Oncofusion

start / graph / drug-plerixafor

Drugmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q33-niche-protein-layer

Plerixafor (AMD3100, Mozobil) - a licensed CXCR4 antagonist that has never been given to an infant with leukaemia, and whose only paediatric efficacy study failed

also called AMD3100 · Mozobil · JM-3100

Plerixafor blocks the receptor leukaemia cells use to stay anchored in the bone marrow, and it is already an approved drug for pushing stem cells into the blood before a transplant. In mice carrying infant MLL-rearranged leukaemia it made a targeted drug work much better, but it has never been tested in an infant with leukaemia and its one paediatric efficacy trial did not work.

The same thing, for a scientist

Plerixafor (CHEMBL18442, small-molecule bicyclam CXCR4 antagonist) is the only agent in this graph's niche layer with a licence. Its FDA indication is narrow and is not oncological treatment: mobilisation of HSCs with filgrastim for autologous transplant in non-Hodgkin lymphoma or multiple myeloma, NDA022311 [E1]. The KMT2A-relevant preclinical result is in the index lineage: in primary infant MLL-rearranged ALL, stroma protects blasts differentially versus non-MLL-rearranged ALL, and AMD3100 markedly enhanced lestaurtinib efficacy in an MLL-R ALL xenograft [E2]. Clinically the picture is thin and should be stated plainly. Two paediatric leukaemia trials of CXCR4 antagonism exist and **both exclude infants**: POETIC NCT01319864, phase 1, completed, n=20, eligibility 3-29 years [E3]; and NCT01655875, phase 1, terminated at n=1 with whyStopped recorded as 'lack of patient enrollment', eligibility 2-22 years [E4]. The label is blunter still: safety and effectiveness have not been established in paediatric patients, and **effectiveness was NOT demonstrated** in the phase 1/2 randomised paediatric mobilisation study of 45 patients aged 1 to under 17 with solid tumours or lymphoma [E1, E5]. Nothing here is an infant leukaemia result.

Go deeper
What the registry and the label say, which the literature does not

Both facts below came from the regulatory record and neither is visible in PubMed - the pattern rule 75 records.

The terminated trial's reason. NCT01655875 stopped with whyStopped = 'lack of patient enrollment', having enrolled 1 patient against a planned pilot 4. That string is absent from the literature and is only returned by the raw ClinicalTrials.gov v2 API, not by the MCP tool. It matters because it says the failure was accrual, not toxicity or futility - so the mechanism is untested rather than tested and rejected, and any new proposal here must state its accrual route before its rationale.

The paediatric efficacy failure. Section 8.4 of the Mozobil label states that safety and effectiveness have not been established in paediatric patients and that effectiveness was not demonstrated in a phase 1/2 randomised open-label comparative study: 45 patients aged 1 to <17 with solid tumours or lymphoma, randomised 2:1, and the day before first apheresis the median peripheral CD34+ count was 35 x 10^6/L in the comparator arm against 15 x 10^6/L in the plerixafor arm - numerically worse in the treated arm 15. No new safety signals were seen. This is a mobilisation endpoint in a different disease, so it does not refute chemosensitisation in leukaemia; it does mean there is no paediatric efficacy precedent of any kind for this drug.

Infants have received plerixafor - but never for leukaemia

Of 190 plerixafor studies returned by the ClinicalTrials.gov v2 API on 2026-09-22, five set a minimum age below 2 years, and every one is mobilisation or transplant conditioning in a non-malignant setting: NCT03547830 (chronic granulomatous disease, 1 month), NCT03019809 (Wiskott-Aldrich syndrome, 1 month), NCT00479115 and NCT03157804 (Fanconi anaemia, 1 year), and NCT01288573 (paediatric cancer mobilisation, 1 year - the study behind the label's negative efficacy statement) 5. So the drug has infant exposure and no infant leukaemia exposure, and the honest summary for the PI is that safety in a one-month-old is not the unknown; efficacy against leukaemia at any paediatric age is.

