Anti-Cancer Drug–Induced Life-Threatening Ventricular Arrhythmias

The article by Salem et al is an analysis of massive amounts of drug safety data highly relevant to the emerging field of Cardio-Oncology. It is timely due to the rapidly escalating number of drugs being marketed for the prevention or treatment of cancer and because many of these drugs have potentially toxic effects on the heart. The authors focus on analysis of reports of drug-induced long QT (diLQT) and ventricular arrhythmias (VA), including torsades de pointes (TdP), for anticancer drugs in the WHO’s voluntary adverse event reporting database known as Vigibase. They confirm their earlier finding that some anticancer drugs (eg, ibrutinib) are associated with VA without evidence of TdP. They summarize and compare their findings of potential signals (for diLQT, TdP, or VA) for these anticancer drugs in Vigibase to two additional sources frequently used by clinicians, ie, information in the manufacturers’ drug label and the CredibleMeds list of drugs that prolong QT and/or cause TdP. The authors correctly attribute discordances between the three sources to the characteristics of the source data for each and their intended purposes. For example, Vigibase analyses are intended to find potential signals for adverse events in a large heterogeneous dataset that require confirmation in subsequent research. The manufacturers’ labels are guides for clinicians that summarize data predominantly from controlled clinical trials conducted during drug development and are heavily influenced by FDA-mandated studies of the drug’s effect on QT interval in small numbers of highly selected patients or normal volunteers. The CredibleMeds drugs list is the product of an expert consensus process that places drugs in risk categories using all available data, ie, adverse event data from the FDA and Vigibase, the drug label and clinical reports in the medical literature.¹

The authors correctly note the limitations inherent to analyses of voluntary report databases such as Vigibase (or the FDA’s Adverse Event Reporting system), especially the lack of balance in drug exposure and unknown denominators. For this reason, the absolute numbers of reports shown in the tables and in figure 2 should not be used by the readers to estimate relative risk of harm. In fact, at this time, there is no source of data that would enable the ranking of drugs by their relative risk of diLQT, TdP or VA. The data summarized in this article can be useful to plan and guide future investigation in the new field of Cardio-Oncology.

References

1. Woosley RL, Romero K, Heise CW, et al. Adverse drug event causality analysis (ADECA): a process for evaluating evidence and assigning drugs to risk categories for sudden death. Drug Saf. 2017;40(6):465-474. https://link.springer.com/article/10.1007/s40264-017-0519-0