Dose-Dependent LET Constraints and Constraint Resolution for Interpreting Proton Therapy Toxicity

Area of Science:

• Radiation Oncology
• Medical Physics
• Radiobiology

Background:

• Dose-averaged linear energy transfer (LET_D) is crucial for predicting normal tissue toxicity in proton therapy.
• Current LET_D constraints often use single-value thresholds, neglecting the coupled dose-LET relationship vital for biological effects.

Purpose of the Study:

• Develop and evaluate a probabilistic framework for dose-dependent LET_D constraints.
• Assess the consistency of these probabilistic constraints with existing clinical toxicity data.

Main Methods:

• Derived dose-LET_D constraint curves using linear quadratic-based variable relative biological effectiveness (RBE) models.
• Incorporated uncertainty from RBE models, tissue radiosensitivity, and fractionation into probabilistic envelopes.
• Compared derived constraints with clinical toxicity data for brainstem necrosis, brain necrosis, RIBI, rib fracture, and osteoradionecrosis.

Main Results:

• Probabilistic constraints aligned well with clinical data for brainstem necrosis, RIBI, and osteoradionecrosis.
• Toxicity-associated dose-LET_D values in patients exceeded constraint envelope upper bounds.
• Limited empirical support for higher LET constraints was found, especially with the newest RBE model.

Conclusions:

• Dose-dependent probabilistic LET_D constraints offer a biologically sound alternative to single-value thresholds in proton therapy.
• This framework better reflects the interplay of dose and LET in predicting normal tissue complications.