Oncofusion

start / graph / drug-blinatumomab

Drugmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q6b-therapeutics2

Blinatumomab in infant KMT2A-rearranged ALL — the largest single outcome gain in thirty years

also called Blincyto · AMG 103 · MT103 · anti-CD19/CD3 BiTE

Blinatumomab is an antibody with two grabbing ends: one holds a leukemia cell by its CD19 marker, the other drags a T cell over to kill it. Adding one course of it to standard infant chemotherapy roughly doubled four-year survival, which is the biggest improvement this disease has seen in decades.

The same thing, for a scientist

Blinatumomab is a CD19×CD3 bispecific T-cell engager (BiTE) delivered by continuous infusion. In a 30-patient Dutch-led study, a single postinduction blinatumomab course added to the Interfant-06 backbone in newly diagnosed KMT2A-rearranged infant ALL produced a 4-year DFS of 83.3% versus 44.0% and a 4-year OS of 93.3% versus 60.2% against a selected Interfant-06 historical control, at a median follow-up of 4.2 years. It is the single largest reported outcome improvement in this disease, and it is being confirmed prospectively in Interfant-21 (NCT05327894) and in the COG/NCI infant trial NCT06317662. The result is not randomised, and that is the caveat that matters most.

Go deeper
The numbers, and exactly how much weight they carry1 deeper

Thirty infants with newly diagnosed KMT2A-rearranged ALL received the standard Interfant-06 chemotherapy backbone plus one additional postinduction course of blinatumomab (15 µg/m²/day by continuous infusion over 28 days in the published protocol; EudraCT 2016-004674-17).

At a median follow-up of 4.2 years (range 3.2-6.0) 1:

Blinatumomab (n=30)Interfant-06 selected historical control
4-year disease-free survival83.3%44.0%
4-year overall survival93.3%60.2%
Post-induction infection-related death0%4%

IVIg was given to 19/30 (63%) and G-CSF to 5/30 (17%). Mean steady-state blinatumomab concentration was 706 ± 194 pg/mL/day and median clearance 0.89 L/h/m² (range 0.57-2.66), comparable across paediatric age groups, so infants do not appear to need a different exposure target 1.

What this is not. It is not randomised. The comparator is a selected Interfant-06 cohort, chosen to match eligibility, and selection of a historical control is exactly the step where an effect of this magnitude can be manufactured. Thirty patients gives a 4-year DFS confidence interval that is wide whatever the point estimate. The honest reading is: a very large effect that is very unlikely to be entirely artefact, whose true size is unknown until Interfant-21 reads out.

Immunoglobulin recovery — a practical answer to the obvious objection

CD19 engagement depletes normal B cells as well as leukemic ones, and in an infant whose adaptive immune system is still being built, prolonged hypogammaglobulinaemia is the predictable harm. In 169 IgG measurements across the same 30 infants, IgG fell after blinatumomab and then rose, with practically all available values back above 4 g/L by maintenance — i.e. near-complete recovery before end of treatment 2. One course is evidently survivable immunologically. Whether repeated courses, or blinatumomab given after CD19 CAR-T, behave the same way is unknown.

Why a T-cell engager should work here at all

KMT2A-rearranged infant ALL is the subtype in which chemotherapy intensification has repeatedly failed: Interfant-99 and Interfant-06 both tested adding myeloid-style blocks and neither rescued the KMT2A-rearranged group. An immune effector mechanism is orthogonal to every resistance route that matters for cytotoxics and for chromatin drugs — it does not care about the HOXA9/MEIS1 program, drug efflux, or apoptotic priming. That orthogonality is the reason it works and the reason it is the natural partner for a menin inhibitor rather than a competitor to it.

Two trials now test exactly that pairing: NCT07636564 (SWOG, blinatumomab ± revumenib in KMT2A-translocated B-ALL/ALAL with persistent MRD) and NCT07283640 (MD Anderson, subcutaneous blinatumomab + revumenib in KMT2A-rearranged ALL) 3.

The bite-back that is already visible

CD19-directed pressure in KMT2A-rearranged disease does not only select CD19-negative B cells; it selects myeloid lineage switch, a route that is essentially unique to this genotype because the fusion holds the cell in a lineage-plastic state. That mechanism is documented separately in this graph. Practically: the leukemia that comes back after blinatumomab in a KMT2A-rearranged infant may not be treatable with any CD19-directed agent, and may not even be the same disease on a flow panel. Any trial adding blinatumomab must count AML relapses as blinatumomab failures, not as second cancers.

