EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Effectiveness of radiotherapy or chemoradiotherapy in older patients with advanced-stage, non-surgically treated head and neck cancer: A machine learning analysis

Rasheed Omobolaji Alabi iD, Alhadi Almangush, Mohammed Elmusrati, Ylva Tiblom Ehrsson, Göran Laurell, Antti A Mäkitie

Medical Oncology · 2026

DOI: 10.1007/s12032-026-03420-5 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Medical Oncology, 2026. Verfasst von Rasheed Omobolaji Alabi und 5 weiteren Autor:innen. Beteiligte Einrichtung: Research Program in Systems Oncology, Faculty of Medicine, University of Helsinki, Helsinki, Finland. rasheed.alabi@helsinki.fi. Hauptthemen laut MEDLINE: Machine Learning, Chemoradiotherapy, Head and Neck Neoplasms, Squamous Cell Carcinoma of Head and Neck. Die Arbeit nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Helsinki.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rising incidence of head and neck squamous cell carcinoma (HNSCC) is linked to both the HPV epidemic and the aging population. As a result, there is an emerging need to develop management strategies for older patients with advanced-stage HNSCC. This study aims to apply machine learning (ML) techniques to compare the effectiveness of definitive concurrent chemoradiotherapy (CRT) and radiotherapy (RT) alone in older patients (≥ 65 years) with stage III-IV non-surgically treated HNSCC. We used Kaplan-Meier (KM) survival curves, Cox proportional hazards models, and ML-based permutation feature importance (PFI) to compare the effectiveness of CRT versus single modality radiotherapy (RT) for disease-specific survival (DSS) in this population of patients with HNSCC. Finally, we used propensity-score matched (PSM) analysis to clarify if there is a significant difference between patients that received either CRT or RT alone. A total of 1525 patients with HNSCC were obtained from the Surveillance, Epidemiology, and End Results (SEER) Program to build the ML model. Furthermore, the PFI analysis evaluated the effectiveness of CRT or RT alone regarding DSS of older (≥ 65 year) patients with HNSCC. Then, we performed a 1:1 propensity-score matched (PSM) analysis to reduce treatment selection bias and to analyze the prognostic role of CRT. Our ML model showed performance accuracy of 55.5% in predicting DSS in this group with HNSCC. The aggregate feature importance showed that treatment option (CRT or RT alone) is among the four most important features for enhancing DSS among these patients. Specifically, our KM curves, Cox proportional hazards model, and PFI analyses all showed that CRT seems more feasible and effective than RT alone for older patients with stage III-IV HNSCC who did not receive surgery. In addition, patients with stage III disease had better survival rates than those with stage IV. The PSM confirmed that there are substantial differences in nearly all covariates between patients who received CRT or RT alone. The results from the SEER database provide insight into a potential survival benefit for CRT in older patients with stage III-IV HNSCC who are not candidates for surgery. This further emphasizes that treatment decisions at an advanced stage may be made according to other patient-related factors rather than chronological (calendar) age alone.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningChemoradiotherapyHead and Neck NeoplasmsSquamous Cell Carcinoma of Head and NeckHumansFemaleAgedMaleSEER ProgramNeoplasm StagingAged, 80 and overTreatment OutcomeKaplan-Meier Estimate

Stichwörter der Autor:innen

ChemoradiotherapyHead and Neck Cancer Squamous Cell Carcinoma (HNSCC)Machine learningOverall survivalPropensity Score Matching

Autor:innen und Institutionen

  • Research Program in Systems Oncology, Faculty of Medicine, University of Helsinki, Helsinki, Finland. rasheed.alabi@helsinki.fi.
  • Cyber-Physical Systems (CPS) Research Team, School of Technology and Innovations, University of Vaasa, Vaasa, Finland. rasheed.alabi@helsinki.fi.
  • Research Program in Systems Oncology, Faculty of Medicine, University of Helsinki, Helsinki, Finland.
  • Institute of Biomedicine, Pathology, University of Turku, Turku, Finland.
  • Department of Pathology, University of Helsinki, Helsinki, Finland.
  • Libyan Authority for Scientific Research, Tripoli, Libya.

ORCID-Profile: Rasheed Omobolaji Alabi

Förderung

  • University of Helsinki

Zitieren

APA

Alabi, R. O., Almangush, A., Elmusrati, M., Tiblom Ehrsson, Y., Laurell, G., & Mäkitie, A. A. (2026). Effectiveness of radiotherapy or chemoradiotherapy in older patients with advanced-stage, non-surgically treated head and neck cancer: A machine learning analysis. Medical Oncology. https://doi.org/10.1007/s12032-026-03420-5

Vancouver

Alabi RO, Almangush A, Elmusrati M, Tiblom Ehrsson Y, Laurell G, Mäkitie AA. Effectiveness of radiotherapy or chemoradiotherapy in older patients with advanced-stage, non-surgically treated head and neck cancer: A machine learning analysis. Medical Oncology. 2026. doi:10.1007/s12032-026-03420-5

BibTeX
@article{Alabi2026,
  author = {Alabi, Rasheed Omobolaji and Almangush, Alhadi and Elmusrati, Mohammed and Tiblom Ehrsson, Ylva and Laurell, Göran and Mäkitie, Antti A},
  title = {Effectiveness of radiotherapy or chemoradiotherapy in older patients with advanced-stage, non-surgically treated head and neck cancer: A machine learning analysis},
  journal = {Medical Oncology},
  year = {2026},
  doi = {10.1007/s12032-026-03420-5}
}
Kostenlos starten →