Biological state-guided oncology: a framework for dynamic precision cancer care through oncological state transitions

Precision oncology has transformed cancer treatment through molecular profiling, but clinical decisions remain largely anchored to baseline assessments that incompletely capture evolving tumor biology and host conditions.

This review proposes the Oncological State Transition (OST) framework as a conceptual approach to cancer as a dynamic process of evolving tumor and host biological states. We examine the biological basis of state transitions and the potential roles of longitudinal liquid biopsy, multi-omics, functional imaging, digital pathology, artificial intelligence, and digital twins in multimodal state assessment. Potential applications across prevention, surgery, systemic therapy, minimal residual disease surveillance, therapeutic resistance, and survivorship are discussed, together with challenges in clinical validation and implementation.

OST provides a testable framework for integrating longitudinal multimodal information to estimate current biological states and potential future transitions. It emphasizes probabilistic state estimation and the distinction between biological and actionable transitions. Rather than replacing conventional staging or molecular classification, OST may complement precision oncology by supporting dynamic state assessment and adaptive decision-making. Prospective studies and randomized comparisons are required to establish its clinical utility. OST therefore represents a potential foundation for Dynamic Precision Oncology, but remains a conceptual framework rather than a validated clinical system.