Comparison of Three Radiobiological Models in Stereotactic Body Radiotherapy for Non-Small Cell Lung Cancer.

Objective: The applicability of the linear quadratic (LQ) model to local control (LC) modeling after hypofractionated radiotherapy to treat lung cancer is highly debated. To date, the differences in predicted outcomes between the LQ model and other radiobiological models, which are characterized by additional dose modification beyond a certain transitional dose (d _T ), have not been well established. This study aims to compare the outcomes predicted by the LQ model with those predicted by two other radiobiological models in stereotactic body radiotherapy (SBRT) for non-small cell lung cancer (NSCLC). Methods: Computer tomography (CT) simulation data sets for 20 patients diagnosed with stage Ⅰ primary NSCLC were included in this study. Three radiobiological models, including the LQ, the universal survival curve (USC) and the modified linear quadratic and linear (mLQL) model were employed to predict the tumor control probability (TCP) data. First, the d _T values for the USC and mLQL models were determined. Then, the biologically effective dose (BED) and the predicted TCP values from the LQ model were compared with those calculated from the USC and mLQL models. Results: The d _T values from the USC model were 29.6 Gy, 33.8 Gy and 44.5 Gy, whereas the values were 90.2 Gy, 84.0 Gy and 57.3 Gy for the mLQL model for 1-year, 2-year and 3-year TCP prediction. The remarkable higher d _T values obtained from the mLQL model revealed the same dose-response relationship as the LQ model in the low- and high-dose ranges. We also found that TCP prediction from the LQ and USC models differed by less than 3%, although the BED values for the two models were significantly different. Conclusion: Radiobiological analysis reveals small differences between the models and suggested that the LQ model is applicable for modeling LC using SBRT to treat lung cancer, even when an extremely high fractional dose is used.
Competing Interests: Competing Interests: The authors have declared that no competing interest exists.