AMLSG 23-14 / Palbo-AL-1 (NCT02310243) — the only trial ever run in MLL-rearranged acute leukemia with a CDK4/6 inhibitor, and it reported nothing
also called Palbo-AL-1 · AMLSG 23-14 · NCT02310243
A German study gave the breast-cancer drug palbociclib to adults whose leukemia carried the KMT2A (MLL) rearrangement — the only trial ever done that matches this exact disease and this exact drug class. It was due to finish in 2019 and has never posted a single result.
The same thing, for a scientist
Phase Ib/IIa, open-label, single-agent palbociclib in adults (>= 18 y, no upper age limit) with acute myeloid or lymphoblastic leukemia carrying an MLL rearrangement, sponsored by the University of Ulm with Pfizer and the NCT Heidelberg. Dose 125 mg once daily, 21 days on, 7 days off, with a 3+3 modified Fibonacci phase Ib and a Simon optimal two-stage phase IIa (21 patients in stage 1, 50 total if 3 or more achieved CR, CRi, PR or an anti-leukemic effect). Primary outcomes: adverse events, maximum tolerated dose, overall response rate. Secondary: relapse-free survival, overall survival, quality of life, and **evaluation of CDK6 target inhibition by palbociclib**. Recorded completion July 2019; registry status 'unknown'; **has_results false**; no publication found in PubMed searches this cycle [E1]. This node exists because the trial's outcome is the single most decisive missing fact about CDK4/6 inhibition in this disease, and because an entry is the only way the graph will remember to go and get it.
Why this trial is the one that matters
Every other trial of this drug class in acute leukemia is either in adults unselected for fusion, or in children unselected for fusion, or in combination with chemotherapy that would mask a single-agent effect. This one is single-agent, and its inclusion criterion is the fusion. Its secondary endpoint even includes direct evaluation of CDK6 target inhibition, which is the pharmacodynamic readout that would say whether the drug engaged the mechanism the pre-clinical case rests on.
Its design also makes a partial answer recoverable. The Simon two-stage structure means that if 21 patients were treated and two or fewer responded, the trial would have stopped at the first stage — and that fact alone, if the Ulm group will confirm it, settles whether single-agent CDK4/6 inhibition does anything here.
What the registry record does and does not tell us
Does: the population, the design, the dose, the endpoints, 24 German sites, a planned enrolment of 50, an intended recruitment window of July 2015 to July 2017, and a last-patient-out of July 2018 with completion recorded as July 2019 1.
Does not: how many patients were actually enrolled, whether stage 1 was completed, any toxicity, any response, or any pharmacodynamic result. The status field reads 'unknown', which on ClinicalTrials.gov means the record has not been updated by the sponsor past its expected completion date.
Inference to avoid. No results is not a negative result. A trial in a subtype forming a minority of adult acute leukemia, needing 21 patients before a decision, opened at 24 sites, is a trial that could easily have failed on accrual. This graph has a standing rule against treating an absence as a finding, and the honest label here is unknown, with a named way to find out.
Who to ask
The sponsor is the University of Ulm; the listed Ulm investigator on the registry record is Peter Paschka. The pre-clinical paper that makes the CDK6 case in MLL-rearranged AML has its senior authors at Ulm and the NCT Heidelberg — Claudia Scholl and Stefan Fröhling — who are the same institutions that sponsored this trial 2. That is almost certainly the chain by which the trial came to exist, and it means the people who know the answer are identifiable and reachable.
A second, cheaper route was not taken this cycle and should be: search ASH and EHA abstracts for 2017 through 2021. Trials that never reach full publication are frequently presented once.
| nct | NCT02310243 |
|---|---|
| phase | Phase Ib/IIa |
| status | unknown (registry); completion recorded July 2019; no results posted |
| sponsor | University of Ulm |
| n | 50 |
| n planned stage 1 | 21 |
| population | adults >= 18 years, no upper age limit, acute leukemia (AML or ALL) with confirmed MLL (KMT2A) rearrangement by WHO 2008 criteria; refractory, relapsed, or newly diagnosed and not eligible for intensive therapy; not immediate candidates for allogeneic HSCT (bridge to transplant allowed) |
| readout | adverse events, maximum tolerated dose, overall response rate; secondary relapse-free and overall survival, quality of life, and evaluation of CDK6 target inhibition |
| sites | 24 |
| country | Germany |
| has results | false |
| why it matters | the only trial ever run in MLL-rearranged acute leukemia with a CDK4/6 inhibitor, single-agent, with a pharmacodynamic CDK6 endpoint; its outcome would settle whether this class does anything in this disease |
| how to find out | contact the Ulm group (registry investigator Peter Paschka; pre-clinical senior authors Scholl and Frohling at Ulm/NCT Heidelberg); search ASH and EHA abstracts 2017-2021 |
| entered graph via | followed from the scan hit gene-ccnd3 |
collaborators
[ "Pfizer", "National Center for Tumor Diseases, Heidelberg" ]
arms
[ "single-agent palbociclib, 125 mg once daily for 21 days then 7 days off (phase Ib starting dose)" ]
- If this trial was clearly negative, then the CCND3 scan hit, the whole CDK4/6 target entry and the biomarker idea built on GFI1-36N are all downstream of a question the field already answered and did not write down. The cost of not asking is a cycle of work on a closed question.
