Answers are grounded in this paper and cite their sources.
Towards Individualized Dynamic Adaptive Radiotherapy: The Head and Neck Use-CasePresented virtually 2026-11-16T0830 Head and neck radiotherapy has advanced through successive paradigm shifts: recognition of HPV-driven disease, the move from 3D-conformal treatment to IMRT and then proton therapy, and now high-frequency on-treatment imaging with the MR-Linac. Each shift reduced dose to normal tissue, yet dose prescription itself remains population-based, resting on the incorrect assumption that all tumors, and all tumor sub-volumes, are equally radiosensitive.This talk argues that spatially resolved, longitudinal dose-response data are radiation oncology’s distinctive asset, and examines what it takes to turn imaging into a decision-grade biomarker. Using the BEST framework and context-of-use statements, it separates technical quantification (QIBA-style metrology) from biological qualification, and proposes that useful quantitative imaging biomarkers must be accurate, anticipatory, actionable, and additive. Diffusion-weighted and IVIM MRI illustrate the promise: early on-treatment ADC change predicts mid-therapy complete response, daily DWI is feasible on the MR-Linac, and functional-imaging-guided sub-volume boosting has been tested prospectively. Dynamic contrast-enhanced MRI illustrates the problem: in an 11-site digital reference object challenge and subsequent in vivo comparison, algorithms disagreed even on whether a patient’s Ktrans fell above or below the median, underscoring the need for acquisition and post-processing standardization before multi-site scaling.The talk closes with Dynamics-Adapted Radiotherapy Dose (DARD), a model that uses early on-treatment tumor volume dynamics to estimate the individual dose needed for local control, cross-validated across CBCT, CT-on-rails, and MR-Linac data. A planned phase II/III randomized trial in HPV-positive oropharyngeal cancer (240 patients, 2:1) will test DARD-personalized dosing of 40–84 Gy against standard 70 Gy, with Q-TWiST superiority as the primary endpoint, gated non-inferiority testing of 3-year progression-free survival, Bayesian futility monitoring, and an HPV ctDNA safety stop. Embedded multiparametric MRI and digital-twin modeling aim to deliver voxel-level recurrence-risk prediction.
No takes yet. Share an insight, caveat, or question.
Clifton David Fuller (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: