Predicting changes in dose distribution to tumor and normal tissue when correcting for heterogeneity in radiotherapy for lung cancer

Purposes: The purposes of this study were to examine dose alterations to gross tumor volume (GTV) and lung using heterogeneity corrections and to predict the magnitude of these changes. Methods: Three separate conformal plans were generated for 37 patients with lung cancer: plan 1 corrected for heterogeneity, plan 2 did not correct for heterogeneity, and plan 3 used identical beams and monitor units from plan 2 but with heterogeneous calculations. Plans 1 and 2 were normalized to the 95% isodose line. Mean dose (MeanD(GTV)), maximum dose (MaxD(GTV)), and minimum dose (MinD(GTV)) to GTV and V-20 were compared between plans 1 and 3. For each patient, the amount of lung in all beam paths of plan 3 was quantified by a density correction factor and correlated with the percent change. Results: The median percent change in MeanD(GTV), MaxD(GTV), and MinD(GTV) between plan 3 and plan 1 was -4.7% (-0.1% to -19.1%, P < 0.0001), -5.59% (0.16% to -31.86%, P < 0.0001), and -4.88% (2.90% to -24.88%, P < 0.0001), respectively. The median V-20 difference was -1% (1% to -8%). The density correction factor correlated with larger differences in MeanD(GTV) on univariate analysis. Conclusions: Heterogeneity correction lowers the dose to GTV by 5%. This difference can be correlated with the density correction factor.