Development of Single Enantiomers or Metabolites of Marketed Racemic Drugs

DEVELOPMENT OF SINGLE ENANTIOMERS OR METABOLITES OF MARKETED RACEMIC DRUGS

The comparison between prenylamine and terodiline described in this chapter shows the strengths of a scientific synthesis of all the available information when evaluating the significance of even a handful of spontaneous reports of an adverse event, and formu-lating the most appropriate regulatory strategies for risk management. This is especially relevant when another member of the same chemical, pharmacologic or therapeutic class is associated with the same low frequency adverse event.

The marketing authorization holder of terodiline has to be commended for the speed and the willingness with which the drug was withdrawn as soon as it became evident that the risk is unlikely to be immediately manageable. Unfortunately, they did not follow up the recommendation from the regulatory assessor to investigate separately the two enantiomers system-atically for their pharmacology, and possibly develop one of these if it can be shown to be devoid of potassium-channel-blocking activity while retaining a beneficial therapeutic effect. In the light of subsequent investigations showing that − -(S)-terodiline does not affect the QTc interval (Hartigan-Go et al., 1996) and does indeed have some anticholinergic proper-ties, the possibility that − -(S)-terodiline might have a much superior risk–benefit profile compared to the racemic mixture is a real one. At the time of its with-drawal in 1991, the development of a single enan-tiomer may have appeared an arduous and potentially unrewarding activity, but paradoxically this has been one of the striking features of new drug develop-ment in the period 1994–2002. This trend has resulted in the development of (S)-ketoprofen, (S)-ofloxacin, (S)-omeprazole, (R)-salbutamol, (S)-citalopram and (S)-ketamine among many others that are still in the pipeline (Shah, 2000).

It is interesting that astemizole has two metabo-lites – desmethylastemizole and norastemizole. Preclinical data show that desmethylastemizole is as cardiotoxic as the parent drug. Since desmethylastem-izole has a very long half-life relative to astemizole, plasma levels of desmethylastemizole are generally about 30-fold higher than that of astemizole, and the clinically observed cardiotoxicity appears to be mainly due to desmethylastemizole. In one patient with astemizole-induced torsade de pointes, plasma desmethylastemizole and astemizole concentrations were 7.7–17.3 ng/mL and < 0 5 ng/mL, respectively (Volperian et al., 1996). Not surprisingly, cardiotox-icity of astemizole is the highest following an over-dose, or when a high loading dose is administered to quickly achieve the steady-state therapeutic concen-trations (Anon, 1987). In both these situations, there is rapid accumulation of desmethylastemizole. Findings such as these not only preclude the development of some metabolites, but also illustrate the strengths of simple observations that should guide the drug devel-opment programme and evaluation of post-marketing case reports of adverse drug reactions.

Development of active but safer metabolites which are devoid of the unwanted secondary cardiotoxic pharmacology, or unwanted metabolic profile and drug interaction potential, has been another trend in drug development (Shah, 2005a). Preclinical data have suggested that the risk–benefit ratio might be superior for the metabolite compared to the corre-sponding parent drug for fexofenadine (a metabo-lite of terfenadine), norcisapride (a metabolite of cisapride), norastemizole (a metabolite of astemizole), desmethylloratadine (a metabolite of loratadine) or norlevacetylmethadol (a metabolite of levacetyl-methadol). These preclinical leads have already been followed up for some of these metabolites, and fexofe nadine and desmethylloratadine are now already on the market.