMT2A-rearranged ALL is a weak surrogate. Reporting a case at this stage as a success risks creating an optimistic prior that the Interfant-06 event-free survival of roughly 36% does not support.
Still open
  1. How often does KMT2A-rearranged infant ALL first present to a general or respiratory service rather than to haematology, and what is the median interval from first contact to blood count? No cohort in this graph answers this, and it is answerable from Interfant registry data.
  2. This child had 21% blasts in the CSF with no neurological signs; what proportion of infants with KMT2A::AFF1 have occult CNS disease at diagnosis, and does CSF blast fraction at diagnosis predict CNS relapse independently of WBC?
  3. Does emergency leukapheresis alter outcome in infants, or only the first 72 hours? The reported evidence is contradictory and the question is answerable only by a registry comparison, not by more case reports.
  4. The absence of any clonal IGH/IGK/IGL rearrangement places the clone before B-cell-receptor assembly; does that pre-rearrangement state track with the fetal pre-pro-B cell of origin, and is it detectable at birth?
Every source behind this entry (1)
  1. E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    A 6-month-old male infant presented with one week of cough and wheeze and was found to have pro-B ALL with 46,XY,t(4;11)(q21;q23) and a KMT2A::AFF1 fusion, WBC 1,755 x 10^9/L with 94% blasts, Hb 49-56 g/L, platelets 31-32 x 10^9/L, hepatomegaly 4 cm and splenomegaly 5 cm, 21.16% abnormal B lymphoblasts in CSF, and a CD34+CD19+cCD79a+TdT+CD10- phenotype; after intubation and two leukaphereses the WBC fell to 55.63 x 10^9/L (96% reduction) with no tumour lysis, and after high-dose methotrexate on the CCCG-iALL-2022 protocol MRD and the fusion transcript became undetectable with a normal repeat karyotype, the child being well at 1 year 1 month of age.

    What this evidence cannot show

    Full text read (PubMed Central, PMC11663633), not abstract only. A single case report: it establishes that this happened to this child and nothing about frequency, and publication bias selects hard for the unusual - a white count of 1,755 x 10^9/L is published precisely because it is extraordinary, so this case must not be read as a picture of typical infant ALL. The reported leukocyte count is far above values elsewhere in the hyperleukocytosis literature and this entry reports it without endorsing it; the PMC text rendering loses superscripts, although the paper's parallel '1,755,000/mm3' makes the intended unit unambiguous. Follow-up is about seven months from diagnosis, short relative to the relapse-risk window in KMT2A-rearranged infant ALL, so 'recovered' in the paper means remission, not cure. The report does not state the breakpoint intron, so the fusion is recorded at gene level only. Single centre, no independent verification of the counts.

    Evidence that argues against this (2)
    • The benefit of leukapheresis in very young children with hyperleukocytosis is contested; the report itself cites evidence, including an international multicentre retrospective study, that support for the procedure in this age group is insufficient, so this single favourable case does not establish efficacy. source 2024
    • Against reading this child's outcome as representative: in the Interfant-06 trial the KMT2A-rearranged group had a 6-year event-free survival of about 36%, so the modal outcome for an infant with this lesion is not the one recorded here. source 2021
Connected to