Facts and figures
chembl IDCHEMBL18442
targetgene-cxcr4
modalitysmall molecule (bicyclam CXCR4 antagonist)
statusFDA approved (NDA022311, Mozobil, Sanofi-Aventis U.S. LLC) for HSC mobilisation with filgrastim in non-Hodgkin lymphoma or multiple myeloma. NOT approved for any leukaemia and NOT established in paediatric patients: label section 8.4 states safety and effectiveness have not been established in children and that effectiveness was not demonstrated in the phase 1/2 paediatric mobilisation study.
chembl max phase4
infant exposureYES to the drug, NO to the indication. Five of 190 registered plerixafor studies go below two years and all are mobilisation or conditioning in non-malignant disease or solid tumours - chronic granulomatous disease and Wiskott-Aldrich syndrome from 1 month, Fanconi anaemia and paediatric cancer mobilisation from 1 year. No infant has received plerixafor for leukaemia in a registered trial, and no CXCR4-antagonist leukaemia protocol has ever had a lower age bound under 2 years.
regulatory noteRule 75 applies and was checked: the decisive facts here are regulatory rather than bibliographic and came from the openFDA label and the raw ClinicalTrials.gov v2 API, not from PubMed. The label's negative paediatric efficacy statement and the terminated trial's 'lack of patient enrollment' string appear nowhere in the literature.
trials
[
  "NCT01319864",
  "NCT01655875",
  "NCT01288573",
  "NCT03547830",
  "NCT03019809",
  "NCT00479115",
  "NCT03157804"
]
paediatric leukaemia trials
[
  {
    "nct": "NCT01319864",
    "name": "POETIC, plerixafor as a chemosensitiser in relapsed paediatric acute leukaemia and MDS",
    "sponsor": "Seattle Children's Hospital",
    "phase": "PHASE1",
    "status": "COMPLETED",
    "enrolment_actual": 20,
    "eligibility": "3 Years to 29 Years",
    "dates": "2011-03 to 2016-06-28",
    "why_stopped": null
  },
  {
    "nct": "NCT01655875",
    "name": "AMD3100 as a sensitising agent before myeloablative allogeneic BMT in chemotherapy-resistant paediatric acute leukaemia",
    "sponsor": "Emory University",
    "phase": "PHASE1",
    "status": "TERMINATED",
    "enrolment_actual": 1,
    "eligibility": "2 Years to 22 Years",
    "dates": "2012-06 to 2014-04",
    "why_stopped": "lack of patient enrollment"
  }
]
key ae
[
  "injection-site reactions",
  "gastrointestinal effects (diarrhoea, nausea)",
  "leukocytosis by mechanism - the drug's purpose is to move cells into blood",
  "in a leukaemia the mechanistic adverse event of concern is blast mobilisation into the circulation, which is not a labelled AE because the licensed population does not have circulating malignant myeloid or lymphoid blasts"
]
resistance
[
  "ACKR3 (formerly CXCR7) is a second CXCL12 receptor that plerixafor does not antagonise; it has 0 mentions in this corpus under either token and has never been measured in KMT2A-rearranged disease",
  "CXCL12-independent CXCR4 signalling, reported in murine MLL-AF9, would be unaffected by blocking ligand-competitive binding at the receptor if the relevant signalling is constitutive"
]
How this could bite us back
  • MOBILISING A LEUKAEMIA IS NOT THE SAME AS MOBILISING A GRAFT. The licensed purpose of this drug is to put haematopoietic cells into the bloodstream. Doing that to a child with leukaemia distributes blasts, and no registered leukaemia protocol has reported extramedullary or CNS outcomes after CXCR4 antagonism. This graph separately holds a CNS-adapted leukaemic cell state and an MLL-AF9 result in which the VCAM1-VLA-4 adhesion axis drives intracerebral invasion.
  • THE TARGET MAINTAINS THE PATIENT'S OWN STEM CELLS. Induced Cxcr4 deletion in adult mice collapses the HSC pool and increases sensitivity to myelotoxic injury. An infant on intensive chemotherapy has minimal marrow reserve, and the mechanism of benefit and the mechanism of harm are the same receptor.
  • THE PAEDIATRIC EFFICACY PRECEDENT IS NEGATIVE. The only randomised paediatric study in the label did not demonstrate effectiveness and the treated arm's median pre-apheresis CD34+ count was numerically lower than the comparator's. That is a different disease and a different endpoint, but it is the only paediatric efficacy evidence that exists for this molecule.
  • THE BYPASS IS UNMEASURED. ACKR3/CXCR7 binds CXCL12 and is not blocked by plerixafor. Neither token appears anywhere in this corpus and no KMT2A-rearranged measurement exists, so the resistance route is not merely plausible - it is unexamined.
  • FEASIBILITY IS A REAL FAILURE MODE HERE. One of the two paediatric protocols terminated having enrolled a single patient for lack of accrual. In a disease this rare, a proposal that does not name its consortium and its accrual route is proposing an experiment that has already failed once for non-scientific reasons.
Still open
  1. Is there any route to an infant KMT2A-rearranged ALL study of CXCR4 antagonism, given that the two existing paediatric protocols enrolled 21 patients between them and one terminated for accrual?
  2. Would a CXCR4 antagonist add anything to a menin inhibitor, or is the axis already suppressed by it? See hyp-cxcr4-function-is-a-hoxa-meis1-output-of-the-fusion.
  3. Has anyone measured ACKR3 on KMT2A-rearranged blasts? A plerixafor-resistant bypass receptor with zero measurements is an unexamined liability for the whole approach.
Every source behind this entry (5)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    Plerixafor (Mozobil, NDA022311, Sanofi-Aventis U.S. LLC) is indicated in combination with filgrastim to mobilise haematopoietic stem cells to peripheral blood for collection and subsequent autologous transplantation in non-Hodgkin lymphoma or multiple myeloma; its label states that safety and effectiveness have not been established in paediatric patients and that effectiveness was not demonstrated in a phase 1/2 randomised paediatric study of 45 patients aged 1 to under 17 with solid tumours or lymphoma.

