Step 1: the blood count and the blood film
also called full blood count · FBC · CBC · peripheral blood smear · blood smear · peripheral film
The whole diagnosis usually starts with one tube of blood: a machine counts the cells, then a person looks at a stained drop of that blood under a microscope and sees cells that should not be in the bloodstream. That takes under an hour, and it is what turns a pale, bruising baby into a leukaemia emergency the same afternoon.
The same thing, for a scientist
The full blood count and peripheral blood film are the entry point to every KMT2A-rearranged infant leukaemia diagnosis. The count typically shows anaemia and thrombocytopenia with a white cell count that is either markedly elevated - hyperleukocytosis above 30 x 10^9/L is characteristic of KMT2A-rearranged ALL, and the Interfant-06 high-risk threshold sits at 300 x 10^9/L - or paradoxically normal-to-low when a packed marrow releases few blasts [E1][E2]. The film provides the first direct look at the blast, but blast morphology in KMT2A-rearranged disease is non-specific: it establishes that an acute leukaemia is present and triggers urgent referral, not which lesion drives it. Because the film cannot lineage-assign or genotype, a bone marrow aspirate with a trephine biopsy follows immediately to supply material for flow cytometry, karyotype, FISH, molecular characterisation and the minimal residual disease baseline.
What the first tube of blood actually says2 deeper
The automated analyser returns three numbers that, taken together, are the signature of a marrow that has been displaced by something:
| Parameter | Typical direction in infant acute leukaemia | Why |
|---|---|---|
| Haemoglobin | low (anaemia) | erythropoiesis crowded out of the marrow |
| Platelets | low (thrombocytopenia) | megakaryopoiesis crowded out; explains the bruising and petechiae that often bring the baby in |
| White cell count | bidirectional - very high or normal/low | depends on whether the blast clone is spilling into the circulation or is trapped in a packed marrow |
The white cell count is the one that misleads. KMT2A-rearranged ALL is specifically characterised by hyperleukocytosis, conventionally a white cell count above 30 x 10^9/L, together with a relatively high rate of central nervous system involvement and an aggressive early-relapsing course 1. The high end is genuinely extreme: Interfant-06 used a white cell count of 300 x 10^9/L or more - roughly thirty times the upper limit for a healthy infant - as one of the criteria defining its high-risk group, alongside KMT2A rearrangement and age over six months 2. In the earlier Interfant-99 cohort a very high white cell count was one of only four factors independently associated with inferior event-free survival, the others being KMT2A rearrangement itself, age under six months, and a poor response to the prednisone prophase 3.
So a white count of 400 x 10^9/L in a three-month-old is not an artefact. It is a prognostic variable and, at that level, a leukostasis risk that changes management before any genetic result exists.
Why the white count can also be low, and why that is not reassuring
A marrow can be more than 90 per cent blasts and still deliver a white cell count in the normal range, or below it. Two things happen at once: residual normal granulocyte and lymphocyte production collapses, and the blast clone is not necessarily released into the blood. The net count can therefore land anywhere. A normal white cell count with a low haemoglobin and low platelets is, if anything, a more dangerous presentation than hyperleukocytosis, because it looks less alarming on the printout and gets attributed to a viral illness or to immune thrombocytopenia.
That misattribution is documented. A child presenting with isolated-looking thrombocytopenia was admitted with a presumptive diagnosis of immune thrombocytopenic purpura and treated with intravenous immunoglobulin; only on expert re-review of the film the next day were blasts identified, after which a marrow aspirate confirmed precursor-B ALL 5. The operational lesson is specific and unglamorous: corticosteroids must not be given for presumed immune thrombocytopenia until leukaemia is excluded, because steroid exposure partially treats the leukaemia, lyses blasts, delays the diagnosis and degrades the diagnostic sample 5.
What a KMT2A-rearranged blast looks like down the microscope
The film is where a human being first sees the cell. A KMT2A-rearranged lymphoblast is typically a medium-to-large cell with a high nuclear-to-cytoplasmic ratio, fine open chromatin, one or more nucleoli, scant agranular basophilic cytoplasm, and no Auer rods. Some cases carry a monocytoid or vacuolated appearance that hints at the myeloid co-expression flow cytometry will later confirm.
