Quantitative Imaging Biomarkers in Precision Oncology: Predicting Therapy Response and Toxicity

The field of medical imaging is witnessing a rapid transformation through the integration of artificial intelligence (AI), particularly in the development of quantitative imaging biomarkers for oncology. Recent progress in AI-driven segmentation, deep learning analytics, and radiomics has shifted imaging assessment from subjective, qualitative interpretations to quantitative, reproducible measurements. These advances allow for detailed evaluation of normal tissue, tumors, and body composition directly from clinical imaging modalities such as CT, MRI, and PET/CT. However, variability in imaging protocols, segmentation techniques, and the interpretation of high-dimensional features continues to challenge reproducibility and generalizability in clinical practice. Recent studies underscore the association between imaging biomarkers—like tumor volume, heterogeneity, and body composition metrics—and patient outcomes, including therapy response, progression, and treatment-related toxicity. Despite promising results, robust validation across multi-center datasets and broader standardization remain unmet needs, limiting clinical translation and routine application in precision oncology.This Research Topic aims to gather innovative and interdisciplinary research that advances AI-driven quantitative imaging biomarkers for predicting therapy response and toxicity. The objective is to highlight and promote methodological developments, clinical applications, and comprehensive validation strategies that leverage automated organ quantification, tumor volumetry, and radiomics. Central questions include how quantitative imaging biomarkers can predict therapy outcomes, inform on potential toxicities, and how they may be integrated with other biological, molecular, or laboratory data for enhanced patient stratification and management. The goal is to foster a deeper understanding of how artificial intelligence can transform imaging into a cornerstone of precision oncology by enabling data-driven, individualized patient care.The scope of this Research Topic encompasses original research, reviews, methodological advancements, and case studies focused on AI-driven quantitative imaging and radiomics in the context of oncology and nuclear medicine. The boundaries are set to imaging-derived biomarkers developed and validated using automated segmentation, volumetric measurement, or feature extraction from CT, MRI, PET/CT, or hybrid modalities.To gather further insights in this evolving area of cancer imaging, we welcome articles addressing, but not limited to, the following themes:- AI-based segmentation and volumetric analysis of organs and tumors- Tumor volume and burden as imaging biomarkers for predicting therapy outcomes- Radiomics and machine learning methods for predicting treatment response and toxicity- Imaging-based predictors of organ damage and systemic effects in oncology- Quantitative analysis of body composition and organ volumes in cancer patients- Integration of imaging biomarkers with clinical, molecular, or laboratory data- Multi-center validation and reproducibility studies for quantitative imaging biomarkers- Applications in radiology, nuclear medicine, and theranostics- Standardization and benchmarking of radiomics and AI imaging workflows- Strategies for clinical translation and workflow integration of AI-based imaging Topic editors: Haissam Ragab, Lisa Steinhelfer Source: https://www.frontiersin.org/research-topics/80239/quantitative-imaging-biomarkers-in-precision-oncology-predicting-therapy-response-and-toxicity