To the Editor: My colleagues and I thank Drs. Sumiyoshi and Meltzer for commenting on our study, and we agree with them when they say that our results extend their previous observations with schizophrenic patients (Sumiyoshi et al., 2001). They comment that we failed to mention the discrepant results regarding the effect of the lower dose (30 mg/day) of tandospirone on memory performance, i.e., the facilitative influence in their studies versus the partially inhibitory effects in our study reported in the Journal. However, the patients in their study received tandospirone, 10 mg t.i.d., while the subjects in our study received the 30 mg of tandospirone at one time. Because of this difference, we cannot directly compare their results with ours. Previous studies have shown that the half-life of the plasma concentration of tandospirone was relatively short (1.2–1.4 hours), and the effect of the accumulation of this drug is reported to be negligible when the subjects received 10 mg t.i.d. of tandospirone for more than 3 months (1, 2). From this point of view, the plasma concentration of tandospirone in their studies might have been lower than that of ours.

They pointed out that the apparent difference between the two studies is that we studied the acute effect of tandospirone while they assessed its long-term effect over 6 weeks. They suggested that some adaptive changes in 5-HT_1A receptor function occurs during prolonged administration of the drug, leading to an altered response of 5-HT_1A receptors to intrinsic 5-HT or the 5-HT_1A agonist. As to this point, we agree with their notion, and future work must address the details of the long-term effect of tandospirone on the adaptive changes in 5-HT_1A receptor function.

They also claimed that the difference was that our study used normal subjects, while their studies were concerned with patients with schizophrenia. In our recent study (3), we measured the availability of 5-HT_1A receptors in patients with schizophrenia and normal comparison subjects using positron emission tomography with [¹¹C]WAY-100635. With regard to the binding of 5-HT_1A receptors in the hippocampus, we found no significant difference between patients and comparison subjects (3). But this result did not contradict the different effect of tandospirone on the explicit memory function between normal subjects and patients. Cellular localization of 5-HT_1A receptors has been demonstrated on both cholinergic and glutamatergic neurons (4, 5), and the effect of tandospirone on memory function might be affected by disease change of these neurotransmitter systems. The interaction between 5-HT_1A receptors and the disease change of several neurotransmitter systems needs to be considered when assessing the effect of tandospirone on memory function in patients with schizophrenia and other psychiatric disorders.

In conclusion, we share the opinion of Drs. Sumiyoshi and Meltzer that agents that activate presynaptic 5-HT_1A receptors, as well as drugs that work as antagonists at postsynaptic 5-HT_1A receptors, may be of value in treating some of the cognitive deficits in schizophrenia and possibly other psychiatric disorders as well.