The Promise of New Drugs for Schizophrenia Treatment

CA Tamminga, MD¹

Schizophrenia is one of the most serious human brain diseases, yet it remains without a known etiologic or pathophysiologic understanding. It affects 1% of the world's population without ethnic, educational, or socioeconomic prejudice. Several factors are known to contribute to its onset without any one of them being essential or critical as far as we know: positive family history; poor birth history; second trimester intrauterine assault; and, possibly, adolescent drug use (1,2). The disease runs a chronic, lifelong course; onset is characteristically in the early 20s for men and in the mid-20s for women. The early illness course is often characterized by repeated acute episodes with florid psychosis and overall psychosocial deterioration. Midlife illness can be rather smooth with occasional exacerbations. The later-life illness course can be smoother, often with frank loss of positive psychotic symptoms. Prognosis is poor overall, but female schizophrenic persons have a better psychosocial outcome than their male counterparts. Psychotic symptoms of the illness are often grouped into positive (hallucinations, delusions, paranoia), cognitive (thought disorder, bizarre behaviour), and negative (anhedonia, alogia, thought paucity) symptoms. Whether these clusters are multiple manifestations of a single illness or multiple similar illnesses remains a topic of hot debate (3,4), but firm evidence for either position is lacking. The single most replicable biologic observation in schizophrenia is the reduction of its psychotic symptoms in response to antidopaminergic treatments.

Chlorpromazine was serendipidously discovered to have specific antipsychotic properties in the 1950s (5); shortly after, its pharmacologic characteristic of dopamine receptor blockade was linked to its antipsychotic action (6). Several different chemical families of neuroleptics were identified, developed, and introduced in the next 2 decades. After this early productivity, drug discovery almost stopped in the area of schizophrenia therapeutics for about 15 years. After 1978 until the approval of clozapine for psychosis in 1990, no new antipsychotic drugs were brought before the United States (US) Food and Drug Administration (FDA). This profound hiatus in drug development for psychosis may have reflected the idea, however unfounded, that schizophrenia was well treated by available agents, or possibly the daunting task of drug discovery in a disease of unknown pathophysiology. Whatever the reason, that “dry period” was demoralizing for patients, families, and caregivers alike. It was not until public pressure for research heightened and governmental funding agencies responded that industry and academia alike became active again in schizophrenia therapeutics. Experience from studies with clozapine may well have fuelled this resurgence of research by suggesting that additional aspects of schizophrenia are responsive to treatment.

Traditional Neuroleptics

It is difficult to begin a discussion of new neuroleptics without making a few critical points about traditional schizophrenia treatments. Chlorpromazine, in the early 1950s, was the first drug reported to have specific antipsychotic actions (5). Its action in schizophrenic psychosis was demonstrated to be brisk and potent, partially because of the fully untreated state of these early study participants. Not long after chlorpromazine’s introduction, the mechanism of its antipsychotic action was proposed to be dopamine receptor blockade (6). Many additional pieces of data have continued to support the idea that antipsychotic action of neuroleptics is mediated by dopamine receptor blockade (7–10). Moreover, additional pharmacologic strategies that decrease overall dopamine-mediated neural transmission also possess antipsychotic activity, including dopamine depletion (11) and partial dopamine agonist treatment (12). The obvious inference, therefore, that dopamine system hyperactivity exists in schizophrenic psychosis is still without clear evidence; conversely stated, no data have been discovered and consistently replicated to document an underlying dopaminergic abnormality in schizophrenic illness. Despite this, the dopamine hypothesis of schizophrenic psychosis is still an attractive hypothesis, and considerable new research in this area continues whenever advances in clinical methods are introduced (13–16).

Clinical use of the traditional neuroleptics to treat schizophrenia has been ubiquitous. Several different chemical classes with many congeners in each class exist (17). The primary antipsychotic activity of all of these compounds is present and is of similar magnitude, dose-adjusted for potency, even though side effect profiles differ across compounds (18). Low-potency phenothiazines have high cardiovascular effects and sedation, whereas the high- potency compounds display potent parkinsonian motor effects and akathisia. All compounds currently marketed for psychosis block the D₂ family of receptors, without exception, albeit with widely different affinities. Whether this will remain the only neurochemical mechanism for antipsychotic action is being challenged, but not yet successfully.

