Relapse of acute myeloid leukemia (AML) after allogeneic hematopoietic stem cell transplantation (allo-HSCT) remains one of the main causes of reduced long-term survival. Modern methods for predicting the risk of AML relapse after allo-HSCT take into account the data on the pre-transplant level of minimal residual disease (MRD) determined by flow cytometry and molecular biological studies of recurrent genetic abnormalities, which are currently widespread in clinical practice. Recent studies of the expression of genes characteristic of leukemic stem cells (LSCs) have shown prognostic significance for children with AML in relation to treatment response and the risk of relapse. The study of LSC persistence in order to predict the risk of recurrence after allo-HSCT in children with AML in addition to standard MRD detection methods may be of great importance. The aim of the work was to evaluate the impact of MRD status, both using classic methods and taking into account the genes characteristic of LSC, on the results of allo-HSCT in children with AML. The study was approved by the Independent Ethics Committee and the Scientific Council of the I.P. Pavlov First Saint Petersburg State Medical University of Ministry of Healthcare of the Russian Federation. To assess MRD using standard diagnostic methods, we analyzed the data of 95 children with AML in their 1st–2nd remission (cohort 1). MRD status was negative in 67 (70.6 %) patients; in 28 (29.4 %) children, MRD status was positive according to molecular genetic studies and / or immunophenotyping results. For pre-transplant evaluation of the expression of genes characteristic of LSC, we investigated bone marrow samples of 50 patients (cohort 2) using real-time polymerase chain reaction. The DNMT3B, GPR56, CD34, SOCS2, SPINK2, FAM30A, and ABL genes were studied by real-time polymerase chain reaction, followed by calculation of the pLSC6 value using the formula: DNMT3b × 0.189 + GPR56 × 0.054 + CD34 × 0.0171 + SOCS2 × 0.141 + SPINK2 × 0.109 + FAM30A × 0.0516. At the time of allo-HSCT, 37 (74 %) children with AML were in their 1st or 2nd remission of the disease, 13 (26 %) were out of the 1st–2nd remission. With a median follow-up of 5 years in the group of patients with a positive MRD status, determined by standard methods (cohort 1), overall survival (OS) was 67.9 % vs 73.1 % for patients with a negative MRD status (p = 0.83). The cumulative incidence of relapse was 50 % and 22 %, respectively; p = 0.012. When assessing the level of expression of genes characteristic of LSC (cohort 2), a pLSC6 level was above the median in 18/37 (49 %) patients. The results of linear regression showed that the pre-transplant level of expression of genes characteristic of LSC was not associated with the number of blasts/MRD (odds ratio 1.002; 95 % confidence interval 0.979–1.025). One-year OS rates did not differ significantly in children in the 1st–2nd remission of AML, depending on pLSC6 level: 84.2% in patients with low pLSC6 and 72.2 % – with high pLSC6 (p = 0.4), event-free survival in the corresponding groups – 68.4 % and 61.1 %, respectively (p = 0.34). The cumulative incidence of early relapse after allo-HSCT in the group of AML patients with a high pLSC6 score was significantly higher than in children with a low pLSC6 score before allo-HSCT (22 % and 0 %, respectively; p = 0.03). MRD does not have a statistically significant effect on OS. However, MRD positivity before allo-HSCT increases cumulative incidence of relapse. The level of expression of genes characteristic of LSC, determined before allo-HSCT, showed a prognostic significance in relation to the development of early AML relapse after allo-HSCT.