TP53 is the most frequently altered tumor-suppressor gene in human cancer, yet efforts to therapeutically target p53 have yielded limited and inconsistent clinical success. We argue that this gap reflects not a lack of druggable biology, but an oversimplified conceptual framework that treats p53 as a binary wild-type versus mutant entity. Here, we synthesize emerging evidence supporting a model in which p53 operates across a spectrum of functional states defined by mutation class, allelic burden, isoform composition, aggregation propensity, post-translational regulation, and cellular context. These states shape distinct biological outputs, including transcriptional activity, dominant-negative and gain-of-function effects, immune modulation, and checkpoint dependency, which collectively determine therapeutic vulnerability. We review current strategies targeting the p53 pathway, including mutant p53 reactivation, targeted degradation, anti-aggregation approaches, immune-directed therapies, restoration of wild-type pathway activity, gene replacement, and synthetic lethal targeting of DNA damage response dependencies. Clinical and preclinical evidence highlights key limitations of each approach, including stoichiometric constraints, mutation specificity, context-dependent efficacy, and adaptive resistance. Notably, emerging evidence from preclinical and correlative clinical studies suggests that therapeutic outcomes may be more closely associated with p53 functional state than with TP53 mutation status alone. We further emphasize the emerging roles of p53 isoforms and the tumor immune microenvironment as critical modifiers of p53 activity and determinants of treatment response. Collectively, these insights support a paradigm shift toward mechanism-matched, biomarker-stratified strategies that align therapeutic modality with the operative p53 network. Future progress will depend on integrating multi-parameter diagnostics with rational combination therapies to fully exploit p53 as a central vulnerability in cancer.
Saunders et al. (Sat,) studied this question.