Since the original hypothesis was introduced that neuroleptic action is mediated through dopamine receptor blockade, much new knowledge has been gained about dopaminergic systems in human brain. Multiple dopamine receptor subtypes were first defined pharmacologically (19) and then identified, cloned, and sequenced (20–22). Much work on structure-function relationships is ongoing. The D₂ family of dopamine receptors, not the D₁ families, mediate antipsychotic response (23). Thus the descriptive studies of D₂-family receptor binding and receptor message expression are generally considered relevant to schizophrenia research and can often suggest receptor function (15,2,24–26). Selective function, interactions, and clinical relevance of the D₂-family subtypes (D₂, D₃, and D₄) remain to be fully described (23). The partial dopamine agonists at each of these D₂ receptors are being explored for their antipsychotic potential and optimal levels of intrinsic activity (27).

In vivo receptor occupancy imaging in schizophrenia has been used to estimate the receptor occupancy of different neuroleptic treatment regimens (28). The idea of identifying a proxy measure of antipsychotic action in humans has always been attractive. In individuals treated with traditional neuroleptics for psychosis, occupancy of the striatal dopamine receptors at a level of 70% to 95% has been associated with antipsychotic action using positron emission tomography (PET) and a ¹¹C-labelled dopamine receptor ligand. Occupancy of dopamine receptors for this purpose has been quantified in striatum because of the strong binding signal in this area, whereas dopamine receptors are known to exist (even though at lower densities) in other areas thought to be more relevant to treatment response. Now new high-affinity ligands are being developed to detect extrastriatal dopamine receptor occupancy. Which of these regional occupancies is related to antipsychotic response, and at what level, still remains an open issue. Not only dopamine occupancy with antipsychotics but also the distribution of their functional actions has been studied recently (29,30). This measure provides information not about the regional binding of the drug, but about the functional actions both the primary and the downstream associated with its receptor occupancies. Holcomb and others showed that the traditional antipsychotic haloperidol increases regional glucose metabolism in the striatum and thalamus and decreases regional metabolism in the anterior cingulate and the middle and inferior frontal cortices (29). It would be presumed that these regional metabolism changes might be associated with the antipsychotic action, a presumption which can be further tested.

It is widely held by treating physicians that traditional neuroleptics have potent therapeutic action on positive psychotic symptoms, incomplete action on negative symptoms, and little action on cognitive symptoms (2). In the original chlorpromazine efficacy trials (with neurolepticnaive schizophrenic persons), traditional neuroleptics had a 60% good to excellent efficacy, with 20% of the cases demonstrating moderate efficacy, and the remaining 20% being patients unaffected or worsened by treatment (17). Thus the traditional neuroleptics, while having some disadvantages, will undoubtedly continue in clinical use for a time because of their efficacy and their economic advantage. Low-dose use of the traditional compounds will reduce the overall incidence of motor side effects. Despite this, it is easy to say that far better clinical treatments are needed for schizophrenia. Side effect-free treatments, with broader and more extensive actions on cognition and negative symptoms, are required.

Clozapine and Risperidone: New Neuroleptics

Clozapine was the first of the new drugs developed for schizophrenia. It is of interest to note that neither clozapine nor its application in schizophrenia are at all new. Clozapine was first introduced into the European market in the 1970s and was used rather widely. It was not for several more years, when agranulocytosis was firmly associated with its use, that clozapine was more conservatively used in Europe and, because of that side effect, not approved in the US until recently. Even after the agranulocytosis was discovered and before 1988, however, the drug still continued to be used in Europe in inpatient settings with suitable monitoring and in the US as a compassionate use drug (in New York state). This experience with the drug, however slight, was enough to suggest to US clinicians that it might have unique antipsychotic actions. This provided the rationale for the multicentre Sandoz Study, which demonstrated clozapine to be superior to chlorpromazine in treatment resistant schizophrenic inpatients (31). This study was pivotal not only for its direct treatment implications for schizophrenia but because it promised new directions and created enthusiasm for new treatments. The study itself showed that clozapine has a larger therapeutic effect than chlorpromazine on positive symptoms of psychosis and also has greatly reduced motor side effects. Subsequent studies have confirmed this differential efficacy of clozapine in neuroleptic nonresponders (32–34).

