EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Mutational Signatures and Machine Learning for Risk Stratification of Acute Myeloid Leukaemia Based on Targeted Sequencing Data

Heba Elhaddad iD, Claudia Chiriches iD, Shuvro Prokash Nandi, Patrick van Eijk iD, Amanda Gilkes iD, Katie Watts, Amy Houseman iD, Charlotte S Wilhelm-Benartzi, Oliver Gerhard Ottmann, Simon H Reed iD, Martin Ruthardt iD

Cancers · 12.06.2026 · 18 (12) · 1925

DOI: 10.3390/cancers18121925 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Cancers, online erschienen am 12.06.2026. Verfasst von Heba Elhaddad und 10 weiteren Autor:innen. Beteiligte Einrichtung: Division of Cancer and Genetics, Section of Haematology, School of Medicine, Cardiff University, Cardiff CF14 4XN, UK. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Newton Musharafa Fund, Newton-Mosharafa Fund, Egyptian Central Department of Missions.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/Objectives: To date, no validated scoring system can accurately predict the responses of acute myeloid leukaemia (AML) patients to induction chemotherapy (CTX). Current risk assessment relies on complex cytogenetic and molecular abnormalities and focuses on mutations in genes considered fundamental to leukaemogenesis.

Methods

We performed bioinformatic analysis of targeted sequencing (TS) data from 111 genes in 1552 AML patients, focusing on mutational patterns derived from single-nucleotide variant (SNV) catalogues. The SNV catalogues were analysed using non-negative matrix factorisation (NNMF), a linear dimensionality-reduction approach, to extract risk-defining recursive signatures (RSs) and to distinguish responders from resistant patients following induction CTX. To enable patient-level prediction, we complemented NNMF with a Random Forest (RF) model. Given the class imbalance between responders and resistant cases, model performance was improved by applying the Synthetic Minority Over-sampling Technique (SMOTE) and by incorporating germline variants alongside somatic mutations.

Results

NNMF-derived RSs captured clinically relevant structures in patients’ mutational profiles and clustered patients by treatment response, indicating that the diagnostic targeted sequencing data contain sufficient information for risk stratification and treatment response prediction. At the single-patient level, RF models incorporating balanced data and germline variation improved predictive performance compared with unbalanced somatic-only models.

Conclusions

These findings demonstrate that machine learning applied to targeted sequencing data can extract clinically informative mutational structures and improve risk stratification in AML, supporting its potential integration into precision treatment decision-making.

Schlagwörter

Stichwörter der Autor:innen

Random ForestSMOTE balancingacute myeloid leukaemiabioinformaticsinduction chemotherapymachine learningrisk assessment

Autor:innen und Institutionen

  • Division of Cancer and Genetics, Section of Haematology, School of Medicine, Cardiff University, Cardiff CF14 4XN, UK.
  • Experimental Cancer Medical Centre (ECMC), School of Medicine, Cardiff University, Cardiff CF14 4XN, UK.
  • Clinical Pathology Department, Faculty of Medicine, Mansoura University, Mansoura 35516, Egypt.
  • Division of Cancer and Genetics, School of Medicine, Cardiff University, Cardiff CF14 4XN, UK.
  • Department of Cellular and Molecular Medicine, University of California San Diego, La Jolla, CA 92093, USA.
  • Centre for Trials Research, School of Medicine, Cardiff University, Cardiff CF14 4XN, UK.

ORCID-Profile: Heba Elhaddad, Claudia Chiriches, Patrick van Eijk, Amanda Gilkes, Amy Houseman, Simon H Reed, Martin Ruthardt

Förderung

  • Newton Musharafa Fund
  • Newton-Mosharafa Fund
  • Egyptian Central Department of Missions
  • Cardiff University

Zitieren

APA

Elhaddad, H., Chiriches, C., Nandi, S. P., van Eijk, P., Gilkes, A., Watts, K., Houseman, A., Wilhelm-Benartzi, C. S., Ottmann, O. G., Reed, S. H., & Ruthardt, M. (2026). Mutational Signatures and Machine Learning for Risk Stratification of Acute Myeloid Leukaemia Based on Targeted Sequencing Data. Cancers, 18(12), 1925. https://doi.org/10.3390/cancers18121925

Vancouver

Elhaddad H, Chiriches C, Nandi SP, van Eijk P, Gilkes A, Watts K, et al. Mutational Signatures and Machine Learning for Risk Stratification of Acute Myeloid Leukaemia Based on Targeted Sequencing Data. Cancers. 2026;18(12):1925. doi:10.3390/cancers18121925

BibTeX
@article{Elhaddad2026,
  author = {Elhaddad, Heba and Chiriches, Claudia and Nandi, Shuvro Prokash and van Eijk, Patrick and Gilkes, Amanda and Watts, Katie and Houseman, Amy and Wilhelm-Benartzi, Charlotte S and Ottmann, Oliver Gerhard and Reed, Simon H and Ruthardt, Martin},
  title = {Mutational Signatures and Machine Learning for Risk Stratification of Acute Myeloid Leukaemia Based on Targeted Sequencing Data},
  journal = {Cancers},
  year = {2026},
  volume = {18},
  number = {12},
  pages = {1925},
  doi = {10.3390/cancers18121925}
}
Kostenlos starten →