Reactive oxygen species (ROS) in cancer: from redox signaling and metabolic plasticity to therapeutic vulnerabilities.

Reactive oxygen species (ROS) are important regulators of cancer biology, acting as tumor-promoting signaling mediators and inducers of oxidative cell death. Oncogenic signaling, mitochondrial dysfunction, metabolic rewiring, and microenvironmental stress lead to increased basal ROS levels in cancer cells, resulting in a state of chronic oxidative pressure. Tumors develop adaptive antioxidant programs such as glutathione and thioredoxin, NADPH regeneration pathways, and sustained activation of the Nrf2-Keap1 axis to adapt to these conditions, leading to redox plasticity and "Nrf2 addiction" in some cancers. This adaptive rewiring allows malignant cells to sustain proliferative signaling while evading ROS-induced cytotoxicity and contributes substantially to therapeutic resistance. Despite the great promise of ROS-targeted therapies in preclinical studies, their translation into the clinic has been challenging for decades. Large antioxidant trials failed or even increased cancer risk. Many pro-oxidant therapies have limited efficacy due to a narrow therapeutic window, systemic toxicity, poor tumor selectivity, and a dynamic ability of tumors to reprogram antioxidant defenses. The significant intra-tumoral and spatial heterogeneity of redox status further complicates these constraints, where different tumor regions and cellular subpopulations exhibit different metabolic states, ROS thresholds, and sensitivities to ferroptosis. Emerging evidence indicates that ferroptosis, an iron-dependent cell death triggered by lipid peroxidation, is a significant therapeutic liability of redox-adapted tumors, particularly when antioxidant buffering systems like GPX4, system Xc