Research Project
10114664
7. Project Summary/Abstract Patients with atherosclerotic cardiovascular disease (ASCVD) are at high risk of suffering from future ASCVD events despite contemporary preventive therapies, and this risk has been termed residual risk. Even with recent advances in cardiovascular interventions including statins, dual antiplatelet therapies, revascularization, and other therapies, ASCVD event rates among patients with prior ASCVD remain high (>3% per year). While a disproportionate number of events occur in patients who have prior ASCVD, risk assessment algorithms for predicting recurrent events occurring in these patients has been limited. Furthermore, patients with ASCVD have a range of risk, and recent guidelines now recommend tailoring the intensity of secondary prevention treatments to the patient's risk level. Yet personalized risk prediction models for patients with ASCVD are lacking. The goal of this project is to use advanced machine learning methods to develop and validate a novel risk score (Residual Risk Score) for predicting absolute long-term risk of future events among patients with ASCVD. Novel machine learning methods can analyze complex interactions and correlations between risk factors and other traits. We hypothesize that these computational approaches could be used to develop and validate a personalized risk assessment for patients with ASCVD. This project will contribute clinically meaningful and readily applicable results to directly impact a common clinical problem with high morbidity and cost burden; namely, reducing residual risk in ASCVD patients and personalizing their treatments. The Residual Risk Score will be provided as a free clinical tool and could be incorporated in the electronic health record or as an application. In this study, we will utilize a combination of data from biobanks, electronic health records, and population studies. In Specific Aim 1, we will use advanced computational methods to design and train an artificial neural network for predicting recurrent ASCVD events among patients with prior ASCVD. In Specific Aim 2, we focus on expanding the computational model with the capability of predicting these events in independent and diverse populations. This study will address an important unmet need for better risk stratification tools for patients with ASCVD. This would aid in identifying higher risk ASCVD patients more likely to suffer a recurrent event, and could be used to tailor novel and costly risk reduction strategies to higher risk patients. It has the potential to impact clinical care for shared decision-making discussions between healthcare providers and patients, as well as for clinical trials that target higher risk ASCVD patients for novel costly therapeutics.