The pharmacologic characteristics of clozapine have provided many new leads for subsequent antipsychotic drug development. For future research, it may be fortuitous that clozapine’s pharmacologic characteristics are numerous. Clozapine is a dibenzodiazepine. It demonstrates significant affinity for the D₂-family, alpha₁, alpha₂, 5-HT₂, 5-HT_1A and _1C, the histaminic, and both nicotinic and muscarinic cholinergic receptors. Its broad antagonism of cholinergic, monoaminergic, and histaminic receptors extends throughout cortical and subcortical structures (35). Which of these affinities or which combination could account for its clinical properties, if any, remains unknown. Possibly, its special clinical properties are based on its combinations of broad receptor actions (36). In addition to its affinity profile, clozapine has another special characteristic in that it selectively produces depolarization block in chronically treated rats (in mesolimbic [A10] dopamine neurons only, not in the nigrostriatal [A9] neurons). This characteristic, simplistically put, suggests a regional selectivity of its action directed toward neurons that may modulate affect and cognition (the mesolimbic, A10 tract) but not toward those which modulate movement (the nigrostriatal, A9 tract). It has been proposed, with some face validity, that this regional selectivity accounts for its lower motor side effect profile. Consistent with this, clozapine does not cause catalepsy in rats, nor does it cause dystonia in neuroleptic sensitized monkeys (37). In addition, with very chronic treatment in laboratory rats, it fails to induce the oral dyskinesia syndrome produced by traditional neuroleptics (38). Several of these characteristics have been mimicked by subsequently developed antipsychotic drugs. It may be that the conclusive identification of clozapine's mechanism of action in neuroleptic resistant schizophrenia will need to await the clinical identification of more selective compounds with the same clinical effects.

Risperidone was introduced to the North American market in 1992. It has affinity for the D₂-family, the 5HT₂ and the alpha₁ noradrenergic receptors. Its development was based on the observation that haloperidol, when it was combined with ritanserin, a potent 5-HT₂ antagonist, produced an enhanced antipsychotic effect with reduced motor side effects (39,40). Considerable evidence already supported the idea that dopaminergic and serotonergic systems positively interact in the mammalian brain (41–43) and could improve antipsychotic response (31,33,44,45). Clinical studies have tended to bear out this idea (46–48), but they have yet to provide a definitive answer. With risperidone, motor side effects are clearly reduced at lower dose levels. In the pharmaceutically sponsored multicentre trials, risperidone showed full antipsychotic efficacy at 6 mg/day and a significantly favourable motor side effect profile compared with haloperidol (47). Since its approval, however, additional studies have been pursued to test more clearly any additional advantages, particularly studies in the elderly.

The opportunity that has existed over the last half decade to use these drugs clinically, as well as to study them in academic settings, has generated several widely accepted clinical generalizations: clozapine, while difficult to use by reason of its broad side effect profile, continues to contribute a significant and unique advantage with respect to antipsychotic efficacy to the treatment of treatment resistant schizophrenic persons. Additional studies with clozapine have continued to report a 30% to 60% response advantage in the treatment resistant person (32–34). The drug also demonstrates an advantage in treating schizophrenia with tardive dyskinesia (49). Risperidone is routinely used within the dose range known to produce a full antipsychotic action and low motor side effects, hence it has a significant advantage over traditional neuroleptics. Many schizophrenic persons have achieved significant improvement with risperidone use. It has, however, proven difficult to use with persons requiring high neuroleptic doses because, at that point, motor side effects appear. In addition, its use in the nonschizophrenic psychotic elderly is broadening. Whether risperidone, like clozapine, provides an advantage in treating treatment resistant psychosis is becoming more doubtful, although the question has not yet been fully answered.

Olanzapine and Sertindole: New Neuroleptics

Two additional new drugs for schizophrenia have been developed in North America over the past 4 to 8 years: olanzapine and sertindole. Both have been considered and approved for use by the American and Canadian authorities and both are expected to provide advantages over existing antipsychotic compounds.

Olanzapine is a thienobenzodiazepine and a congener of clozapine. It has an in vitro receptor affinity binding profile similar to that of clozapine, with affinity for the D₁, D₂-family, histamine H₂, alpha₁-adrenergic, 5-HT₂, 5-HT_2C, 5-HT₆, and 5-HT₇ serotonergic and M1 muscarinic receptors; its affinity at all of these sites is generally somewhat greater than that of clozapine, as is its clinical potency (50). Neurochemically, olanzapine antagonizes dopamine, serotonin, alpa₁-adrenergic, and muscarinic receptor actions (51). Electrophysiologically, olanzapine mimics the action of clozapine in its differential activity on A10 (mesolimbic) dopamine neurons to the exclusion of A9 neurons, at least in the lower dose ranges (52,53). Consistent with this action is the regionally selective c-fos immediate early gene (IEG) activation found with olanzapine, localized chiefly to nucleus accumbens shell, medial frontal cortex, with only slight activation in dorsal striatum (54). Consistently, the behavioural pharmacology of olanzapine in animals resembles the profile of clozapine: both block dopamine agonist induced behaviours, inhibit the conditioned avoidance response, but fail to produce catalepsy (55,56). Because it is still unclear which of these preclinical characteristics are important for therapeutic drug effects, it is impossible to exclude any of these actions as unimportant to a unique clinical effect.

