Withdrawal of Terodiline: A Tale of Two Toxicities

Withdrawal of Terodiline: A Tale of Two Toxicities

INTRODUCTION

Apart from drug-induced prolongation of the QT interval, and its subsequent degeneration into torsade de pointes, it is difficult to think of another type A pharmacological adverse drug reaction that has been responsible for the withdrawal of so many drugs from the market over the last two decades. With-drawal of prenylamine in 1988, followed by that of lidoflazine in 1989 and terodiline in 1991, was to herald a similar misfortune for many other drugs such as terfenadine, astemizole, cisapride, sertindole, grepafloxacin, droperidol, thioridazine and levacetyl-methadol. A number of other drugs, such as pimozide, halofantrine, lumefantrine and mizolastine to name just four, had severe prescribing restrictions placed on their clinical use for similar reason, while others such as moxifloxacin, gatifloxacin and ziprasidone have had their approval greatly delayed in some Member States of the European Union (EU) because their ‘QT-liability’ was determined to adversely affect their risk–benefit ratio. Not surprisingly, many drugs have recently had their clinical development terminated, some at a fairly advanced stage, as a result of their potential to prolong the QT interval (Shah, 2002).

Withdrawal of terodiline has a number of important lessons for drug development and pharmacovigilance. Firstly, from a regulatory perspective, terodiline is almost too perfect an example of drugs whose more potent secondary pharmacological effects, observed as adverse drug reactions during their originally intended clinical uses, have led to their clinical re-development for completely different indications. In the case of terodiline, this concerned its potent anticholinergic side effect observed during its approved use as an antiang-inal agent. Terodiline illustrates how such a strategy can be eclipsed by the virulent appearance of additional secondary pharmacological effects that are not fully explored. With terodiline, this additional activity was its adverse effect on cardiac repolarization and QT inter-val duration on the surface electrocardiogram (ECG). Indeed, terodiline might therefore be described as a ‘pharmaceutical boomerang’. It serves as a reminder of the limitations of drug development programmes in characterizing a relatively rare, but potentially fatal, clinical hazard. Secondly, it emphasizes both the perils of failing to appreciate the problems associated with other members of the same chemical, pharmacologi-cal or therapeutic class of drugs (prenylamine in the case of terodiline), and the necessity of applying all available techniques to characterize a potential class-related safety issue when developing a new drug. This is particularly unfortunate, since drug-induced QT inter-val prolongation is a concentration-dependent type A adverse drug reaction that can be investigated during preclinical and clinical phases of drug development, and therefore ought to be predictable. Finally, the post-marketing identification of the proarrhythmic risk asso-ciated with terodiline through a spontaneous report-ing system emphasizes the strengths of systems such as the United Kingdom (UK) Yellow Card Scheme in comparison with formal post-marketing surveil-lance studies that had continued to assert its cardiac safety.

This chapter will focus on a comparison between terodiline and prenylamine with a view to providing a framework of some of the major issues that need to be considered when preparing the pre-marketing Safety Specification of a new drug, as required by the International Conference on Harmonization (ICH) E2E guideline, and discussing the potential risks that require further evaluation. In this context, it will also discuss the ICH E1A guideline on the clinical safety dataset required to assess the safety of medicines intended for chronic use, and the recently adopted ICH S7B and ICH E14 guidelines on pre-approval investigation of drugs for their potential to prolong QT interval.