The honest point is that none of that is diagnostic. Morphology cannot distinguish a KMT2A-rearranged pro-B blast from an ETV6::RUNX1 blast, and in infants with myeloid co-expression it cannot reliably distinguish ALL from AML at all. The film's job is to answer one binary question - are there blasts in this baby's blood - and to answer it in minutes. Every subsequent step exists because the film cannot do more than that. The clinical entity is defined by immunophenotype and genetics, which is why the disease is classified by KMT2A break-apart status rather than by cell appearance 1.
Turnaround is the whole argument for doing this first
An automated count takes minutes on the analyser. A stained film can be ready in roughly half an hour and read in the same session. Nothing else in the diagnostic pathway is remotely that fast: a G-banded karyotype needs 24-48 hours of culture before anyone can look at a chromosome, FISH is typically an overnight hybridisation, and full molecular characterisation of the fusion partner runs from days to weeks.
That asymmetry is why the film sits first. It is the test that decides tonight's decisions - transfer to a paediatric oncology unit, central venous access, hydration and tumour lysis prophylaxis, whether leukapheresis or exchange transfusion is on the table for hyperleukocytosis - while the tests that will actually name the disease are still incubating. The Interfant protocols built a deliberately slow, cheap read-out into the same window: the seven-day prednisone prophase, whose end point is measured on the peripheral blood, was used to stratify risk before any of the genetics returned 3.
Why a bone marrow aspirate and trephine always follows1 deeper
The film cannot end the work-up, for four reasons.
- Quantification. Diagnosis and response assessment are anchored on marrow blast percentage, not blood blast percentage.
- Material for the assays that matter. Flow cytometry, karyotype, FISH and molecular studies all need viable marrow cells in the right anticoagulant. A KMT2A break-apart result cannot be obtained from a stained slide.
- The MRD baseline. The diagnostic marrow is the only chance to establish the patient-specific target that later minimal residual disease measurement will track - see
concept-mrd-monitoring-genomic-breakpoint. Miss it and the child is monitored on a generic assay for the rest of treatment. - Architecture. A trephine core shows whether the marrow is packed, fibrotic or hypocellular, which matters particularly in infants where a dry or diluted aspirate is common.
In infants this step is not trivial. The procedure needs general anaesthesia, the available marrow volume is small, and a heavily packed marrow frequently yields a poor aspirate. Getting enough material in one pass, and splitting it correctly between the immunophenotyping tube, the cytogenetics culture, the FISH slide and the molecular aliquot, is the single highest-leverage technical decision in the entire diagnostic pathway.
The clinical picture the film is read against
The haematologist is never reading the film blind. KMT2A-rearranged blast surface phenotype was mapped to its clinical correlates in the original prospective NG2 screen: cases carrying the KMT2A-associated antigen were significantly more likely to be infants, to present with hyperleukocytosis, and to have central nervous system involvement, as well as being CD10-negative and expressing the myeloid-associated antigens CD15 and CD65 4. Hepatosplenomegaly and, in infants specifically, leukaemia cutis are part of the same picture.
Put differently: by the time the film is on the microscope stage, the combination of a very young infant, a very high white count, an enlarged liver and spleen and blasts with scant cytoplasm already carries a strong prior for a KMT2A rearrangement. The rest of the pathway exists to convert that prior into a fact.
| assay | full blood count plus May-Grunwald-Giemsa or Wright-stained peripheral blood film |
|---|---|
| specimen | EDTA peripheral blood, typically 0.5-1 mL in an infant |
| next step | bone marrow aspirate plus trephine biopsy, split between flow cytometry, cytogenetic culture, FISH slide and molecular aliquot |
| morphology notes | high nuclear-to-cytoplasmic ratio, fine chromatin, scant agranular basophilic cytoplasm, no Auer rods; monocytoid or vacuolated forms occur; not distinguishable from other B-ALL genotypes by eye |
turnaround
{
"automated_count": "minutes",
"stained_film_ready": "about 30 minutes",
"morphologist_report": "same working session",
"comparison_karyotype": "24-48 h culture before first read",
"comparison_molecular": "days to weeks"
}presenting signals
[
{
"parameter": "haemoglobin",
"direction": "low",
"note": "anaemia from displaced erythropoiesis"
},
{
"parameter": "platelets",
"direction": "low",
"note": "thrombocytopenia; bruising and petechiae are common presenting complaints"
},
{
"parameter": "white cell count",
"direction": "high or low",
"note": "hyperleukocytosis above 30 x 10^9/L is characteristic of KMT2A-rearranged ALL, but a packed marrow can give a normal or low count"
},
{
"parameter": "blasts on film",
"direction": "present",
"note": "the finding that converts a general paediatric presentation into an oncological emergency; morphology alone is not lesion-specific"
}
]thresholds
[
{
"label": "hyperleukocytosis, conventional definition",
"value": 30,
"unit": "x10^9/L",
"source": "Gorecki 2023 review (PMID 36979800)"
},
{
"label": "Interfant-06 high-risk white cell count criterion",
"value": 300,
"unit": "x10^9/L",
"source": "Pieters 2019 Interfant-06 (PMID 31283407)"
}
]- A normal or low white cell count makes the presentation look benign on the printout while the marrow is over 90 per cent blasts. Any diagnostic pathway keyed to a high white count will systematically miss the quiet presentations.
