Cancer in transition: discovery of tumor-intrinsic transcriptional programs shaping the immune and microenvironmental landscape

Abstract

Cancer biomarker discovery has traditionally focused on individual molecular features; however, tumor behavior and therapeutic response are governed by integrated transcriptional and epigenetic programs that shape immune and microenvironmental states. Deciphering the determinants of metastatic evolution and the complexity of cancer ecosystems is imperative for designing novel preventive and targeted therapies. Tumor complexity, driven by intrinsic cellular heterogeneity and the dynamic plasticity of cancer cells in response to microenvironmental cues, complicates therapeutic strategies based solely on defined molecular or genetic traits. Consequently, there is an urgent need for reliable predictive biomarkers that reflect cancer vulnerabilities, indicate disease progression, or predict patient-specific therapeutic responses to enable truly individualized treatment strategies. Current biomarkers encompass genetic and epigenetic alterations, non-coding RNAs, epithelial-mesenchymal transition-, stemness-, and metastasis-associated transcription factors, as well as cellular components of the tumor microenvironment, including immune cell subsets and cancer-associated fibroblasts. In immuno-oncology, additional biomarkers such as tumor mutational burden, mismatch repair deficiency/microsatellite instability-high status, and PD-L1 expression are widely used for patient stratification. Importantly, the reversible nature of epigenetic modifications, aberrant transcription factor activity, and cell-intrinsic signaling alterations, together with dynamic interactions within the tumor microenvironment, profoundly influence cancer behavior and treatment outcomes. This review summarizes recent advances in cancer and immune-related biomarker research. It outlines a regulatory, systems-level framework that integrates tumor-intrinsic gene control programs with multi-omic, cellular, spatial, and AI-enabled biomarkers. This framework aims to capture tumor plasticity more effectively and advance precision oncology.