Several large, multicentre, controlled studies have been carried out in acutely psychotic schizophrenic inpatients, including a 6-week dose comparison study (placebo; 1 and 10 mg/day olanzapine) (57); a 6-week dose range comparison study (olanzapine 2.5 to 7.5 mg/day, 7.5 to 12.5 mg/day and 12.5 to 17.5 mg/day; haloperidol 10 to 15 mg/day; and placebo) (58); and a large, 6-month haloperidol (10 to 20 mg/day) versus olanzapine (10 to 20 mg/day) comparator study in schizophrenic outpatients. Olanzapine, by the time of its National Drug Administration submission, had been tested in over 2500 persons and in many more by launch, to comprise a substantial safety data base. The efficacy studies showed only minor outcome differences and were highly consistent across studies with respect to magnitude and extent of clinical outcome. The antipsychotic action of olanzapine on positive symptoms in doses of 10 to 20 mg/day is potent and of at least equivalent magnitude to that of haloperidol (10 mg/day) (59). Onset of this antipsychotic action with olanzapine is brisk, with measurable symptom decreases at one week after drug initiation. A dose of 10 mg/day seems to be the lowest effective dose, with upper doses, especially for difficult-to-treat persons, recommended up to 25 mg/day. Notably, therapeutic doses can be initiated with the first dose without any need for dose titration. This pharmacologic characteristic will be advantageous for a fast onset of antipsychotic action.

Olanzapine had a greater therapeutic effect on negative psychotic symptoms in schizophrenia than placebo in 2 of its studies (57,58) and greater than haloperidol in the large, 6-month study. In response to the question of whether this advantage on negative symptoms is “merely” secondary (that is, secondary to its lower motor side effects, antidepressive action, and/or antipsychotic action) or is a primary effect, a Path analysis was done. This analysis suggests that after covarying the negative symptom response with these 3 other olanzapine effects, there remained a significant difference in the effect of olanzapine, presumably its primary antinegative symptom action (60).While this analysis is insufficient to establish a firm basis for such a position, these data surely recommend a specifically designed study in predominantly negative symptom persons with schizophrenia to test this effect directly. Whether this action on negative symptoms is primary or secondary, the clinical advantage to the schizophrenic person remains significant.

Olanzapine produces low or no acute motor side effects like Parkinsonism in the dose range of 10 to 20 mg/day. In the ongoing clinical studies, olanzapine has produced similar ratings on the Simpson Angus Scale (SAS), as has placebo—markedly different ratings than those produced by haloperidol. Olanzapine produces some akathisia, although mild, and requires some anticholinergic drug use, although at near-placebo levels (57,58). Overall, the extrapyramidal side effects with olanzapine are very mild, if present at all.

The longer-term treatment trials comparing olanzapine with haloperidol confirm a lower incidence of treatment emergent dyskinesias with olanzapine than with haloperidol, suggesting the potential for a lower incidence of tardive dyskinesia with olanzapine. Although these studies were not prospectively designed for this purpose and tested previously treated patients, they are encouraging. The haloperidol rate of treatment emergent dyskinesias in this analysis was 4.5% over the more than 200 study days, a figure that matches rather closely the 5% annual incidence of tardive dyskinesia found in properly designed prospective studies; in comparison, the 1% incidence of treatment emergent dyskinesia with olanzapine is striking. This observation, coupled with several preclinical studies generating consistent results in laboratory rats (61) and monkeys (37), suggests that olanzapine may well be associated with lower tardive dyskinesia potential than traditional antipsychotics.

The other side effects of olanzapine are mild but do include transient evidence of hepatotoxicity, mild prolactin elevation, and weight gain.

In summary, olanzapine is a clozapine congener that has been demonstrated to have potent antipsychotic actions on positive symptoms, possibly an advantage in treating negative symptoms, and a clear motor side effect advantage over traditional antipsychotics. Olanzapine produces almost none of the usual motor or cardiac side effects associated with neuroleptics, no Parkinsonism, and little akathesia. Its advantage here has been firmly demonstrated.

