Development and validation of a multivariable prediction model for major adverse cardiovascular events after early stage breast cancer: A population-based cohort study

Abstract

Aims Develop a score to predict the risk of major adverse cardiovascular events (MACE) after early stage breast cancer (EBC) to facilitate personalized decision-making about potentially cardiotoxic treatments and interventions to reduce cardiovascular risk. Methods Using administrative databases, we assembled a cohort of women diagnosed with EBC in Ontario between and results 2003 and 2014, with follow-up through 2015. Two-thirds of the cohort were used for risk score derivation; the remainder were reserved for its validation. The outcome was a composite of hospitalizations for acute myocardial infarction, unstable angina, transient ischaemic attack, stroke, peripheral vascular disease, heart failure (HF), or cardiovascular death. We developed the score by regressing MACE incidence against candidate predictors in the derivation sample using a Fine–Gray model. Discrimination was assessed in the validation sample using Wolber’s c-index for prognostic models with competing risks, while calibration was assessed by comparing predicted and observed MACE incidence. The risk score was derived in 60 294 women and validated in 29 810 women. Age, hypertension, diabetes, ischaemic heart disease, atrial fibrillation, HF, cerebrovascular disease, peripheral vascular disease, chronic obstructive pulmonary disease, and chronic kidney disease were significantly associated with MACE incidence and incorporated into the score. Ten-year MACE incidence was >40-fold higher for patients in the highest score decile compared to the lowest. The c-index was 81.9% (95% confidence interval 80.9–82.9%) at 5 years and 79.8% (78.8–80.8%) at 10 years in the validation cohort, with good agreement between predicted and observed MACE incidence. Conclusion Cardiovascular prognosis after EBC can be estimated using patients’ pre-treatment characteristics.