Predicting risk of cardiotoxicity among young and emerging adult breast cancer patients from treatment to survivorship

Breast cancer is the most common cancer among women and cardiovascular disease (CVD) is prevalent among breast cancer survivors. This is due to shared risk factors between CVD and cancer, but also that breast cancer therapies are often cardiotoxic, which may later cause heart failure (HF). Cardiotoxicity from breast cancer chemotherapy affects between 10-20% of patients with enhanced risk in the presence of traditional CVD risk factors. However, there is a significant gap in our knowledge of cardiotoxicity among the rapidly growing population of young and emerging adult (YEA) breast cancer survivors, which comprise 5-12% of all breast cancer diagnoses. As survival from breast cancer increases, exposure to cardiotoxic chemotherapies at a younger age may enhance HF risk among YEA breast cancer survivors. Moreover, YEA breast cancer patients are more likely to have gene mutations that may also impair cardiac tissue function combined with a unique pattern of health behaviors and CVD risk factors. However, we are currently unable to predict which patients are at highest risk of cardiotoxicity. Studies suggest that gene expression may refine identification of women at increased risk of cardiotoxicity. To date, no studies examined whether combining gene expression and genetic mutations with CVD risk factors can identify YEA patients at increased risk of cardiotoxic effects of chemotherapy. To address this issue, I will complete the following specific aims: 1) Develop a predictive model combining psychosocial and traditional CVD risk factors to identify YEA breast cancer patients at increased risk of cardiotoxicity as defined by a decline in global longitudinal strain (GLS) or left ventricular ejection fraction (LVEF); 2) Investigate if the risk factor profile at diagnosis is associated with trajectory of GLS and LVEF during and after breast cancer treatment; and 3) investigate the impact of molecular biomarkers to risk prediction models. We will recruit a longitudinal cohort of n=300 YEA breast cancer patients treated at Northwestern Medicine. Among these participants, in a nested case-control design, we will select cases diagnosed with decline in GLS during chemotherapy (n=50) with age-matched controls without cardiotoxicity (n=50). For all participants, we will combine electronic health record (EHR) data with psychosocial and traditional CVD risk factors at three timepoints. For the nested case-control study, we will additionally measure gene expression at two timepoints. This directly informs my short-term career development goals to 1) Gain experience in HF and CVD etiology, epidemiology, and risk factors; 2) Develop skills in machine learning and bioinformatics approaches for prediction; and 3) Refine health informatics methods to integrate EHR with epidemiologic and molecular data. The skills and pilot data generated through this K01 will enable me to address the NHLBI compelling question (5.CQ.10) to reduce cardiac morbidity and mortality in cancer survivors. I will thus achieve my long-term career goal to identify and intervene on the CVD threats to the health and longevity of YEA cancer survivors.