Next-Generation Cancer Biomarkers: Integrating Computational Prediction and Experimental Validation

The field of precision oncology is rapidly evolving, driven by the imperative to tailor cancer care to the biological individuality of each patient. Despite notable progress achieved through advances in genomic and proteomic profiling, such as improved cancer subtyping and the identification of actionable mutations, crucial obstacles persist. Tumor heterogeneity and adaptive resistance to therapy continue to hinder patient outcomes, challenging the reliability and universal application of known biomarkers. The advent of high-throughput multi-omics technologies has resulted in an unprecedented accumulation of complex biological data, creating both an opportunity and a challenge: how best to extract meaningful biomarkers that reliably inform diagnosis, prognosis, and treatment selection.



Recent studies have leveraged artificial intelligence (AI), machine learning, and sophisticated bioinformatic pipelines to mine large-scale datasets for candidate biomarkers with high predictive potential. These computational approaches have enhanced our ability to detect subtle patterns and rare molecular features linked to disease subtypes or therapeutic response. Nonetheless, many statistically promising biomarkers fail to make the transition from in silico discovery to clinical application due to inadequate experimental validation. This gap slows the integration of novel biomarkers into real-world oncology practice and underscores the necessity for closer collaboration between computational and experimental research.



This Research Topic aims to accelerate the translation of computationally predicted cancer biomarkers into clinically meaningful tools through an integrated multidisciplinary approach. The primary objectives are to spotlight innovative in silico methodologies, including AI-driven models and multi-omics data integration, while simultaneously emphasizing rigorous experimental studies that validate these computational predictions. By fostering collaboration across bioinformatics, molecular biology, and clinical oncology, the topic seeks to identify, validate, and advance biomarkers capable of transforming cancer diagnosis, prognosis, and therapeutic monitoring.



The scope of this Research Topic encompasses the entire pipeline of biomarker discovery—ranging from computational hypothesis generation to preclinical and clinical validation—but excludes studies not directly linked to both prediction and experimental assessment. To gather further insights in this multidisciplinary field, we welcome articles addressing, but not limited to, the following themes:



- Computational mining of public multi-omics datasets for novel cancer biomarkers

- Application of AI and machine learning models for biomarker identification, patient stratification, and therapy response prediction

- Integration of in silico results with experimental design for biomarker validation

- Innovative wet-lab approaches (including in vitro, in vivo, and patient-derived xenograft models) for the experimental confirmation of computational predictions

- Emerging biomarkers from non-invasive sampling techniques such as liquid biopsies