- If it failed on accrual instead, that is its own warning: a trial restricted to a fusion subtype could not fill 50 slots across 24 German sites in two years, and an infant-restricted trial in the same disease would be very much harder. `ana-infant-trial-feasibility-arithmetic` already makes this point in general and this is a concrete instance of it.
- How many patients were enrolled, and did the trial pass the first Simon stage? Two or fewer responses in 21 patients would have stopped it and would be a definitive answer.
- Was CDK6 target inhibition measured, and did palbociclib engage it at the tolerated dose? If the drug did not reach the target, a negative result says nothing about the biology.
- Was any result presented at ASH or EHA between 2017 and 2021 without reaching full publication?
- Were any of the enrolled patients ALL rather than AML, and did lineage matter? The eligibility criteria allow both, and this project's population is lymphoid.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry
NCT02310243 (AMLSG 23-14 / Palbo-AL-1) is a phase Ib/IIa open-label study of single-agent palbociclib in adults with acute leukemia carrying an MLL rearrangement, sponsored by the University of Ulm with Pfizer and the National Center for Tumor Diseases Heidelberg, planning 50 patients on a 3+3 phase Ib and a Simon optimal two-stage phase IIa across 24 German sites, with adverse events, maximum tolerated dose and overall response rate as primary outcomes and evaluation of CDK6 target inhibition among the secondary outcomes; its recorded start is July 2015, its completion July 2019, its status 'unknown', and it has no results posted.
Phase Ib/IIa Study of Palbociclib in MLL-rearranged Acute Leukemias AMLSG 23-14/Palbo-AL-1 · University of Ulm; Pfizer; National Center for Tumor Diseases, Heidelberg · ClinicalTrials.gov 2015What this evidence cannot show
Full registry record read via the ClinicalTrials.gov API on 2026-09-18, including eligibility criteria, outcome measures and site list. The record's has_results flag is false and its status is 'unknown', which means the sponsor has not updated it past the expected completion date; nothing in the record states actual enrolment, toxicity or response. No PubMed search run this cycle found a publication; conference abstracts were not searched. Do not cite this entry as evidence that the trial was negative.
Evidence that argues against this (1)
- Registry records go stale for administrative reasons and a 'status: unknown' flag says more about record maintenance than about science. Reading a scientific conclusion into it would be the same error this graph made when it treated an absent endpoint in menin-inhibitor registries as evidence about endocrine safety. Raetz EA et al. 2023
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The pre-clinical rationale for this trial — that MLL-AF9-driven AML cells are exceptionally reliant on CDK6 but not CDK4, that CDK6 is transcriptionally activated by mutant MLL, and that a small-molecule CDK6 inhibitor phenocopies CDK6 depletion — was published by senior authors based at Ulm University and the National Center for Tumor Diseases and German Cancer Research Center in Heidelberg, the institutions that sponsored the trial.
Requirement for CDK6 in MLL-rearranged acute myeloid leukemia · Placke T, Faber K, Nonami A, Putwain SL, Salih HR, Heidel FH, Kramer A, Root DE, Barbie DA, Krivtsov AV, Armstrong SA, Hahn WC, Huntly BJ, Sykes SM, Milsom MD, Scholl C, Frohling S · Blood 124:13-23 2014What this evidence cannot show
Read as abstract only; the author affiliations were read from the PubMed record. The institutional link between paper and trial is an inference from affiliation and timing, not a stated fact in either source, and it is recorded because it identifies who to ask rather than as a scientific claim.
Evidence that argues against this (1)
- A trial designed by the authors of its own pre-clinical rationale is the situation in which publication bias is strongest, which cuts both ways: an unreported result from such a group is somewhat more likely to be uninformative or negative than a neutral base rate would suggest. University of Ulm 2015
single-agent palbociclib in fusion-selected adults with a CDK6 pharmacodynamic endpoint; no results posted
eligibility was acute leukemia with MLL rearrangement, AML or ALL, adults only
the clinical test of the axis the CCND3 scan hit sits on