Program Information
Comparison Between Mechanistic Radiobiological Modeling Vs. Fowler BED Equation in Evaluating Lung Cancer Radiotherapy Outcome for a Broad Range of Fractionation
J Jeong*, J Oh , A Fontanella , M Crispin-Ortuzar , J O Deasy , Memorial Sloan Kettering Cancer Center, New York, NY
Presentations
SU-E-FS1-2 (Sunday, July 30, 2017) 1:00 PM - 1:55 PM Room: Four Seasons 1
Purpose: Fowler’s BED equation with repopulation term has long been demonstrated its usefulness in evaluating therapeutic effect of various fractionations in radiation therapy. In this work, we compared the mechanistic modeling of lung cancer outcome for various fractionations with the simple BED equation based outcome analysis.
Methods: The mechanistic model incorporating key radiobiological effects, such as hypoxic, cell cycle, reoxygenation, and repopulation effects, was used to derive equivalent doses in 2 Gy/fx for a given α/β ratio (EQD2_α/β,model) for non-conventional fractionation regimes. The model was applied to lung cancer outcome data of various fractionation schedules with 38 cohorts (n=2701). Various radiosensitivity parameters were tested to find the best-fit parameter values. Considering the treatment duration, Fowler’s BEDs were calculated for the same cohorts, using standard parameter values (α=0.35 Gy⁻¹, α/β=10 Gy, T_k=28 days, and T_p=3 days). The best-fit EQD2_α/β,model values were compared against BED₁₀ values. In the scatter plots of the outcome vs. EQD2_α/β,model or BED₁₀, logistic regressions were performed, assuming the slope of the dose response curve (γ₅₀) of 1.5. Chi-squared tests were performed to test the validity of the fits.
Results: For the model, the best-fit was found with the parameter values of α=0.305 Gy⁻¹, α/β=2.8 Gy, and OER_I=1.7. The resulting EQD2₂.₈ values are highly correlated with the BED₁₀ (R²=0.985). The dose response curves from both EQD2₂.₈ and BED₁₀ precisely fit the data with similar TD₅₀ values (62.1 and 64.0 Gy, respectively) and high Chi-squared p-values (p=0.999 and 0.995, respectively).
Conclusion: The mechanistic model including major radiobiological factors can accurately predict the outcome of lung cancer for a broad range of fractionation, in agreement with the Fowler’s BED equation that has long been used in clinic. The difference in radiosensitivity values (α/β=2.8 vs. 10 Gy) suggests that the intrinsic radiosensitivity might be different from the clinically observed one.
Funding Support, Disclosures, and Conflict of Interest: This research was supported by research grants from Varian Oncology and the NIH (R01 CA85181 as well as MSKCC Core Grant P30 CA008748).
Contact Email: