Radiomics-driven multimodal integration in precision radiotherapy for nasopharyngeal carcinoma— current evidence, clinical barriers, and future directions
Nasopharyngeal carcinoma (NPC), a head and neck malignancy, is highly sensitive to radiation. Accordingly, radiotherapy has long been the primary curative modality for patients with non-metastatic NPC. Although intensity-modulated radiotherapy has significantly improved target conformality and normal tissue protection, local recurrence, distant metastasis, and late toxicity remain important issues in long-term management. To achieve a balance between tumor control and normal tissue protection, the continuous integration of multi-source information throughout the treatment workflow is critical. Radiomics has the potential to quantify spatial heterogeneity that may not be fully captured by conventional visual assessment. In this review, radiomics-driven multimodal integration refers to the task-specific combination of imaging-derived phenotypes with complementary clinical variables, Epstein–Barr virus DNA (EBV-DNA), hematological biomarkers, dose distributions, pathological features, molecular data, longitudinal imaging, or treatment-delivery information for defined radiotherapy decisions. This review focuses on the clinical timeline of precision radiotherapy for NPC and provides an overview of radiomics-driven multimodal integration studies. It discusses the current applications and limitations of this framework in pretreatment risk stratification and treatment selection, target volume and organ-at-risk definition during radiotherapy planning, on-treatment monitoring and adaptive radiotherapy, toxicity prediction and normal-tissue protection, and post-treatment recurrence assessment and re-irradiation decision-making. This synthesis of current evidence, clinical barriers, and methodological considerations may inform future validation and implementation of radiomics-driven multimodal integration in NPC precision radiotherapy.
Authors
- Nianyong Chen (ORCID: https://orcid.org/0000-0003-0825-3048)
- Qitao Gou (ORCID: https://orcid.org/0000-0002-3798-7115)
- Yuxin Xie (ORCID: https://orcid.org/0009-0007-6659-5886)
- Qiheng Gou (ORCID: https://orcid.org/0000-0002-9264-2216)
- Yuhan Ma
Institutions
- Sichuan University (CN)
- West China Hospital of Sichuan University (CN)
Publication Details
- Journal
- Frontiers in Oncology
- Published
- 2026-11-04
- DOI
- https://doi.org/10.3389/fonc.2026.1902266
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00