BARBE-Q: Bench to bedside: Applying Relative Biological Effectiveness modeling based on beam quality Q

• Applicant: Professor Dr. Armin Lühr
• Subject Area: Medical Physics, Biomedical Technology
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 551492504

In radiotherapy with protons, a relative biological effectiveness (RBE) of 1.1 is applied to account for the higher cell killing efficiency of protons compared to conventional photons. However, this simplistic constant RBE model neglects a large body of preclinical and emerging clinical evidence for a variable proton RBE and impacts patient treatment. The RBE depends on physical and biological factors, which vary across the treatment field. Therefore, accurate RBE values for individual irradiation conditions require adequate biomathematical models, while proton therapy practitioners demand simple models. Currently, the biggest problem for proton RBE modeling is a shortage of experimental in vivo and clinical RBE data. Existing RBE models are proton-specific, as they parameterize in vitro RBE as a function of linear energy transfer (LET). As an alternative to LET, the beam quality parameter Q is hypothesized to allow simple and ion-independent RBE modeling, which has been demonstrated for monoenergetic cell irradiation. Thus, RBE data from other ions could improve proton RBE prediction by overcoming the lack of proton RBE data. However, important steps to apply Q-based modeling to clinical treatment are still missing. The aim of the proposed project is to translate Q-based RBE modeling from cells to patients. Initially, a Q definition is established that robustly describes mixed particle and energy spectra, and then the Q parameter is analytically examined for its fundamental relationship to mechanistic RBE models for ion therapy. This is contrasted by a data-driven approach to extent Q-based RBW modeling along the translation chain from monoenergetic pencil beam irradiation to spread-out Bragg peak (SOBP) patient treatment fields and from cell survival to late toxicity, respectively. First, the Q definition is validated using a large dataset of SOBP cell irradiation experiments with photons (X) and different ions (H, He, C, Ne). In a second step, a clinically relevant in vivo toxicity endpoint (spinal cord injury) is considered using data from rat irradiations (X, H, He, C, O). Finally, the predictive power of Q-based RBE and normal tissue complication probability models is tested on a clinical data set consisting of patients with and without late radiation-induced brain toxicity after proton irradiation. High-precision Monte Carlo simulations will be used to calculate Q, LET and dose for each experimental and clinical irradiation setup. The performance of the Q-concept will be tested against established LET-based proton and mechanistic ion-independent RBE predictions. If successful, the Q-concept will enable data-driven clinical proton RBE modeling based on a large amount of (pre-)clinical photon and ion outcome data. By reducing uncertainties in predicting biological effectiveness, the proposed project will contribute to improved cancer treatment with protons and help reduce the risk of radiation-induced toxicity in patients.

• DFG Programme: Research Grants