    What this evidence cannot show

    The US label only; EMA and other regulators are not checked here and a geographic asymmetry of the kind rule 75 documents for treosulfan cannot be excluded. The paediatric study described in section 8.4 is a MOBILISATION study in solid tumours and lymphoma, so its negative efficacy result does not speak to chemosensitisation in leukaemia - it speaks to the absence of any paediatric efficacy precedent. Retrieved 2026-09-22.

    Evidence that argues against this (1)
    • A negative mobilisation result in solid tumours and lymphoma is a poor guide to a different mechanism in a different disease: the preclinical leukaemia rationale is chemosensitisation by disrupting stromal protection, which the paediatric mobilisation endpoint does not measure at all. Sison EA et al. 2013
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In primary infant MLL-rearranged ALL, stroma differentially protects blasts relative to non-MLL-rearranged ALL and protects both bulk blasts and leukaemia stem cells from the FLT3 inhibitor lestaurtinib; in an MLL-R ALL xenograft, CXCR4 inhibition with AMD3100 (plerixafor) markedly enhanced lestaurtinib efficacy.

    PMID 23294096DOI
    What this evidence cannot show

    MLL-rearranged as a class; the abstract does not break the primary samples down by fusion partner, so the contribution of KMT2A::AFF1 is unknown from it (rule 80). The potentiated partner is lestaurtinib, which was not adopted clinically, so the specific combination tested is not a live regimen. Read from the abstract.

    Evidence that argues against this (1)
    • In syngeneic murine MLL-AF9, Cxcr4 deletion eradicated leukaemia WITHOUT impairing marrow homing and with CXCL12 dispensable, the phenotype tracking to oxidative stress and differentiation. If that is the operative mechanism then plerixafor's benefit here is not de-adhesion, and a short antagonist pulse timed around chemotherapy - the mobilisation logic - is the wrong schedule. Ramakrishnan R et al. 2020
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry

    POETIC NCT01319864, 'A Phase I Study Using Plerixafor as a Chemosensitizing Agent for Relapsed Acute Leukemia and MDS in Pediatric Patients', Seattle Children's Hospital, is COMPLETED with 20 actual participants, ran 2011-03 to 2016-06-28, and its eligibility is 3 years to 29 years.

    NCT01319864
    What this evidence cannot show

    Registry record only; no results are read here and no efficacy claim is made from it. The relevant fact for this beat is the lower age bound of 3 years, which places every infant outside it. Fetched from the ClinicalTrials.gov v2 API 2026-09-22; whyStopped is absent because the study completed.

    Evidence that argues against this (1)
    • A completed phase 1 with 20 patients and no posted efficacy signal in this graph is weak support for the approach; absence of a follow-on phase 2 in the decade since is itself information, and this entry does not have a published reason for it. Sison EA et al. 2015
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry

    NCT01655875, a phase 1 pilot of AMD3100 as a sensitising agent before myeloablative allogeneic transplant in chemotherapy-resistant paediatric acute leukaemia (Emory University), was TERMINATED after enrolling 1 actual participant, with whyStopped recorded in the registry as 'lack of patient enrollment', and eligibility 2 years to 22 years.

    NCT01655875
    What this evidence cannot show

    The whyStopped string is only returned by the raw ClinicalTrials.gov v2 API and is absent from the MCP tool's response - this graph has a recorded case of an entry asserting 'the registry gives no reason' when the registry gave it plainly. Fetched 2026-09-22. n=1 means nothing about the drug is learnable from this record.

    Evidence that argues against this (1)
    • An accrual failure is not evidence against the mechanism, but it is evidence about feasibility, and it points the other way from the enthusiasm in the preclinical literature: two paediatric CXCR4-antagonist leukaemia protocols have existed and between them enrolled 21 patients over roughly a decade. ClinicalTrials.gov 2016
  5. E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.trial_registry

    Of 190 plerixafor studies returned by the ClinicalTrials.gov v2 API, five set a minimum age below two years - NCT03547830 (chronic granulomatous disease, 1 month), NCT03019809 (Wiskott-Aldrich syndrome, 1 month), NCT00479115 (Fanconi anaemia, 1 year), NCT03157804 (Fanconi anaemia gene therapy, 1 year) and NCT01288573 (paediatric cancer mobilisation, 1 year) - and all five are mobilisation or transplant-conditioning protocols rather than leukaemia treatment.

    NCT01288573
    What this evidence cannot show

    The 190 is what the API returned for query.intr=plerixafor with pageSize=200 on 2026-09-22; a registry query is only as complete as its indexing, and studies naming only AMD3100 or Mozobil may be missed by an intervention-name search, which would make five a floor rather than a ceiling. The age screen was a pattern match on minimumAge for Month, Week, Day or '1 Year', so a study with no minimumAge recorded would be missed.

    Evidence that argues against this (1)
    • Two of the 190 records in the same listing carry no minimumAge at all (NCT01362985, an EBMT registry analysis, and NCT01068301, a paediatric second-transplant study), so the five-study count is a floor and the claim is written as one. ClinicalTrials.gov 2026
Connected to