POE 10-03 / NCT01319864 — the only completed paediatric trial of niche disruption, which mobilised blasts 3.4-fold and produced three responses in nineteen children, none of them in ALL
also called NCT01319864 · POE 10-03 · plerixafor chemosensitisation in paediatric relapsed acute leukaemia
This was the one completed children's trial that tried to treat leukaemia by knocking cells out of their bone-marrow hiding place before giving chemotherapy. The drug did exactly what it was supposed to do biologically — it pushed leukaemia cells out into the bloodstream — but only three of nineteen heavily pretreated children responded, and none of them had lymphoblastic leukaemia.
The same thing, for a scientist
A phase 1 study of the CXCR4 antagonist plerixafor with high-dose cytarabine and etoposide in children and young adults with relapsed or refractory ALL, AML and MDS, run by the Pediatric Oncology Experimental Therapeutics Investigators' Consortium (POETIC) and sponsored by Seattle Children's Hospital with Children's Healthcare of Atlanta. Plerixafor was given daily for five days at four dose levels (6, 9, 12, 15 mg/m2) followed four hours later by high-dose cytarabine twelve-hourly and daily etoposide. **Nineteen patients were treated** (13 AML, 5 ALL, 1 MDS); the registry records 20 enrolled. **No dose-limiting toxicities**; the commonest grade 3 or higher non-haematological toxicities attributable to plerixafor were febrile neutropenia and hypokalaemia. Exposure increased with dose and clearance was stable between day 1 and day 5. Eighteen were evaluable: **two complete remissions and one CRi, all three in AML; no responses in ALL or MDS**. **Plerixafor mobilized leukaemic blasts into the peripheral blood in 14 of 16 evaluable patients, median 3.4-fold, and the degree of mobilization correlated with surface CXCR4 expression** [E1]. The registry adds a fact absent from the paper that matters for this graph: the **minimum age was 3 years**, so infants - the population in which KMT2A rearrangement and the stromal-protection data actually live - were excluded by protocol [E2].
Target engagement without clinical benefit is the informative result
This trial is unusually interpretable for a phase 1 because it measured the mechanism directly. Plerixafor worked: blasts were mobilized in 14 of 16 evaluable patients, at a median 3.4-fold, and the magnitude tracked surface CXCR4 expression, which is what a target-engagement biomarker is supposed to do. The drug was safe, with no dose-limiting toxicities across four dose levels.
And the clinical result was modest: three responses in eighteen evaluable, heavily pretreated patients, all in AML.
That combination is the single most useful piece of clinical evidence in this beat, because it separates two things the preclinical literature runs together. Mobilising cells out of the niche is achievable. Achieving it is not the same as achieving benefit. Any future proposal in this project to disrupt the marrow niche has to explain what it would do differently from a trial that demonstrably hit its target and still did not move the endpoint.
The candidate explanations are worth listing because they are testable: the population was heavily pretreated and refractory; the chemotherapy partner (cytarabine/etoposide) may not have been the right one; the schedule gave plerixafor four hours before chemotherapy, which assumes a mobilisation window nobody has validated; or the retention model is wrong and CXCR4's role in these cells is cell-intrinsic signalling rather than positional (mechanism-cxcl12-cxcr4-marrow-retention).
The registry fact the paper does not carry: infants were excluded
The published report describes the population as 'children and young adults'. The registry record gives the eligibility as 3 years to 29 years 2.
That is decisive for this graph. The preclinical case for CXCR4 antagonism in this disease rests on primary infant KMT2A-rearranged ALL samples, where stroma protects differentially relative to non-rearranged ALL and AMD3100 reverses it. The trial that tested the idea in children could not enrol an infant. Its five ALL patients were aged at least three and are not described by genotype in the abstract, so whether any of them carried a KMT2A rearrangement is unknown from the sources read here.
So the honest statement is not 'CXCR4 antagonism failed in KMT2A-rearranged leukaemia'. It is that the hypothesis has never been tested in the population it was generated from. The registry sweep in ana-cxcr4-directed-trials-kmt2a-infant confirms that no CXCR4-directed trial in KMT2A-rearranged or infant leukaemia exists at all.
| nct | NCT01319864 |
|---|---|
| phase | Phase 1 |
| status | COMPLETED |
| start date | 2011-03 |
| completion date | 2016-06-28 |
| n | 20 |
| n treated per publication | 19 |
| sponsor | Seattle Children's Hospital |
| significance | the only completed paediatric trial of niche disruption; demonstrates target engagement without proportionate clinical benefit, and excluded the infant population from which the preclinical rationale was generated |
| registry retrieved | raw ClinicalTrials.gov v2 API, 2026-09-19 |
age eligibility
{
"minimum": "3 Years",
"maximum": "29 Years",
"note": "infants excluded by protocol"
}collaborators
[ "Children's Healthcare of Atlanta", "Pediatric Oncology Experimental Therapeutics Investigators' Consortium (POETIC)" ]
arms
[ "plerixafor 6, 9, 12 or 15 mg/m2/dose daily x5, each followed 4 h later by high-dose cytarabine q12h and daily etoposide" ]
population — the numbers behind the chart
{
"AML": 13,
"ALL": 5,
"MDS": 1
}readout
{
"dlt": "none",
"common_attributable_grade3plus_nonhaem": [
"febrile neutropenia",
"hypokalaemia"
],
"responses": {
"CR": 2,
"CRi": 1,
"evaluable": 18,
"all_responses_in": "AML",
"responses_in_ALL": 0
},
"pharmacodynamic": "blast mobilization in 14 of 16 evaluable patients, median 3.4-fold, correlating with surface CXCR4 expression"
}- Mobilising blasts into the circulation in a disease prone to hyperleukocytosis is a real hazard, and this trial's population had low disease burden by virtue of being in relapse after treatment; the same manoeuvre at diagnosis would be different.
