Abstract / Summary
Osteoradionecrosis of the jaws (ORNJ) remains a formidable late complication of head and neck radiotherapy (RT), despite advances in intensity-modulated RT, proton therapy, and contemporary dose-optimization paradigms. Prevailing dosimetric methodologies typically conceptualize the mandible or jaw as a singular organ at risk, employing whole-jaw summary metrics—such as mean dose, maximum dose, and dose–volume parameters—that, while establishing the dose dependence of ORNJ, may inadequately reflect the spatial complexity of radiation-induced injury, particularly in markedly asymmetric dose distributions or localized disease. This critical review synthesizes anatomical, developmental, neurovascular, radiobiological, dosimetric, and clinical evidence to advance the field beyond uniform whole-jaw assessment and advocates for a conditional, multiscale dosimetric approach. We delineate an FDI-based framework that proposes candidate dosimetric compartments for comparative evaluation according to lesion distribution and clinical endpoints: hemijaw dose–volume metrics for well-lateralized ORNJ; anterior-midline compartment dosimetry for anterior midline-related disease; multicompartment evaluation for extensive or bilateral disease; and integrated socket-/implant-site-specific and regional dose–volume analysis for extraction-related disease. The framework further proposes parallel reporting of physical-dose and EQD2-calibrated parameters to enhance comparability across conventional, hypofractionated, stereotactic, proton, and reirradiation regimens. While current clinical evidence supports the biological plausibility of regional and site-specific assessment, definitive superiority over whole-jaw dosimetry remains unproven. Thus, the proposed framework is presented as a structured, empirically testable model requiring validation in retrospective and prospective cohorts. If validated, this paradigm may refine ORNJ risk stratification, support dental decision-making, personalize survivorship care, and guide biologically nuanced jaw-sparing RT planning.