Abstract 16332: Novel Risk Model for Predicting Cardiovascular Outcome in Breast Cancer Survivors

Introduction: Both conventional cardiovascular (CV) risk factors and breast cancer (BC) therapy-related CV risk factors play an important role in CV mortality among BC survivors. Hypothesis: A risk stratification scheme which is based on both conventional CV risk factors and cancer treatment-related risk factors may have a good performance toward estimating the risk of CV the event among these survivors. Methods: The cohort of the study consists of 1,256 female BC patients from 4 medical centers in South Korea with a mean follow up duration of 51.6 ± 30.8 months. The cohort was randomized on a 1:1 ratio into the derivation group and validation group. A competing risk analysis model was used to derive the risk scheme. The primary endpoint was the composite of CV mortality, myocardial infarction (MI), congestive heart failure (CHF), and transient ischemic attack (TIA)/stroke. Results: We developed the CHEMO-RADIAT score (CHF [2 point], Elderly [age≥60; 1 point], prior MI/peripheral artery disease [2 points], Obesity [body mass index ≥30 kg/m 2 ; 1 point], Renal failure [glomerular filtration rate <60 ml/min/1.73 m2; 1 point], Abnormal lipid profile [1 point], Diabetes mellitus [1 point], Irradiation to left side breast [1 points], Anthracycline dose [1 point per 100mg/m 2 doxorubicin equivalent dose], TIA /stroke [2 points]) from the development group. The time-dependent C-index improved significantly when conventional CV risk factor data was incorporated into the BC treatment-related factors in the development (0.739 vs. 0.582, p=0.017). The time-dependent C-index of the CHEMO-RADIAT score at 5 years was 0.739 (95% confidence interval [CI] 0.615-0.864), and 0.848 (95% CI 0.721-0.976) in the derivation and validation group, respectively. Conclusions: A simple score for predicting CV risk in breast cancer patients was developed and validated well in this multi-center cohort. The CHEMO-RADIAT score may provide overall CV risk stratification in breast cancer.