Sertindole is a phenylindol derivative discovered and patented in Denmark by Lundbeck and clinically developed in North America by Abbott Laboratories Ltd. It has a relatively restricted receptor profile compared with clozapine and olanzapine, with nanomolar affinity to the D₂ dopamine receptors, the 5-HT_2A and 5-HT_2C serotonin receptors, and to the alpha₁-adrenergic receptor; it lacks affinity for the D₁ dopamine family, other serotonergic receptors, histamine, or acetylcholine receptors (62). In addition to this affinity profile, this drug was prospectively designed as a D₂-family antagonist that would produce selective depolarization blockade with chronic treatment in only mesolimbic (A10) dopamine neurons, without affecting the nigrostriatal neurons (A9) (63). Thus, prior to clinical testing, sertindole was hypothesized to be associated with low (or no) motor side effects. Other animal behavioural characteristics of sertindole were consistent with this prediction as well: its lack of catalepsy, lack of motor sedation, and relative resistance to dystonias in rats with chronic treatment (64). In addition, sertindole stimulates c-fos IEG expression in a central nervous system distribution characteristic of clozapine: in the medial prefrontal rat cortex, the shell of the nucleus accumbens, but not in the dorsal striatum (65). It is curious that sertindole shares these regional action characteristics with both clozapine and olanzapine despite its vastly different receptor affinity profile and chemical structure.

Abbott has carried out several controlled trials in the US, including first, a 6-week dose comparison study (8, 12, and 20 mg/day versus placebo) (66); second, an 8-week, 7-arm study with a dose range of sertindole (12, 20, 24 mg/day tested against a dose range of haloperidol (4, 8, 16 mg/day) (67); and third, an 8-week, 4-arm study (sertindole 20 mg/day and 24 mg/day, haloperidol 10 mg/day, and placebo). In addition, Lundbeck has carried out a multicentre European trial comparing a dose range of sertindole (8 to 24 mg/day) to haloperidol at 10 mg/day (68). Several long-term extension studies augment data from a 12-month, random assignment, sertindole (24 mg/day) versus haloperidol (10 mg/day) study designed to demonstrate maintenance efficacy (69).

Clinical efficacy data from all of the sertindole studies are remarkably consistent. Placebo response rates are low. The lowest effective dose appears to be 12 mg/day. Response to 20 mg/day and 24 mg/day appears to be equivalent, thus establishing the upper effective dose. Between these doses, sertindole shows robust antipsychotic efficacy at least equivalent to haloperidol. Across this dose range, it has proved impossible to demonstrate a dose response relationship with positive symptom reduction (67) or a dose–plasma level relationship of clinical importance in the large multicentre designs (70). Because the second study included a range of haloperidol doses (the first contemporary efficacy study to do so), it has been demonstrated that this is also true for haloperidol, now the prototypical traditional neuroleptic. Between the doses of 4 mg/day and 16 mg/day, haloperidol shows a significant antipsychotic action on total Brief Psychiatric Rating Scale and Positive and Negative Syndrome Scale scores. Only a slight suggestion of a dose effect becomes apparent from analyzing positive symptom response where 4 mg/day haloperidol lacks statistical significance compared with placebo (even though the magnitude of this effect is considerable).

The effect of sertindole on negative symptoms is significant at the 20 mg/day dose in the second study and at the 24 mg/day dose in the third study, but in neither study at any dose does haloperidol produce an antinegative symptom effect. For sertindole, as with olanzapine, a Path analysis was done in an attempt to control for potentially confounding effects of sertindole on positive symptoms, depression, and motor side effects. This analysis showed that the direct effect of sertindole on negative symptoms after removing these confounds was significantly greater than placebo in all drug groups (71).

Since motor side effects were predicted, a priori, to be reduced or nonexistent with sertindole, the clinical trial designs were constructed to examine this question. Results of all trials showed that sertindole produces motor side effects at the same rate as placebo, both for Parkinsonism (SAS score) and for akathisia (Barnes Akathisia Scale score), a rate significantly lower than found for haloperidol. In the second study, which utilized dose ranges of both drugs, sertindole at all its doses produced placebo-level motor side effects, showing lower extrapyramidal side effects than even the lowest (4 mg/day) dose of haloperidol.These clinical data are consistent with primate (37) and rat (61) data with sertindole from the laboratory, suggesting low or no extrapyramidal side effects or dyskinesias at clinically relevant doses.

