Impact of Breast volume, Body Mass Index, and Treatment Position on Heart and Lung Dose and Development of Acute Toxicities in Hypofractionated Whole Breast Irradiation
2020; Elsevier BV; Volume: 108; Issue: 3 Linguagem: Inglês
10.1016/j.ijrobp.2020.07.1152
ISSN1879-355X
AutoresAbhinav V. Reddy, Jean L. Wright, Amanda J. Walker, T.R. McNutt, Jason Hirshberg, F. Asrari, Victoria Croog, Susan Stinson, S.A. Dudley, Sara R. Alcorn,
Tópico(s)Cancer survivorship and care
ResumoPast studies have shown that breast volume (BV), body mass index (BMI), and prone vs. supine treatment position may impact rates of adverse events from whole breast irradiation (WBI). In this study, we evaluate complex relationships between these 3 inter-related candidate predictor variables on (1) heart and lung dose, and (2) risk of developing acute toxicities during hypofractionated WBI. Patients treated with tangent field WBI to 4050 cGy in 15 fractions, with or without boost, from January 2017 to February 2018 were identified using electronic medical records. Plans were contoured to include whole breast, ipsilateral lung, and heart volumes, and corresponding dosimetric data was collected. Acute toxicities were documented weekly during WBI, including dermatitis, hyperpigmentation, and fatigue per CTCAE v4, pain intensity scale (0-10), and presence of twinging, burning, tenderness, and rash. Univariable relationships between the 3 candidate predictor variables and lung and heart doses were evaluated using Pearson correlations and t-tests. Multivariable relationships between the 3 predictor variables and (1) lung and heart dose, and (2) acute toxicities were analyzed using linear and logistic regressions, respectively. Of 135 patients, WBI was delivered to the right breast in 57%, prone position was used in 54%, and boost was delivered in 78% of cases. Mean volume of the treated breast was 695 cm3 [standard deviation (SD) 453.] Ipsilateral mean lung dose (MLD) was 251 cGy (SD 220). Mean heart dose for left-sided WBI was 116 cGy (SD 28). Increasing BV and BMI were both correlated with lower ipsilateral MLD (rho = -0.38 and -0.30, respectively; both p<0.001). There were no associations between either variable and heart dose. Prone position was significantly associated with decreased MLD [96 cGy (SD 7) vs. 431 cGy (SD 26) for supine; t-test p<0.001). On multivariable analysis including BV and BMI, only BV remained associated with MLD (beta = -0.17, p<0.001); however, this relationship did not persist after controlling for treatment position. For acute toxicities, multivariable analyses including the 3 candidate variables plus boost delivery showed that increasing BV was associated with higher odds of pain score ≥2 (OR = 1.003, p = 0.02). BMI and treatment position were not independently associated with any toxicity outcomes. In this modern WBI series, an association between larger BV and lower MLD appears to be primarily explained via effect mediation by prone positioning. BV and BMI were not significantly linked to cardiac dose. These data suggest that historic predictors of lung and cardiac dose may no longer be relevant or may demonstrate inverted or complex associations in the era of prone WBI. Treatment position did not mitigate the association between larger BV and development of acute pain.
Referência(s)