An ongoing challenge in acute myeloid leukemia (AML) is the reliable identification of patients who may have resistant disease when treated with standard induction chemotherapy, or who may relapse early after remission. In this article titled “Real-world validation study of the LSC17 score for risk prediction in patients with newly diagnosed acute myeloid leukemia”, the authors present the first prospective, multicenter study aiming to evaluate the prognostic value of the leukemic stem cell 17-gene (LSC17) score in patients with newly diagnosed AML.1
The European LeukemiaNet (ELN 2022) remains the current standard clinical guidelines for risk stratification, using cytogenetic and molecular profiling data to broadly define risk categories and guide treatment. However, ELN 2022 does not fully account for resistance mechanisms to chemotherapy and predictors of relapse, which are often driven by leukemic stem cells (LSC).2 In 2016, Ng and colleagues developed the LSC17 score: a panel which includes 17 genes associated with LSC biology,2 specifically with a focus on identifying prognostic biomarkers to predict resistance to standard induction therapy or predict relapse after therapy in patients lacking conventionally defined adverse risk factors.2,3 The LSC17 score has been validated in retrospective cohorts,4,5 with a prognostic impact found in both younger (<60 years) and older (>60 years) patients.4 Patients with a high LSC17 score were shown to have lower rates of complete response, shorter overall survival, and higher rates of positive measurable residual disease (MRD) status.5
In the ALFA-0702 trial, LSC17 score independently predicted outcomes in NPM1-mutated adult patients receiving intensive therapy, and added prognostic value when used in combination with NPM1-MRD.5 LSC-based MRD testing was also shown to predict relapse after allogeneic hematopoietic stem cell transplantation (allo HSCT) earlier than traditional MRD testing.6
In the first prospective cohort utilizing the LSC17 score, 276 AML patients were recruited, of which 190 were treated with curative-intent intensive chemotherapy. Importantly, the cohort included patients of various ages, treatment intensities and mutational backgrounds. Samples from bone marrow aspiration or peripheral blood were obtained at time of suspected diagnosis, and both patients and physicians were blinded to the LSC17 score. Patients with a high LSC17 score had lower complete remission (CR) rate, higher MRD after induction, and shorter overall survival.1 The LSC17 score was the best predictor of MRD over ELN 2022 adverse risk, and remained an independent predictor in multivariable models that included age, white blood cell count, and ELN risk.
Through this work, the LSC17 score was prospectively validated as a clinical tool with benefit in upfront risk assessment, and to guide therapy in patients with AML. Although the ELN classification system has improved prognostication, it does not comprehensively account for the biological complexity of the LSC in AML. The data presented in this study reinforces the concept of LSC stemness as a fundamental driver of treatment response in AML. With conventional risk stratification systems, there are challenges in certain risk categories of patients, specifically in the ELN 2022 intermediate-risk group. In this study, the LSC17 score showed a significant advantage in predicting survival outcomes in patients deemed intermediate risk, being able to better predict relapse in patients with a high LSC17 score. The LSC17 score was also the best predictor for predicting the presence of MRD following induction chemotherapy. Patients in the high LSC17 score group benefited from allo HSCT whereas patients in the low-score group did not, suggesting LSC17 could also help to guide transplant decisions.
Another advantage of the LSC17 score is its feasibility and ease of integration into the real-world clinical setting for patients with AML. Samples were collected without additional procedural burden to the patient. Standardized, reproducible determination of the LSC17 score from diagnostic RNA was performed in a Clinical Laboratory Improvement Amendments certified laboratory using a NanoString-based assay,1 with a rapid turnaround time. These factors demonstrate the therapeutic implications of incorporating the LSC17 score into molecular modeling, alongside MRD and mutational testing.
Figure 1.Integration of the LSC17-gene expression score into acute myeloid leukemia risk stratification helps to refine risk assessment, including risk of relapse and measurable residual disease, and to inform post-remission therapy. AML: acute myeloid leukemia; LSC17: leukemic stem cell 17-gene; WBC: white blood cell count; PS: performance status; allo HSCT: allogeneic hematopoietic stem cell transplantation; ELN: European LeukemiaNet; MRD: measurable residual disease.
The majority of the patients in the study received intensive induction chemotherapy, either with 7+3 or FLAG-Ida. However, the landscape of treatment in AML is evolving, with new treatment options including targeted agents, combinations with venetoclax, and the use of maintenance therapies. The utility of the LSC17 score to predict outcomes in this subset of patients, who may be frailer and receive less intensive regimens is not yet defined.
In addition to its clinical relevance, the LSC17 score reinforces the importance of understanding the biology of the LSC and the need for superior biomarkers in the management of AML. A set of LSC death genes were identified based on the expression of the programmed cell death gene.7 These genes were used to develop a LSCD score that, when high, could predict sensitivity to drugs used in the management of AML.7 Venetoclax + hypomethylating agent combinations have been shown to target LSC,8 and the use of these biomarkers will aid in identifying patients who may benefit from alternative regimens.
In summary, in this landmark study, Murphy and colleagues have shown that the LSC17 score adds independent prognostic value in AML, and testing at diagnosis is feasible in the clinical setting. Looking forward, the incorporation of the LSC17 score in prospective clinical trials will be essential to further refine the utility of this score, and help to identify the selection of targeted therapies and indications for allo HSCT.
Footnotes
- Received November 6, 2025
- Accepted November 21, 2025
Correspondence
Disclosures
No conflicts of interest to disclose.
Contributions
Both authors wrote and edited the editorial.
References
- Murphy T, Zhang B, King I. Real-world validation study of the LSC17 score for risk prediction in patients with newly diagnosed acute myeloid leukemia. Haematologica. 2026; 111(3):843-851. Google Scholar
- Ng SWK, Mitchell A, Kennedy JA. A 17-gene stemness score predicts clinical outcome in acute myeloid leukemia. Nature. 2016; 540(7633):433-437. Google Scholar
- Röllig C, Bornhäuser M, Thiede C. Long-term prognosis of acute myeloid leukemia according to the new genetic risk classification of the European LeukemiaNet recommendations: evaluation of the proposed reporting system. J Clin Oncol. 2011; 29(20):2758-2765. Google Scholar
- Bill M, Nicolet D, Kohlschmidt J. Mutations associated with a 17-gene leukemia stem cell score and the score’s prognostic relevance in the context of the European LeukemiaNet classification of acute myeloid leukemia. Haematologica. 2020; 105(3):721-729. Google Scholar
- Vasseur L, Fenwarth L, Lambert J. LSC17 score complements genetics and measurable residual disease for risk assessment in acute myeloid leukemia. Blood Adv. 2023; 7(15):4024-4035. Google Scholar
- Li SQ, Xu LP, Wang Y. Leukemia stem cells for relapse prediction in AML patients receiving allografts: long-term follow-up of a prospective study. Bone Marrow Transplant. 2025; 60(11):1472-1479. Google Scholar
- Wang F, Li M, Cui N, Fu Q, Li F, Gu Z. Signature of leukemia stem cell death pattern predicts prognosis and therapeutic response of acute myeloid leukemia patients. Sci Rep. 2025; 15(1):31612. Google Scholar
- Stelmach P, Trumpp A. Leukemic stem cells and therapy resistance in acute myeloid leukemia. Haematologica. 2023; 108(2):353-366. Google Scholar
Figures & Tables
Article Information
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.