Sertindole shows other side effects of potential relevance. Nasal congestion is caused by sertindole; although benign, it can produce discomfort. In 30% of treated men, sertindole can produce low-volume ejaculation. While sertindole does not affect libido or ejaculation itself, the low-volume ejaculate can cause concern in some patients, particularly if they are not warned. This side effect is relatively benign, reversible, and sometimes shows tolerance with continued treatment. The side effect receiving more attention has been the tendency of sertindole to prolong the QT interval (72). This side effect has been intensively studied because of its potential relevance to safety. The QT interval prolongation is of the magnitude of approximately 15 to 20 msec, a 5% increase over baseline; the percentage of treated persons with QT above 500 msec is less than 1%. Adjusting QT interval for heart rate (sertindole slightly increases cardiac rate) using the QT_c calculation increases the QT_c change from baseline slightly and the number of persons with QTc over 500 to 3%. QT_c values were increased most in the lowest quartile of baseline QT_c persons (for example, 37 msec) and least in the highest baseline quartile of QT_c persons (for example, less than 5 msec). The danger of prolonged QT or QTc intervals is the possibility of tachyarrhythmias, the most conspicuous one of which is torsades de pointe. In the total sertindole US safety data base, however, no person with QT prolongation was detected to have such an arrhythmia, and more significantly, no one with QT prolongation experienced clinical symptoms of a tachyarrhythmia (for example, syncope). The conclusion of a panel of expert cardiologists who reviewed these data was that sertindole has a favourable cardiovascular risk profile for its psychosis benefits. Prescribing physicians should nevertheless remain aware of the QT prolongation and should probably avoid sertindole use in individuals with preexisting heart disease.

In summary, sertindole is a potent antipsychotic compound. It treats positive symptoms like haloperidol and may have an advantage in treating negative symptoms. Its motor side effects match those of placebo, with no detected Parkinsonism or akathisia in these multicentre studies.

Future Antipsychotics

In addition to these new antipsychotic drugs, which are already becoming available for use, 2 additional compounds are in final phases of testing, quetiapine (Seroquel) and ziprasidone (Zeldox).

Quetiapine is a dibenzothiazepine derivative from Zeneca Pharmaceuticals. Its structure is reminiscent of that of clozapine. Moreover, its binding characteristics (except for its lack of muscarinic cholinergic affinity) parallel those of clozapine. Furthermore, the whole preclinical pharmacologic profile is very similar to clozapine’s profile, including dose selectivity for A10 dopamine neurons with chronic treatment (73). Quetiapine has been studied in phase II and phase III clinical efficacy and safety designs. Its antipsychotic efficacy is superior to placebo; it produces changes in mental status that are clinically meaningful and statistically significant. Quetiapine produces decrements in both positive and negative symptoms in schizophrenia (74). A single comparator study with chlorpromazine suggests comparable efficacy with traditional neuroleptics (75), but too few comparator efficacy studies are available to be certain of comparative efficacy. The minimal effective dose of quetiapine is 250 mg/day; possibly 300 to 450 mg/day is optimal. Doses up to 750 mg/day have been tested, however, with positive results. Motor side effects are extremely low. Other side effects include somnolence, mild agitation, and dry mouth.

Ziprasidone is a new antipsychotic drug from Pfizer. Its development is several months behind that of quetiapine, with its submission to the US FDA expected in the first quarter of 1997. The drug is relatively unique with respect to its receptor binding profile. The drug binds significantly to D₂ but not to D₁ receptors; its serotonin affinities are for the 5-HT_2A, 5-HT_2C, 5-HT_1A, and 5-HT_1D receptors. The alpha₁-noradrenergic affinity is relatively high, but its alpha₂ and histamine H₁ affinities are low. The drug also has moderate serotonin and noradrenergic reuptake blockade properties. The clinical profile is one of a potent antipsychotic that has both positive symptom and negative symptom efficacy. Its side effect profile is very favourable with respect to motor side effects, including the production of little, if any, Parkinsonism and akathisia. Moreover, other side effects are manifest at a low rate or are nonexistent: specifically, there are no cardiac, hematologic, or hepatic changes or weight gain. Little information on ziprazidone is published, but the literature can be expected to increase over the next year, when phase III studies finish. It will be particularly interesting to link clinical actions and side effects (or lack of side effects) with the unique binding profile of ziprasidone.

