Author: Joana Antunes, Scott James Bright, David B. Flint, Mojtaba Hoseini-Ghahfarokhi, Mandira Manandhar, Poliana Marinello, Gabriel O. Sawakuchi, Tingshi Su 👨🔬
Affiliation: MD Anderson Cancer Center, The University of Texas MD Anderson Cancer Center, Laboratory of Instrumentation and Experimental Particle Physics; Faculty of Science of University of Lisbon, Department of Radiation Physics, The University of Texas MD Anderson Cancer Center, Department of Radiation Physics, MD Anderson Cancer Center 🌍
Purpose: To study the accuracy of relative biological effectiveness (RBE) models to predict the RBE of a comprehensive panel of pancreatic cancer cell lines.
Methods: Clonogenic cell survival was acquired for a panel of 45 pancreatic cancer cell lines irradiated with photons (6 MV x-rays at 10 cm depth) and protons (200 MeV with 16.5 cm range, SOBP of 4 cm at 15.6 cm depth, 3.85 keV/mm dose-weighted LET in water). We fit the clonogenic survival curves to extract radiosensitivity metrics including D10%, D50%, mean inactivation dose (MID), survival fraction at 2 Gy (SF2Gy) and their corresponding RBE values. Predictions of 11 previously reported RBE models were compared to experimental values among the panel of cell lines.
Results: The cell survival curves exhibited a wide range of responses to both photons and protons with up to 153±8% and 151±10% variability in D10% for photons and protons, respectively. The RBE at D10% varied from 0.83 ± 0.06 to 1.76 ± 0.15 across the cell lines. The RBE models presented a wide range of accuracies depending on which survival metric was assessed, with some models substantially outperforming others for a given metric. For RBE at D10%, the models predicted the measured RBE within a root mean square error (RMSE) of 0.100 to 0.416 depending on the model selected. The models performed better when estimating the RBE of radioresistant cell lines (mean RSME=0.112 for RBE at D10%) versus radiosensitive ones (mean RSME=0.249 for RBE at D10%) across the cell line panel.
Conclusion: Pancreatic cancer cell lines showed a wide range of variabilities in radiosensitivity to both photons and protons as well as in RBE values. Discrepancies observed between the experimental and modeled RBE highlight the importance of refining the RBE models to integrate other parameters related to radiosensitivity.