- Blast morphology is not lesion-specific, so a film-based diagnosis can never identify a KMT2A rearrangement. Treating the film as more informative than it is delays the tests that actually stratify risk.
- Corticosteroids given for a presumed benign cytopenia partially treat the leukaemia: they lyse blasts, can normalise the count, and degrade the diagnostic marrow so that immunophenotyping and cytogenetics are run on a depleted sample.
- Emergency cytoreduction for hyperleukocytosis (leukapheresis, exchange transfusion, early steroid) can reduce blast burden before the diagnostic marrow is taken, compromising cell yield for cytogenetic culture and the MRD baseline - the one sample that cannot be repeated.
- In infants the aspirate volume is small and dry or dilute taps are common, so how the sample is split is effectively irreversible; over-allocating to any one assay silently forecloses another.
- What is the actual distribution of presenting white cell count, haemoglobin and platelet count in KMT2A-rearranged versus KMT2A-germline infants in a modern consecutive cohort? The literature reports protocol thresholds and categorical 'hyperleukocytosis' far more often than it reports medians and interquartile ranges.
- Can digital morphology with machine learning flag a KMT2A-rearranged blast from the film alone, hours before flow cytometry, at a precision that would change the first day of management?
- How often is the initial film in a cytopenic infant reported as unremarkable, and would mandatory haematopathologist re-review change time to diagnosis?
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.review
KMT2A-rearranged ALL is characterised by hyperleukocytosis (white cell count above 30 x 10^9/L), a relatively high incidence of CNS involvement and an aggressive early-relapsing course, and KMT2A rearrangement is present in more than 70% of new ALL diagnoses in infants under one year.
Updates in KMT2A Gene Rearrangement in Pediatric Acute Lymphoblastic Leukemia · Gorecki M, Koziol I, Kopystecka A, Budzynska J, Zawitkowska J, Lejman M · Biomedicines 2023What this evidence cannot show
A narrative review, not a primary cohort. The 30 x 10^9/L hyperleukocytosis threshold is a conventional definition rather than a measured median, and the review reports no distribution of presenting counts. The same review notes that older children with KMT2A-rearranged disease tend to present with counts below 30 x 10^9/L, so the association is age-conditional rather than intrinsic to the lesion.
Evidence that argues against this (1)
- Hyperleukocytosis is characteristic but far from universal: a substantial minority of infants present with a normal or low white cell count, and in the Interfant-06 multivariable model white cell count entered only after KMT2A status and age, making it a modifier rather than a defining feature. Pieters R et al. 2019
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Interfant-06 defined its high-risk group as KMT2A-rearranged infants older than six months with a white cell count of 300 x 10^9/L or more or a poor prednisone response, and found KMT2A rearrangement to be the strongest prognostic factor for event-free survival, followed by age, white cell count and prednisone response.
Outcome of Infants Younger Than 1 Year With Acute Lymphoblastic Leukemia Treated With the Interfant-06 Protocol: Results From an International Phase III Randomized Study · Pieters R, De Lorenzo P, Ancliffe P, et al. · Journal of Clinical Oncology 2019What this evidence cannot show
651 infants across 18 study groups, so the cohort is large but heterogeneous in supportive care. The paper reports outcome by risk group rather than the distribution of presenting blood counts, so 300 x 10^9/L is a protocol cut-off, not a descriptive statistic. Six-year event-free survival was 46.1% overall and differed by about 10-12 percentage points between West European/North American groups and the rest, so threshold-based stratification interacts with care setting.
