Thioridazine and the neuroleptic radioreceptor assay

When the neuroleptic radioreceptor assay (NRRA) has been used to monitor total neuroleptic-like activity (NLA) in the blood of patients taking thioridazine, the NLA values obtained from the NRRA are much lower than values calculated in the same sample by measuring the actual concentrations of parent drug and active metabolites and multiplying these values by the relative potency of each compound. The present paper demonstrates that in the NRRA for thioridazine or its active metabolites, the normal displacement of [³H]-spiperone from striatal membranes by thioridazine is altered in the presence of sera. The inclusion of serum (50 μliter/ml) distorts the sigmoidal displacement curves, such that the resulting log-logit (or Hill) slope is markedly decreased. Similar serum-induced changes in the log-logit slope are seen for two active metabolites of thioridazine but not for chlorpromazine of haloperidol. As a consequence, when one of these latter drugs is used as a standard, the NRRA substantially underestimates the actual NLA (chlorpromazine equivalents) values for patients treated with thioridazine. Moreover, because of differences in the magnitude of the effect with serum from different individuals, it is not possible to control completely for this effect. Thus, these data reconcile discrepancies that have been reported for data from the NRRA versus that from direct analytical measurements, and demonstrate that the use of the NRRA as a quantitative tool in the clinical pharmacology of thioridazine may lead to erroneous estimations of active drug and metabolites in the blood.