Abstract
Background
Children with leukemia should have timely access to effective therapies. In pediatric acute myeloid
leukemia (pedAML), the rarity of the disease makes randomized clinical trials difficult due to slow accrual,
which can delay the implementation of promising therapies. In this context, well-characterized historical
control cohorts can enable single-arm confirmatory trials and accelerate drug development. In this
context, well-characterized historical control cohorts can enable single-arm confirmatory studies and
accelerate drug development.
Patients and Methods
The AML-BFM 2012 and 2017 registries, population-based and highly granular, were used to select
patients diagnosed between 2004 and 2022. A total of 500 patients with de novo AML and relevant
genetic aberrations were identified. Among them, n=133 patients had relapsed (n=123) or primary
refractory (n=10) disease. Median age at relapse was 6.3 years (range 0.1–17.9), f/m 1:1. Median duration
of first remission was 0.89 years (range 0.13–8.74).
Second event-free survival (2nd EFS) was defined as time from start of salvage therapy to non-response
(NR) (blasts >5%), second relapse, second malignancy, or death from any cause (including treatmentrelated
mortality or infection). Remission categories were defined as CR: no evidence of blasts,
Neutrophils >1000/μ, platelets>80.000/μl, CRi: no blasts but incomplete hematological regeneration, CRp:
no blasts, low platelets, NR: blasts>5%.
Kaplan-Meier and log-rank test were used for survival analysis.
Results
Genetic subgroup distribution: t(9;11) 29%, t(10;11) 27%, t(4;11)/t(6;11) 11%, other KMT2Ar (e.g., t(11;17),
t(11;19), t(1;11), t(X;11)) 9%, NPM1 16%, NUP98 translocations 8%. Best remission after two salvage cycles
(CR, CRi, CRp, NR) was t(9;11) 21%, 13%, 16%, 50%; t(10;11): 23%, 24%, 0%, 62%; KMT2Ar others (t(11;17);
t(11;19); t(1;11); t(X;11) etc.): 15%, 8%, 15%, 62%; NPM1: 19%, 14%, 5%, 62%; NUP98: 0%, 23%, 8%, 69%.
Treatment periods are 2004–2011 (n=53), 2012–2017 (n=30), post-2017 (n=50) with
3-year 2nd EFS: 33±9%, 31±7%, 41±7% (p=0.79) and 3-year OS: 43±7%, 47±9%, 42±8% (p=0.98).
In particular: 1y-EFS 48.8±4%, 2y-EFS 38.3±4%, 1y-, 2y-, 3y-OS: 59±4%, 47.3±4%, 45.4±4%.
Discussion
This fully characterized R/R pedAML cohort with KMT2Ar, NUP98, or NPM1 mutations represents a valid
historical control population for single-arm trials with menin inhibitors. Despite high-risk features,
survival outcomes were consistent across treatment eras, underscoring the unmet need for more
effective salvage therapies.
Importantly, hematologic response (CR/CRi/CRp) did not correlate with overall survival, consistent with
previous findings. This highlights the prognostic value of persistent blasts after reinduction and supports
early MRD-guided treatment strategies.
Our data provides a regulatory-grade benchmark to support and accelerate prospective trials with
targeted agents in pediatric AML.
