Cardiovascular Events in Pivotal Cancer Trials

Study Questions:

How are cardiovascular (CVD) events reported in clinical trials of contemporary cancer therapies?


The authors queried the Drugs@FDA database to identify all anticancer therapies that received new drug application approval between 1998 and 2018. They subsequently reviewed pertinent published phase II and III clinical trial data to collect data on the reporting of CVD events. The primary outcome was the rate of reporting of major adverse cardiovascular events (MACE), which included myocardial infarction, stroke, heart failure, atrial fibrillation, coronary revascularization, and cardiac death. The secondary outcome was the report of any CVD event during the trial. Statistical analysis accounted for extensive covariates including cancer type, therapeutic class, trial size, funding source, time and duration of trial, whether patients with CVD were excluded, prior cardiotoxicity report within the same class of drugs, and excess CVD risk reported in prior trials. The authors used the Multi-Ethnic Study of Atherosclerosis observational prospective cohort as a comparison population.


Overall, 189 trials met the authors’ criteria, evaluating 123 drugs and enrolling 97,365 participants (mean age, 59 years; 46% female; 73% on biologic, targeted, or immune-based therapies; median follow-up, 30 months). Only two thirds of trials reported CVD events. Trial characteristics were not associated with reporting of CVD. MACE and any CVD events were more common in intervention arms compared to controls (792 vs. 356 for MACE and 3,142 vs. 1,597 for any CVD). Efficacy of drugs was not associated with CVD reporting. Overall, the reported incidence rate for MACE in the cancer trials was significantly lower than that of similar aged noncancer population (524 vs. 1,408 per 100,000 person-years).


Close to 40% of cancer therapy trials did not report CVD events. The overall reported CVD incidence rate in the trials was lower than in similar aged noncancer population.


As CVD is increasingly being recognized as a major barrier for the long-term survival and quality of life of patients with cancer, the markedly lower rate of reported CVD events in cancer trials compared to noncancer patients gives one pause. For example, trials of trastuzumab report an incidence of heart failure of 2.5%, while the observed incidence in practice is closer to 20%. The authors astutely discuss the potential reasons for the discrepancy, including the heterogeneity in definitions used and interpretation of signs or symptoms of CVD, selection bias towards excluding patients with CVD in trials, in addition to limitations of the study such as the lack of uniform primary source of data and the variation of CVD adjudication across trials. Nonetheless, this study is a call for a systematic re-evaluation of CVD reporting in clinical trials of cancer therapy to improve the assessment of CVD adverse effects of upcoming cancer therapies.

Clinical Topics: Arrhythmias and Clinical EP, Cardiac Surgery, Cardio-Oncology, Heart Failure and Cardiomyopathies, Invasive Cardiovascular Angiography and Intervention, Prevention, Implantable Devices, SCD/Ventricular Arrhythmias, Atrial Fibrillation/Supraventricular Arrhythmias, Cardiac Surgery and Arrhythmias, Cardiac Surgery and Heart Failure, Acute Heart Failure

Keywords: Arrhythmias, Cardiac, Atherosclerosis, Atrial Fibrillation, Biological Products, Cardiotoxicity, Death, Sudden, Cardiac, Heart Failure, Myocardial Infarction, Myocardial Revascularization, Neoplasms, Quality of Life, Secondary Prevention, Stroke

< Back to Listings