Facts and figures
targetCD19 on B-lineage blasts, engaged to CD3 on autologous T cells (bispecific T-cell engager)
modalitybispecific T-cell engager (BiTE), continuous IV infusion; subcutaneous formulation in development
statusFDA and EMA approved for B-cell precursor ALL including paediatric use; use in KMT2A-rearranged infant ALL is protocol-based, not a separate approval
eudract2016-004674-17
pediatric eligibletrue
trials
[
  "NCT05327894",
  "NCT06317662",
  "NCT05848687",
  "NCT07636564",
  "NCT07283640"
]
key results
{
  "n": 30,
  "backbone": "Interfant-06",
  "courses": 1,
  "dfs_4y_pct": 83.3,
  "dfs_4y_control_pct": 44,
  "os_4y_pct": 93.3,
  "os_4y_control_pct": 60.2,
  "median_followup_years": 4.2,
  "control_type": "selected historical Interfant-06 cohort (not randomised)"
}
pharmacokinetics
{
  "css_pg_per_ml_per_day": 706,
  "css_sd": 194,
  "clearance_l_per_h_per_m2_median": 0.89,
  "clearance_range": [
    0.57,
    2.66
  ]
}
key ae
[
  "hypogammaglobulinaemia (IVIg in 63%)",
  "cytokine release syndrome",
  "neurotoxicity",
  "neutropenia (G-CSF in 17%)"
]
resistance
[
  "CD19 antigen loss",
  "myeloid lineage switch (genotype-specific to KMT2A-rearranged disease)",
  "T-cell exhaustion / inadequate effector fitness in a heavily pretreated infant"
]
How this could bite us back
  • Lineage switch: CD19-directed pressure in KMT2A-rearranged disease selects myeloid escape, converting a treatable B-ALL into a refractory AML rather than simply failing.
  • Antigen loss: CD19-negative relapse removes not only blinatumomab but every downstream CD19 CAR-T option, so sequencing matters and burning CD19 early has a real cost.
  • Immune-effector dependency: blinatumomab needs functional autologous T cells. In an infant already through induction, T-cell number and fitness are the hidden limiting reagent and are not routinely measured.
  • Essential-function toxicity: normal B-cell aplasia in the first year of life coincides with the period of primary immunisation; whether vaccine responses are permanently blunted has not been studied.
  • Delivery: 28 days of continuous infusion in a neonate or small infant means a central line, a pump and hospital dependence, with all of the line-sepsis risk that carries.
  • Late effects: neurotoxicity from a T-cell engager in a developing brain has no long-term dataset in this age group.
Still open
  1. Does the benefit survive randomisation? Interfant-21 and NCT06317662 are the only honest answers and neither has read out.
  2. Is one course optimal? The trial used one course because of toxicity caution, not because one was shown to be enough. More courses might improve DFS further or might simply select lineage switch harder.
  3. Does blinatumomab before a menin inhibitor, or after, matter? Menin inhibition drives differentiation, which in a KMT2A-rearranged B-ALL may alter CD19 density in either direction. Nobody has measured CD19 surface density under menin inhibition in primary infant ALL.
  4. Do infants who relapse after blinatumomab relapse as AML? If lineage switch is the dominant failure mode, the CD19 arms race is already lost and the field should be building CD33/CD123-inclusive strategies from the start.
Every source behind this entry (4)
  1. E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In 30 infants with KMT2A-rearranged ALL treated on the Interfant-06 backbone plus one postinduction blinatumomab course, 4-year DFS was 83.3% versus 44.0% and 4-year OS 93.3% versus 60.2% against a selected Interfant-06 historical control, at median follow-up 4.2 years, with no postinduction infection-related fatality versus 4% in Interfant-06.

    Sustained Benefit of Blinatumomab in Infants With KMT2A-Rearranged ALL: Long-Term Outcomes, Toxicity, and Pharmacokinetics · Vieira Martins M, de Lorenzo P, Kotecha RS, et al. · Journal of Clinical Oncology 2026
    PMID 41505662DOI
    What this evidence cannot show

    Single-arm, n=30, compared with a SELECTED historical Interfant-06 cohort rather than a randomised control. Confidence intervals for the survival estimates are not given in the abstract and should be read from the paper before being quoted. Historical-control comparisons in infant ALL are particularly fragile because supportive care, MRD-guided decisions and transplant practice all improved over the same period. The effect size is large enough that it is unlikely to be pure artefact, but its magnitude is not established.

    Evidence that argues against this (1)
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    Across 169 IgG measurements in the same 30 infants, blinatumomab caused an IgG reduction with near-complete recovery to normal (>4 g/L) before the end of treatment.

    Immunoglobulin Depletion and Recovery Following Blinatumomab in Infants With KMT2A-Rearranged ALL · Vieira Martins M, de Lorenzo P, Kotecha RS, et al. · Pediatric Blood & Cancer 2026
    PMID 42496646DOI
    What this evidence cannot show

    Observational, n=30, confounded by IVIg supplementation in 63% of patients, which by construction raises measured IgG. 'Recovery' therefore partly reflects replacement rather than endogenous B-cell return. Only one blinatumomab course was given; repeated exposure is untested.