Conclusions

The need for new drugs to treat schizophrenia clearly exists. Studies with clozapine (and now with the other new drugs) have demonstrated that new treatments can provide additional antipsychotic efficacy. Traditional antipsychotic drugs, while they treat some patients and some symptoms successfully, leave many schizophrenic patients partially or extensively untreated. Some symptom complexes are characteristically inadequately treated by neuroleptics, like social and cognitive deficits; negative symptoms can even worsen with neuroleptics. Moreover, side effectsof current neuroleptics are burdensome and sometimes serious. The long-lasting neuroleptic-induced motor side effect of tardive dyskinesia is especially limiting.

Which are the viable antipsychotic strategies to pursue now and into the 21st century? What are the main candidatedrugs in those areas? These are major questions still to be addressed. Antipsychotic drug development strategies have advanced in 4 major directions: one is the refined clozapine like strategies, which are discussed in this review, but the other directions—novel antipsychotic approaches, experimental strategies based on speculative hypotheses of psychosis and schizophrenia and more serendipity cannot be overlooked as potential high-yield areas (76). The last item deserves separate mention. Serendipity has contributed pivotally to many advances in somatic treatments in psychiatry. For none of the major psychiatric diseases, including schizophrenia, is pathophysiology known. Without known pathophysiology to direct drug discovery, informed and creative serendipity becomes important. A clue to successful antidepressant treatment came with the observation of antidepressant benefit with isoniazid during tuberculosis treatment. The direction of antipsychotic treatment was fixed for a half century by the observation of selective antipsychotic activity of a previously known sedative, chlorpromazine. Other examples could also be raised to support the importance of critical clinical observation in new drug development. Recognizing the role of serendipity does not diminish the ultimate goal of creating rational strategies for new drug development. Rational therapies can be developed based on already successful strategies but not yet on pathophysiology.

Current research has produced new antipsychotic compounds of real use in the treatment of schizophrenia. These have characteristics based on aspects of the traditionalneuroleptics as well as incorporating new dimensions. Opportunity now exists to identify still more unique and improved antipsychotic strategies. The rate of drug discovery today has increased enormously from a decade ago, promising new therapeutic agents for today and for tomorrow. This increased rate of discovery has been due to the efforts and attention of academic scientists, to pharmaceutical research, and to the rapid growth in basic neuroscience knowledge. It is a rich time for antipsychotic drug research and holds considerable promise for even more new treatments and surely more new information in the not-too-distant future.

References

1. Tamminga CT. The neuropsychiatry of schizophrenia. In: Yodofsky S, editor. The third edition of the APA textbook of neuropsychiatry. Forthcoming.

2. Carpenter WT Jr, Buchanan RW. Schizophrenia. N Eng J of Med 1994;330:681–90.

3. Carpenter WT, Buchanan RW, Kirkpatrick B, Tamminga CA, Wood F. Strong inference, theory testing and the neuroanatomy of schizophrenia. Arch Gen Psychiatry 1993;50:825–31.

4. Weinberger DR. Implications of normal brain development for the pathogenesis of schizophrenia. Arch Gen Psychiatry 1987;44:660–9.

5. Delay J, Deniker P. Le traitement des psychoses par une méthode neurolytique dérivée de l'hibernothérapie. In: Congrès des Médecins Aliénistes et Neurologistes de France. Luxembourg; 1952. p 497–502.

6. Carlsson A, Lindquist M. Effect of chlorpromazine and haloperidol on formation of 3–methoxytyramine and normetanephrine in mouse brain. Acta Pharmacologica and Toxicologica 1963;20:140–4.

7. Snyder SH, Creese I, Burt D. The brain's dopamine receptor: labeling with [3H]dopamine and [3H]haloperidol. Psychopharmacology Communications 1975;1:663–73.

8. Seeman P, Lee T, Chau–Wong M, Wong K. Antipsychotic drug doses and neuroleptic/dopamine receptors. Nature 1976;261:717–9.

9. Seeman P, Ulpian C, Bergeron C, Riederer P, Jellinger K, Gabriel E, and others. Bimodal distribution of dopamine receptor densities in brains of schizophrenics. Science 1984;225:728–31.

10. Pickar D, Labarca R, Doran AR. Longitudinal measurement of plasma homovanillic acid levels in schizophrenic patients. Arch Gen Psychiatry 1986;43:669–76.

11. Walinder J, Skott A, Carlsson A, Roos BE. Potentiation by metyrosine of thioridazine effects in chronic schizophrenics. Arch Gen Psychiatry 1976;33:501–5.

