Emerging Biomarkers, AI-Driven Liquid Biopsy, and Multi-Modal Data Integration for Early Cancer Detection

Early cancer detection remains one of the most important challenges in oncology, with major implications for patient outcomes, treatment selection, and disease monitoring. Recent advances in biomarker discovery, liquid biopsy technologies, artificial intelligence, and integrative data science are reshaping how malignancies may be detected at their earliest and most treatable stages. At the same time, important scientific and translational questions remain regarding analytical validity, biological interpretability, clinical utility, and implementation across diverse cancer types and patient populations.This Research Topic will highlight emerging biomarkers and computational strategies that support early cancer detection through minimally invasive and multi-modal approaches. Of particular interest are studies investigating circulating tumor DNA, cell-free nucleic acids, methylation signatures, circulating tumor cells, extracellular vesicles, proteins, metabolites, immune-related markers, and other blood- or body fluid-based indicators of early malignant transformation. We also welcome work on AI-driven liquid biopsy analysis, including machine learning and deep learning methods for signal detection, feature extraction, classification, risk stratification, tissue-of-origin prediction, and integration of weak or complex biomarker signals.A central focus of this Topic is multi-modal data integration. Combining liquid biopsy data with imaging, digital pathology, genomic, transcriptomic, epigenomic, proteomic, metabolomic, and clinical data has the potential to improve sensitivity, specificity, and robustness in early detection models. We are particularly interested in contributions that address how heterogeneous data types can be harmonized and translated into clinically meaningful insights, as well as studies exploring model interpretability, reproducibility, validation, and generalizability across cohorts and settings.This Research Topic also seeks to capture challenges that must be addressed for real-world impact, including pre-analytical and analytical standardization, biomarker qualification, low-signal detection in early-stage disease, false positive reduction, bias in AI systems, and performance in underrepresented populations. By bringing together advances in biomarker science, computational oncology, and translational cancer research, this collection aims to provide a focused overview of emerging frameworks for earlier, more precise, and more scalable cancer detection.Please note: Manuscripts consisting solely of bioinformatics, computational analysis, or predictions of public databases which are not accompanied by validation (independent clinical or patient cohort, or biological validation in vitro or in vivo, which are not based on public databases) are not suitable for publication in this journal. Topic editors: Sonam Mittal, Saima Shakil Malik, Ramakrishna Sompallae Source: https://www.frontiersin.org/research-topics/85470/emerging-biomarkers-ai-driven-liquid-biopsy-and-multi-modal-data-integration-for-early-cancer-detection