    Evidence that argues against this (1)
    • 'Near-complete recovery to normal' describes a cohort that was being supplemented. Long-term follow-up of the SAME 30 infants (median 4.2 years, range 3.2-6.0) records that intravenous immunoglobulin was administered in 19 of 30 (63%) and G-CSF in 5 (17%). Recovery of serum IgG under immunoglobulin replacement in two-thirds of patients is not evidence of endogenous B-cell recovery, and the IgG trajectory cannot be read as a measure of restored humoral immunity without separating supplemented from unsupplemented patients. Vieira Martins M et al. 2026
  3. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    In the primary report of the same 30-infant study, one 28-day continuous infusion of blinatumomab at 15 µg/m²/day added to Interfant-06 chemotherapy met no clinically relevant toxicity endpoint, 28/30 (93%) were MRD-negative or had low-level MRD after the course, and 2-year DFS was 81.6% (95% CI 60.8-92.0) versus 49.4% (95% CI 42.5-56.0) in Interfant-06 historical controls, with 2-year OS 93.3% (95% CI 75.9-98.3) versus 65.8% (95% CI 58.9-71.8).

    Blinatumomab Added to Chemotherapy in Infant Lymphoblastic Leukemia · van der Sluis IM, de Lorenzo P, Kotecha RS, et al. · New England Journal of Medicine 2023
    PMID 37099340DOI
    What this evidence cannot show

    Single-arm, n=30, median follow-up only 26.3 months (range 3.9-48.2) at this report; ten serious adverse events occurred (fever 4, infection 4, hypertension 1, vomiting 1). The primary endpoint was toxicity, not efficacy, so the survival comparison is a secondary, non-randomised analysis against historical Interfant-06 controls. The control confidence intervals are narrow because that cohort is large, which makes the contrast look sharper than the 30-patient estimate justifies. **Regraded moderate -> established on 2026-09-20** under `concept-the-confidence-grading-rule`: this is the trial's own reported result, which is established as a fact about what the trial found.

    Evidence that argues against this (2)
    • The comparator is a historical, Interfant-06-SELECTED cohort, not a randomised control, and the effect size is large enough that selection deserves explicit scrutiny. At a median 4.2 years the same 30 infants show 4-year DFS 83.3% versus 44.0% and OS 93.3% versus 60.2% against that control, with no post-induction infection-related fatality versus 4% in Interfant-06 — and the authors themselves state the frontline benefit 'awaits further confirmation in ongoing trials'. A control cohort assembled retrospectively to match the treated infants cannot exclude that the 30 differed in unmeasured ways that predict outcome; the randomised test is NCT06317662, which has not reported. Vieira Martins M et al. 2026
    • A two-year DFS endpoint cannot see the escape route specific to CD19-directed pressure. Whole-genome reconstruction of the phylogeny in a child who received anti-CD19 cellular and antibody therapy after multiply relapsed ALL showed that the AML which followed was derived from the INITIAL ALL, carrying distinct driver mutations that were detectable before the lineage switch emerged — extensive prior clonal diversification followed by selection under CD19-directed therapy. Lineage switch is therefore not a rare curiosity but a selectable pre-existing subclone, and an MRD-negative, DFS-based readout of a CD19-directed course is blind to it by design. Coorens THH et al. 2023
  4. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.trial_registry

    Two trials are prospectively testing blinatumomab combined with the menin inhibitor revumenib in KMT2A-rearranged acute lymphoblastic leukemia.

    NCT07636564
    What this evidence cannot show

    Both are registry records for trials that are NOT YET RECRUITING as of 2026-09-18; neither has enrolled a patient or produced data. The MD Anderson study is NCT07283640. Registration is not evidence of feasibility or benefit.

    Evidence that argues against this (1)
    • Neither trial will test this combination in infants, which is where blinatumomab's benefit in KMT2A-r ALL was actually shown. Checked against the raw ClinicalTrials.gov v2 API on 2026-09-19: NCT07636564 (SWOG, phase 2, blinatumomab with or without revumenib in KMT2A-translocation B-ALL/ALAL) is NOT_YET_RECRUITING with a planned start of 2026-10-14, estimated n=90, primary completion 2031-04-16, and a MINIMUM AGE OF 1 YEAR — so infants under twelve months are excluded by eligibility. Its primary endpoints are dose-limiting toxicity, MRD-negativity rate and MRD-event-free survival, not overall survival. 'Being prospectively tested' is true of the combination and false of the population this lab cares about. SWOG Cancer Research Network 2026
Connected to