Hypoxia-informed RBE-weighted beam orientation optimization for intensity modulated proton therapy

Area of Science:

• Medical Physics
• Radiation Oncology
• Biophysics

Background:

• Proton therapy planning can be optimized using variable relative biological effectiveness (RBE) models.
• Proton RBE reduction in hypoxic tumors presents an opportunity to improve radioresistance via RBE-weighted optimization.

Purpose of the Study:

• To develop and evaluate a voxel-level RBE-weighted optimization algorithm incorporating partial oxygen pressure estimations.
• To integrate hypoxia-informed RBE values into a beam orientation optimization (BOO) algorithm for glioblastoma treatment planning.

Main Methods:

• Utilized [18F]-fluoromisonidazole (FMISO)-PET/CT images to derive voxel-level oxygen values in three glioblastoma patients.
• Calculated McNamara RBE, weighted by oxygen enhancement ratios (OER), and incorporated into the BOO algorithm's dose fidelity term.
• Compared the hypoxia-informed RBE-weighted method (HypRBE) against constant RBE (cRBE) and normoxic McNamara RBE (RegRBE) models using tumor homogeneity index, biological dose metrics, and organ at risk therapeutic index.

Main Results:

• HypRBE significantly increased tumor homogeneity index, maximum biological dose, and D95% compared to cRBE and RegRBE.
• Despite minor increases in organ at risk doses, HypRBE improved the average therapeutic index by over 20%.
• In best-case scenarios, HypRBE escalated tumor coverage without compromising organ at risk sparing, enhancing the therapeutic ratio.

Conclusions:

• A novel optimization algorithm (HypRBE) was developed, incorporating hypoxia-informed RBE values into the dose fidelity term.
• HypRBE demonstrated the ability to deliver higher biological doses to hypoxic tumor regions with lower RBE.
• The method effectively spares radiosensitive normal tissues, indicating potential for improved glioblastoma treatment outcomes in proton therapy.