Regrettably, despite the elegant methods, the antipsychotic concentrations applied were not clinically relevant and would all be toxic or lethal to humans if found at those concentrations in the human plasma water or in the spinal fluid.

For example, haloperidol was used at 500 to 5,000 nanomolar. However, the therapeutic concentration of haloperidol in the spinal fluid is between 1 and 5 nanomolar. The toxic concentration is about 10 times higher, and the comatose-lethal concentration is approximately 20 times higher (see Regenthal et al., 1999), all these concentrations being far lower than those used by Tischbirek et al.

In addition, chlorpromazine was used at 500 to 5,000 nanomolar. But the therapeutic concentration of chlorpromazine is between 3 and 10 nanomolar, the toxic concentration is about 10 times higher, and the comatose-lethal concentration is about 100 times higher (Regenthal et al., 1999), all these concentrations being far lower than those used by Tischbirek et al.

All the concentrations mentioned here are in the spinal fluid or in the plasma water, where allowance has already been made for binding to plasma proteins.

Finally, it should be noted that the toxic concentrations used in vitro in this study are known to be membrane-lytic.

Considering the toxic and lethal concentrations used by Tischbirek et al., their proposed mechanism of antipsychotic drug action is not clinically relevant.

References:

Regenthal R, Krueger M, Koeppel C, Preiss R. Drug levels: therapeutic and toxic serum/plasma concentrations of common drugs. J Clin Monit Comput . 1999 Dec ; 15(7-8):529-44. Abstract

View all comments by Philip Seeman

Round and Round They Go
Tischbirek and colleagues are to be congratulated on completing a complicated, elegant in-vitro as well as in-vivo microdialysis study of the accumulation of a variety of antipsychotic drugs in synaptic vesicles in hippocampal tissue, their release upon neuronal depolarization, and their re-accumulation in neuronal terminals, where they appear to inhibit the further release of vesicles, leading to what they call use-dependent auto-inhibitory effects on synaptic transmission. The latter is the least convincing part of the study because they did not compare the concentrations of drugs in vesicles or in dialysates over periods that might parallel clinical treatment periods in patients, for example, weeks and months. Follow-up longitudinal studies of treatment effects on the levels of drugs within and releasable from vesicles will be of interest.

Tischbirek et al. claim their results are consistent with the delayed onset of clinical response to antipsychotic drugs, while others, for example, Kane and Kapur, are pressing the view that clinical response, when it does occur, does so rapidly. I would agree with Tischbirek on the issue of the slower response rate. The suggestion that this mechanism may explain why ECT, which would be expected to lead to vesicular emptying, is effective, as augmentation to antipsychotic drug treatment is intriguing and imaginative. Some specific tests of this in rodents would be welcome.

To the extent that accumulation of drugs into synaptic vesicles and their release into the synapse, leading to higher local concentrations of drugs than has been the common expectation does occur, polypharmacy would have to lead to complex mixtures which might be highly variable over time, although some steady state might eventually be reached.

View all comments by Herbert Meltzer