Reports of cardiac adverse events from oncology clinical trials often are at variance with reports derived from clinical observations or data-base reviews. These differences may lead to confusion, as different levels of risks abound in the literature, and the true cardiac risk of using some agents is uncertain. Additionally, such discrepancies may lead to the creation of over-cautious surveillance algorithms. Reasons for these reported differences are complex and often reflect subtleties in the criteria for individual patient evaluation. Both clinical trial data and real-world data have potential flaws that make reconciliation problematic. Importantly, however, both provide crucial information regarding the risk of adverse events. Major factors contribute to these differences including different tools used to diagnose events, and how those tools are interpreted. Additionally, differences in the populations of clinical trial participants and real-world populations play a crucial role. This paper looks at these differences and provides a perspective intended to help clinicians interpret reported variations in event rates derived from highly scrutinized clinical trials and broader real-world data.
Keywords: Clinical trial adverse event rates; Disproportionality analysis; Pharmacoepidemiology; Post-approval event rates; Reconciliation of event rate discrepancies.
© 2022. The Author(s).