Collier, Patrick

Background Cardiovascular disease (CVD) is a leading cause of death in cancer survivors. Predicting CVD risk in this population remains challenging. Risk prediction models that use machine learning algorithms could provide an objective method for accurate risk prediction to facilitate the prevention and management of CVD in cancer survivors. We evaluated previously tested machine learning algorithms and logistic regression (regularized by machine learning methods), in addition to testing and validating newer, more complex machine learning algorithms, for CVD prediction in cancer survivors. Methods This multicenter study used a database of 3835 multiracial cancer survivors with 89 clinical, laboratory, and echocardiographic features over 20 years. Models were trained using repeated random and time‐split samples and tested on a separate cohort of 329 patients. Model performance was assessed using the area under the receiver operating characteristic curve. Results Regularized logistic regression achieved an area under the receiver operating characteristic curve of 0.845 (heart failure), 0.783 (atrial fibrillation), 0.792 (coronary artery disease), and 0.806 (composite CVD). These are comparable to 0.837 and 0.848 for heart failure, using Bayesian additive regression tree and random forest as more advanced machine learning models, respectively. De novo composite CVD (post‐cancer diagnosis) was also predicted with an area under the receiver operating characteristic curve of 0.826 using regularized logistic regression, compared with 0.735 and 0.802 using decision tree and random forest, respectively. Conclusions Regularized logistic regression and advanced machine learning models demonstrated similar predictive performance, with institutional transferability. These tools may support risk stratification and prevention strategies in cardio‐oncology using longitudinal data. Registration URL: https://www.clinicaltrials.gov ; Unique identifier: NCT05377320.

BACKGROUND: Global collaboration in cardio-oncology is needed to understand the prevalence of cancer therapy-related cardiovascular toxicity in different risk groups, practice settings, and geographic locations. There are limited data on the socioeconomic and racial/ethnic disparities that may impact access to care and outcomes. To address these gaps, we established the Global Cardio-Oncology Registry, a multinational, multicenter prospective registry. METHODS: We assembled cardiologists and oncologists from academic and community settings to collaborate in the first Global Cardio-Oncology Registry. Subsequently, a survey for site resources, demographics, and intention to participate was conducted. We designed an online data platform to facilitate this global initiative. RESULTS: A total of 119 sites responded to an online questionnaire on their practices and main goals of the registry: 49 US sites from 23 states and 70 international sites from 5 continents indicated a willingness to participate in the Global Cardio-Oncology Registry. Sites were more commonly led by cardiologists (85/119; 72%) and were more often university/teaching (81/119; 68%) than community based (38/119; 32%). The average number of cardio-oncology patients treated per month was 80 per site. The top 3 Global Cardio-Oncology Registry priorities in cardio-oncology care were breast cancer, hematologic malignancies, and patients treated with immune checkpoint inhibitors. Executive and scientific committees and specific committees were established. A pilot phase for breast cancer using Research Electronic Data Capture Cloud platform recently started patient enrollment. CONCLUSIONS: We present the structure for a global collaboration. Information derived from the Global Cardio-Oncology Registry will help understand the risk factors impacting cancer therapy-related cardiovascular toxicity in different geographic locations and therefore contribute to reduce access gaps in cardio-oncology care. Risk calculators will be prospectively derived and validated.