Cardiac risk index as a simple geometric indicator to select patients for the heart-sparing radiotherapy of left-sided breast cancer

Introduction This is a dosimetric study to identify a simple geometric indicator to discriminate patients who meet the selection criterion for heart-sparing radiotherapy (RT). The authors proposed a cardiac risk index (CRI), directly measurable from the CT images at the time of scanning. Methods Treatment plans were regenerated using the CT data of 312 consecutive patients with left-sided breast cancer. Dosimetric analysis was performed to estimate the risk of cardiac mortality using cardiac dosimetric parameters, such as the relative heart volumes receiving ≥25 Gy (heart V25). For each CT data set, in-field heart depth (HD) and in-field heart width (HW) were measured to generate the geometric parameters, including maximum HW (HWmax) and maximum HD (HDmax). Seven geometric parameters were evaluated as candidates for CRI. Receiver operating characteristic (ROC) curve analyses were used to examine the overall discriminatory power of the geometric parameters to select high-risk patients (heart V25 ≥ 10%). Results Seventy-one high-risk (22.8%) and 241 low-risk patients (77.2%) were identified by dosimetric analysis. The geometric and dosimetric parameters were significantly higher in the high-risk group. Heart V25 showed the strong positive correlations with all geometric parameters examined (r > 0.8, p < 0.001). The product of HDmax and HWmax (CRI) revealed the largest area under the curve (AUC) value (0.969) and maintained 100% sensitivity and 88% specificity at the optimal cut-off value of 14.58 cm2. Conclusions Cardiac risk index proposed as a simple geometric indicator to select high-risk patients provides useful guidance for clinicians considering optimal implementation of heart-sparing RT.