(P124) Effect of Inhomogeneity on Cardiac and Lung Dose in Partial-Breast Irradiation Using HDR Brachytherapy

Stuti Ahlawat, MD, Leonard Kim, MS, Venkat Narra, PhD, Evelyn Sebastian, BS, Amy Limbacher, BS, Peter Chen, MD, Atif Khan, MD; Department of Radiation Oncology, Rutgers Cancer Institute of New Jersey; Department of Radiation Oncology, William Beaumont Hospital

Introduction: Recent expert consensus statements have endorsed use of accelerated partial-breast irradiation (APBI) in select groups of low-risk women with early-stage breast cancer. APBI in the form of balloon brachytherapy is increasingly selected as a method of radiation treatment (RT). In APBI, accurate dose distribution representation is essential when considering organs at risk (OARs). The presence of inhomogeneities, like air pockets and contrast material in the bone, creates challenges for accurate dose estimation. Current treatment planning systems are based on the TG-43 dose calculation formalism, which assumes full scatter conditions and does not correct for inhomogeneities. Recent advances in dose calculation software have made estimation of the dose that is corrected for inhomogeneities possible. The aim of the current study is to examine the effect of inhomogeneity effects on cardiac and lung dose calculation as estimated by the Acuros BV (Varian) dose calculation algorithm, compared with uncorrected TG-43 doses. Accurate heart and lung dose calculation may be important to minimize long-term toxicity.

Materials and Methods: This study looked at 29 MammoSite balloon brachytherapy patients treated with a single central dwell position to 3,400 cGy in 10 bid fractions. Balloon density was overwritten to water in the treatment planning system so as to isolate the tissue inhomogeneity effect. Doses were calculated using both TG-43 and Acuros. Maximum and mean heart dose and lung V20 were recorded. A linear fit was used to characterize the relationship between Acuros and TG-43 calculations of these parameters.

Results: Maximum cardiac doses calculated using TG-43 doses ranged from 232 to 4,422 cGy. Maximum cardiac doses as calculated by Acuros were proportional to TG-43 doses. For these patients, a linear fit shows Acuros dose = 0.97 * TG-43 dose (R² = .999). The 95% confidence interval (CI) for the proportionality constant was 0.963–0.978. For any individual patient, the proportional relationship agreed with the actual Acuros dose to 3 ± 2%. When examining lung doses, the Acuros lung V20 is about 80% of TG-43 (R² = .962).

Discussion: In this small cohort of partial-breast balloon brachytherapy patients, we found that differences in cardiac dose between TG-43 and Acuros (~3%) were less than those reported in previous studies on other critical structures, such as ribs (4%–5%) and skin (7%–8%). Compared with these other structures, we hypothesize that the low-density lung between the balloon and heart may be counterbalancing dose decrease due to lack of full scatter, but further study is warranted. As with ribs and skin, Acuros maximum point dose calculation for heart showed a proportional relationship to TG-43. However, the dose-volume parameter, lung V20, displayed larger variance from proportionality, as shown by the lower R² value. With evolving technology, it is important to elucidate accurate normal structure dosages to deliver high-quality care in this important patient population. Further study with a larger number of patients is needed to make any definitive recommendations. Our current results appear to be reassuring, in that Acuros-based heart and lung doses are lower than or comparable with TG-43-based calculations.