Biologically Adaptive Dose Compensation Based on Tumor Radiosensitivity Derived from In-Treatment Volume Dynamics
Abstract
Purpose
Tumor regression during radiotherapy reflects underlying radiosensitivity, yet adaptive radiotherapy is largely guided by geometric changes and physical dose metrics. We developed a biologically adaptive radiation therapy (BART) framework that derives tumor-specific radiosensitivity from in-treatment tumor volume dynamics and compensates dose using biological effective dose (BED), and evaluated its TPS-based dosimetric feasibility.
Methods
Longitudinal tumor volumes from weekly cone-beam CT in head-and-neck cancer patients were analyzed. Latent class mixed modeling (LCMM) identified tumor response patterns, including a subgroup requiring dose escalation. In this subgroup, a mechanistic ordinary differential equation (ODE) model incorporating logistic proliferation and linear–quadratic (LQ) cell kill estimated tumor radiosensitivity (α/β) using tumor volume at the 20th fraction. Required BED for biological equivalence was calculated relative to an ideal reference regimen (2 Gy/fraction, α/β = 10). In a simulated adaptive scenario (20 fractions followed by replanning), deviations from the reference BED distribution were compared for uncompensated and BART-compensated plans after TPS re-optimization and BED-based DVH analysis.
Results
LCMM identified a tumor response subgroup requiring additional biological dose. The ODE model accurately reproduced longitudinal tumor volume dynamics and yielded an estimated α/β of 8.7 at the 20th fraction. Incorporating this radiosensitivity into BED calculations revealed a biological dose deficit of approximately 13.0% in uncorrected adaptive plans relative to the reference regimen. After BART-based compensation and TPS optimization, target BED–DVH metrics (BD2%, BD50%, BD95%, and BD98%) were restored to within ±1% of reference values. BED metrics for organs at risk showed minimal changes and remained within clinically acceptable limits.
Conclusion
BED-based biologically adaptive dose compensation derived from in-treatment tumor volume dynamics can be translated into deliverable TPS plans. BART corrects substantial biological dose deficits while preserving conventional dosimetric constraints, supporting its feasibility as a biologically informed adaptive radiotherapy strategy.