- Plerixafor mobilises normal haematopoietic stem cells as effectively as blasts, which is its licensed use, so a chemosensitisation strategy also sensitises the compartment needed for count recovery.
- A phase 1 that hits its pharmacodynamic target and misses clinically is the hardest kind of result to build on, because it removes the easy explanations: the drug reached the target, engaged it, and it was not enough.
- Would plerixafor plus chemotherapy have worked in infants with KMT2A-rearranged ALL, the population the preclinical work was done in? No trial has asked, and this one could not.
- Was the four-hour interval between plerixafor and chemotherapy right? The mobilisation kinetics that would justify that schedule have not been established in leukaemic children.
- Mobilization correlated with surface CXCR4. Does surface CXCR4 correlate with KMT2A rearrangement, and would a CXCR4-high selection strategy identify a responsive subgroup?
- The three responders were all AML. Is that a lineage effect, a genotype effect, or the fact that 13 of 19 patients had AML?
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In a phase 1 POETIC study, plerixafor 6-15 mg/m2 daily for 5 days followed 4 h later by high-dose cytarabine and etoposide was given to 19 children and young adults with relapsed or refractory acute leukaemia or MDS (13 AML, 5 ALL, 1 MDS); there were no dose-limiting toxicities, the commonest attributable grade >=3 non-haematological toxicities were febrile neutropenia and hypokalaemia, exposure increased with dose and clearance was similar on days 1 and 5; of 18 evaluable patients two achieved CR and one CRi, all three with AML and none with ALL or MDS; and plerixafor mobilized leukaemic blasts into the peripheral blood in 14 of 16 evaluable patients with a median 3.4-fold increase, the degree of mobilization correlating with surface CXCR4 expression.
A phase 1 study of the CXCR4 antagonist plerixafor in combination with high-dose cytarabine and etoposide in children with relapsed or refractory acute leukemias or myelodysplastic syndrome: A Pediatric Oncology Experimental Therapeutics Investigators' Consortium study (POE 10-03) · Cooper TM, Sison EAR, Baker SD, Li L, Ahmed A, Trippett T, Gore L, Macy ME, et al. · Pediatric Blood and Cancer 2017What this evidence cannot show
Phase 1, n=19, no control arm, heavily pretreated relapsed/refractory population in which any regimen would be expected to perform poorly; the study was not powered for efficacy and the response rate should not be read as one. Genotypes of the five ALL patients are not given in the abstract, so KMT2A status is unknown. Mobilization is a pharmacodynamic surrogate, not a clinical endpoint. Read: abstract only.
Evidence that argues against this (1)
- The preclinical rationale is specifically about infant KMT2A-rearranged ALL, where stroma protects differentially relative to non-rearranged ALL, soluble stromal factors alone suffice, and AMD3100 markedly enhanced drug efficacy in xenograft. This trial excluded infants (minimum age 3 years) and produced no ALL responses, so it is not a test of that rationale. A negative or modest result in a different population does not refute the preclinical claim; it leaves it untested. Read: abstract only. Sison EA et al. 2013
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry
ClinicalTrials.gov record NCT01319864, 'A Phase I Study Using Plerixafor as a Chemosensitizing Agent for Relapsed Acute Leukemia and MDS in Pediatric Patients', is COMPLETED with an actual enrolment of 20, started 2011-03 and completed 2016-06-28, sponsored by Seattle Children's Hospital with Children's Healthcare of Atlanta and the Pediatric Oncology Experimental Therapeutics Investigators' Consortium as collaborators, intervention 'Plerixafor Dose Escalation', with eligibility from 3 years to 29 years and no whyStopped field.
NCT01319864 - A Phase I Study Using Plerixafor as a Chemosensitizing Agent for Relapsed Acute Leukemia and MDS in Pediatric Patients · Seattle Children's Hospital; Children's Healthcare of Atlanta; POETIC consortium · ClinicalTrials.gov 2016What this evidence cannot show
Registry fields are sponsor-reported and can lag or disagree with publication: the registry gives enrolment 20 (actual) while the publication reports 19 treated, a one-patient discrepancy this entry does not resolve. Retrieved from the raw v2 API on 2026-09-19 rather than through an MCP tool, so whyStopped was visible and is absent. Read: full registry record for the fields quoted.
Evidence that argues against this (1)
- A minimum age of 3 years is not by itself proof that infants were excluded for a scientific reason rather than a practical one, and plerixafor has been given to younger children in transplant settings - a separate phase 1 studied its safety and pharmacokinetics in children undergoing a second allogeneic transplant for relapsed or refractory leukaemia. So the age floor reflects this protocol's design rather than an absolute barrier to studying the drug in infants. Read: title and PubMed record only. Not verified beyond the PubMed record retrieved 2026-09-19 2014
tested CXCR4 blockade as chemosensitisation; target engaged (3.4-fold blast mobilization) with modest clinical benefit
the only completed paediatric clinical test of reversing niche-mediated chemoprotection
excluded infants by protocol (minimum age 3 years), so the population generating the rationale was not studied
the only completed paediatric therapeutic record the sweep found
paediatric plerixafor chemosensitisation trial, eligibility 3-29 years
POETIC NCT01319864, phase 1, 3-29 years, n=20, completed