Abstract：
OBJECTIVE: Doses to the coronary arteries in breast cancer (BC) radiotherapy (RT) have been suggested to be a risk predictor of long-term cardiac toxicity after BC treatment. We investigated the dose-risk relationships between near maximum doses (Dmax) to the right coronary artery (RCA) and left anterior descending coronary artery (LAD) and ischemic heart disease (IHD) mortality after BC RT.
METHODS: In a cohort of 2,813 women diagnosed with BC between 1958 and 1992 with a follow-up of at least 10 years, we identified 134 cases of death due to IHD 10-19 years after BC diagnosis. For each case, one control was selected within the cohort matched for age at diagnosis. 3D-volume and 3D-dose reconstructions were obtained from individual RT charts. We estimated the Dmax to the RCA and the LAD and the mean heart dose (MHD). We performed conditional logistic regression analysis comparing piecewise spline transformation and simple linear modeling for best fit.
RESULTS: There was a linear dose-risk relationship for both the Dmax to the RCA (odds ratio [OR]/Gray [Gy] 1.03 [1.01-1.05]) and the LAD (OR/Gy 1.04 [1.02-1.06]) in a multivariable model. For MHD there was a linear dose-risk relationship (1,14 OR/Gy [1.08-1.19]. For all relationships, simple linear modelling was superior to spline transformations.
CONCLUSIONS: Doses to both the RCA and LAD are independent risk predictors of long-term cardiotoxicity after RT for BC In addition to the LAD, the RCA should be regarded as an organ at risk in RT planning.
METHODS: In a cohort of 2,813 women diagnosed with BC between 1958 and 1992 with a follow-up of at least 10 years, we identified 134 cases of death due to IHD 10-19 years after BC diagnosis. For each case, one control was selected within the cohort matched for age at diagnosis. 3D-volume and 3D-dose reconstructions were obtained from individual RT charts. We estimated the Dmax to the RCA and the LAD and the mean heart dose (MHD). We performed conditional logistic regression analysis comparing piecewise spline transformation and simple linear modeling for best fit.
RESULTS: There was a linear dose-risk relationship for both the Dmax to the RCA (odds ratio [OR]/Gray [Gy] 1.03 [1.01-1.05]) and the LAD (OR/Gy 1.04 [1.02-1.06]) in a multivariable model. For MHD there was a linear dose-risk relationship (1,14 OR/Gy [1.08-1.19]. For all relationships, simple linear modelling was superior to spline transformations.
CONCLUSIONS: Doses to both the RCA and LAD are independent risk predictors of long-term cardiotoxicity after RT for BC In addition to the LAD, the RCA should be regarded as an organ at risk in RT planning.
摘要：
目的：乳腺癌（BC）放疗（RT）中的冠状动脉剂量被认为是BC治疗后长期心脏毒性的风险预测因素。我们调查了BCRT后右冠状动脉（RCA）和左前降支（LAD）的近最大剂量（Dmax）与缺血性心脏病（IHD）死亡率之间的剂量风险关系。
方法：在1958年至1992年之间诊断为BC的2,813名妇女中，进行了至少10年的随访，我们确定了134例BC诊断后10-19年因IHD死亡的病例.对于每种情况，在诊断时年龄匹配的队列中选择了一个对照.从单个RT图表获得3D体积和3D剂量重建。我们估计了RCA和LAD的Dmax以及平均心脏剂量（MHD）。我们进行了条件逻辑回归分析，比较了分段样条变换和简单线性建模以获得最佳拟合。
结果：在多变量模型中，Dmax与RCA（比值比[OR]/Gray[Gy]1.03[1.01-1.05]）和LAD（OR/Gy1.04[1.02-1.06]）呈线性剂量风险关系。对于MHD，存在线性剂量风险关系（1,14OR/Gy[1.08-1.19]。对于所有的关系,简单的线性建模优于样条变换。
结论：RCA和LAD的剂量是BCRT后长期心脏毒性的独立风险预测因子。在RT计划中，RCA应被视为有风险的器官。
方法：在1958年至1992年之间诊断为BC的2,813名妇女中，进行了至少10年的随访，我们确定了134例BC诊断后10-19年因IHD死亡的病例.对于每种情况，在诊断时年龄匹配的队列中选择了一个对照.从单个RT图表获得3D体积和3D剂量重建。我们估计了RCA和LAD的Dmax以及平均心脏剂量（MHD）。我们进行了条件逻辑回归分析，比较了分段样条变换和简单线性建模以获得最佳拟合。
结果：在多变量模型中，Dmax与RCA（比值比[OR]/Gray[Gy]1.03[1.01-1.05]）和LAD（OR/Gy1.04[1.02-1.06]）呈线性剂量风险关系。对于MHD，存在线性剂量风险关系（1,14OR/Gy[1.08-1.19]。对于所有的关系,简单的线性建模优于样条变换。
结论：RCA和LAD的剂量是BCRT后长期心脏毒性的独立风险预测因子。在RT计划中，RCA应被视为有风险的器官。
