This is the author manuscript accepted for publication and has undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as

Longitudinal molecular measurable residual disease (MRD) sampling after completion of therapy serves as a refined tool for identification of imminent relapse of acute myeloid leukaemia (AML) among patients in long-term haematological complete remission. Tracking of increasing quantitative polymerase chain reaction MRD before cytomorphological reappearance of blasts may instigate individual management decisions and has paved the way for development of preemptive treatment strategies to substantially delay or perhaps even revert leukaemic regrowth. Traditionally, MRD monitoring is performed using repeated bone marrow aspirations, albeit the current European LeukemiaNet MRD recommendations acknowledge the use of peripheral blood as an alternative source for MRD assessment. Persistent MRD positivity in the bone marrow despite continuous morphological remission is frequent in both core binding factor leukaemias and nucleophosmin 1-mutated AML. In contrast, monthly assessment of MRD in peripheral blood superiorly separates patients with imminent haematological relapse from longterm remitters and may allow pre-emptive therapy of AML relapse.

Summary Data on occurrence, genetic characteristics and prognostic impact of complex and monosomal karyotype (CK/MK) in children with acute myeloid leukaemia (AML) are scarce. We studied CK and MK in a large unselected cohort of childhood AML patients diagnosed and treated according to Nordic Society for Paediatric Haematology and Oncology (NOPHO)‐AML protocols 1993–2015. In total, 800 patients with de novo AML were included. CK was found in 122 (15%) and MK in 41 (5%) patients. CK and MK patients were young (median age 2·1 and 3·3 years, respectively) and frequently had FAB M7 morphology (24% and 22%, respectively). Refractory disease was more common in MK patients (15% vs. 4%) and stem cell transplantation in first complete remission was more frequent (32% vs. 19%) compared with non‐CK/non‐MK patients. CK showed no association with refractory disease but was an independent predictor of an inferior event‐free survival (EFS; hazard ratio [HR] 1·43, P = 0·03) and overall survival (OS; HR 1·48, P = 0·01). MK was associated with a poor EFS (HR 1·57, P = 0·03) but did not show an inferior OS compared to non‐MK patients (HR 1·14, P = 0·62). In a large paediatric cohort, we characterized AML with non‐recurrent abnormal karyotype and unravelled the adverse impact of CK and MK on prognosis.

Background Measurable/minimal residual disease (MRD) monitoring can predict imminent hematological relapse in acute myeloid leukemia (AML). The majority of childhood AML patients do not harbor fusion genes or mutations applicable as MRD markers and overexpression of Wilms tumor gene 1 (WT1) may constitute a useful monitoring target. However, age‐specific reference values in healthy hematopoiesis and standardization of WT1 assessment are prerequisites for clinical utility. Procedure We investigated WT1 expression across age in hematologically healthy controls (n = 109), during suspected infection (n = 90) and bone marrow (BM) regeneration (n = 13). WT1 expression in AML at diagnosis (n = 91) and during follow‐up (n = 30) was compared with age‐specific reference values. Results WT1 expression correlated with age and showed higher levels in both BM and peripheral blood (PB) in children compared with adults (P < 0.001 and P = 0.01). WT1 expression from healthy hematopoiesis was lower in PB compared with BM (WT1BM/WT1PB = 8.6, 95% CI: 5.3–13.7) and not influenced by infection nor BM regeneration. At AML diagnosis, 66% had more than 20‐fold WT1 overexpression in PB or BM (PB 74%; BM 45%). WT1 was quantified in 279 PB samples during follow‐up. All 11 patients with PB sampling within 4 months of disease recurrence displayed WT1 overexpression by a median of 1.9 months (range, 0.7–9.7) before hematological relapse. Conclusions This study defines child‐specific reference values for WT1 expression in healthy hematopoiesis and demonstrates that WT1 expression in PB is a useful post‐treatment monitoring tool in childhood AML. Based on these observations, we propose definitions for childhood AML molecular relapse using WT1 overexpression.

No abstract