Evidence that argues against this (1)
- A white cell count of 300 x 10^9/L is an administrative cut-point chosen to define a trial stratum, not a biological threshold, and dichotomising a continuous predictor in this way discards information and makes the apparent effect depend on where the line was drawn. The count is also a measure of tumour burden and proliferation rate rather than a property of the fusion, which is why it predicts outcome without being a target. Statistics in Medicine 2006
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
In Interfant-99, risk was stratified on the peripheral blood response to a seven-day prednisone prophase, and a very high white blood cell count was independently associated with inferior outcome alongside KMT2A rearrangement, age under six months and poor prednisone response.
A treatment protocol for infants younger than 1 year with acute lymphoblastic leukaemia (Interfant-99): an observational study and a multicentre randomised trial · Pieters R, Schrappe M, De Lorenzo P, et al. · Lancet 2007What this evidence cannot show
482 infants treated on the hybrid regimen with a median follow-up of 38 months. The prognostic analysis is observational within a single protocol, and the prednisone-prophase read-out is protocol-specific and no longer used everywhere. 'Very high' white cell count is a categorical variable in the model, not a continuous measurement.
Evidence that argues against this (1)
- Prednisone prophase response is itself a treatment effect measured in vivo, so it confounds disease biology with drug handling. Infants under six months have immature hepatic metabolism and different volume of distribution, which means a "poor prednisone response" in the youngest patients is in part a statement about exposure at the dose given. That matters here because age under six months and poor response are both in the same risk definition. Statistics in Medicine 2006
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Blasts expressing the KMT2A-associated NG2 antigen were significantly more likely to come from infants and to be associated with hyperleukocytosis and central nervous system involvement, to be CD10-negative, and to express the myeloid-associated antigens CD15 and CD65.
Human homologue of the rat chondroitin sulfate proteoglycan, NG2, detected by monoclonal antibody 7.1, identifies childhood acute lymphoblastic leukemias with t(4;11)(q21;q23) or t(11;19)(q23;p13) and MLL gene rearrangements · Behm FG, Smith FO, Raimondi SC, Pui CH, Bernstein ID · Blood 1996What this evidence cannot show
104 consecutive children were prospectively screened but only 9 were antigen-positive, so the clinical associations rest on single-digit numbers from one institution. Follow-up was explicitly described as short. The paper predates KMT2A nomenclature and used Southern blot as the molecular reference.
Evidence that argues against this (1)
- This association is close to circular. NG2/CSPG4 is expressed in about 90% of KMT2A-rearranged cases, so in a cohort containing both rearranged and germline patients, "NG2-positive blasts come from infants and associate with hyperleukocytosis, CNS disease and CD10-negativity" largely re-reports KMT2A status under a different name. Showing that NG2 adds anything requires the comparison within KMT2A-rearranged patients. Oncofusion knowledge graph (internal analysis) 2026
- E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
A peripheral blood film initially reported as normal can contain blasts that only expert haematopathology review identifies, and corticosteroids given for a presumed benign cause before that review delay the leukaemia diagnosis and may worsen outcome.
Teetering on a liver's edge: a case report highlighting clinical decision-making in thrombocytopenia · Yan A, Erdman L, Sung L, Bernstein S · BMC Cancer 2019What this evidence cannot show
A single case report in a 10-year-old, not an infant, and the failure mode (film called normal, blasts found on re-review) is anecdotal rather than quantified. The authors note that a literature search found no primary data assessing the markers (LDH, urate) they used to justify re-evaluation, so the generalisable claim is about process, not a validated triage rule.
Evidence that argues against this (1)
- The American Society of Hematology stopped recommending routine bone marrow aspiration in suspected immune thrombocytopenia in 2011, which is evidence that in the large majority of such presentations the film and count are in fact sufficient to avoid an invasive test. The case above is the tail of that distribution, not its centre. source 2019
step 1 of the diagnostic path
step 1 to step 2: the film raises the suspicion, flow assigns lineage
first diagnostic step in the infant leukaemia workup
- related to →MRD by genomic breakpoint: the only assay in this disease that is fusion-specific by construction
the diagnostic marrow is the only chance to establish the MRD baseline
the blood count, not the presenting complaint, made the diagnosis