12. Tamminga CA, Cascella NG, Lahti RA, Lindberg M, Carlsson A. Pharmacologic properties of (–)–3PPP (preclamol) in man. J Neural Transm 1992;88:165–75.

13. Laruelle M, Iyer RN, Al–Tikriti MS, Zea–Ponce Y, Malison R, Zoghbi SS, and others. Microdialysis and SPECT measurements of amphetamine–induced dopamine release in nonhuman primates. Synapse 1997. Forthcoming.

14. Wong D, Wagner Jr HN, Tune LE, Dannals RF, Pearlson GD, Links JM, and others. Positron emission tomography reveals elevated D2 dopamine receptors in drug–naive schizophrenics. Science 1986;234:1558 – 63.

15. Seeman P, Guan HC, Van Tol HHM. Dopamine D4 receptors elevated in schizophrenia. Nature 1993;365:441–5.

16. Seeman P. Dopamine receptors and psychosis. Science and Medicine 1995;5:28–37.

17. Klein DF, Davis JM. Review of antipsychotic drug literature. Diagnosis and drug treatment of psychiatric disorders. Baltimore: Williams & Wilkins; 1969. p 61–3.

18. Tamminga CA, Shirakawa O, Conley RR. Nontarget actions of neuroleptics. In: Brunello N., Mendlewicz J, Racagni G, editors. New generations of antipsychotic drugs: novel mechanisms of action. Volume 4. Basel: Karger; 1993. p 67–76.

19. Kebabian JW, Calne DB. Multiple receptors for dopamine. Nature 1979;277:93–6.

20. Bunzow JR, Van Tol HH, Grandy DK, Albert P, Salon J, Christie M, and others. Cloning of the dopamine D2 receptor. Nature 1988;336:783–7.

21. Sokoloff P, Giros B, Martres MP, Bouthenet ML, Schwartz JC. Cloning of the dopamine D3 receptor. Nature 1990;347:146–51.

22.  Van Tol HH, Bunzow JR, Guan HC, Sunahara RK, Seeman P, Niznik HB, and others. Cloning of the gene for a human dopamine D4 receptor with high affinity for the antipsychotic clozapine. Nature 1991;350:610.

23. Lahti RA, Lahti AC, Tamminga CA. D2–family receptors in schizophrenia: distribution and implications for treatment. Clin Neuropharmacol 1995;18 (3 Suppl):110S–120S.

24. Meador–Woodruff JH, Grandy DK, Van Tol HHM, Damask SP, Little KY, Civelli O, and others. Dopamine receptor gene expression in the human medial temporal lobe. Neuropsychopharmacology 1994;10:239–48.

25. Lahti RA, Robert RC, Tamminga CA. D2–family receptor distribution in human postmortem tissue: an autoradiographic study. Neuroreport 1995;6:2505–12.

26. Lahti RA, Roberts RC, Conley RR, Cochrane EV, Mutin A, Tamminga CA. D2–type dopamine receptors in postmortem human brain sections from normal and schizophrenic subjects. Neuroreport 1996;7:1945–8.

27. Lahti RA, Figur LM, Piercy MF, Ruppel PL, Evans DL. Intrinsic activity determinations at the dopamine D2 guanine nucleotide–binding protein–coupled receptor: utilization of receptor state binding affinities. Mol Pharmacol 1992;42:432–8.

28. Tamminga CA, Dannals RF, Frost J, Wong D, Wagner HN. Neuroreceptor and neurochemistry studies with positron emission tomography in psychiatric illness: promise and progress. In: Oldham JM, Riba MB, Tasman A, editors. American Psychiatric Press, Review of Psychiatry. Volume 12. Washington (DC): American Psychiatric Press; 1993. p 487–510.

29. Holcomb HH, Tamminga CA, Cascella NG, Thaker GK, Medoff DR, Gastineau EA, and others. Functional sites of neuroleptic drug action in human brain: PET/FDG studies with and without haloperidol. Am J Psychiatry 1996;153:141–9.

30. Bartlett EJ, Wolkin A, Brodie JD, Laska EM, Wolf AP. Importance of pharmacologic control in PET studies: effects of thiothixene and haloperidol on cerebral glucose utilization in chronic schizophrenia. Psychiatry Res 1991;40:115–24.

31. Kane J, Honigfeld G, Singer J, Meltzer HY. Clozapine for the treatment–resistant schizophrenic: a double–blind comparison with chlorpromazine. Arch Gen Psychiatry 1988;45:789–96.

32. Wilson WH. Clinical review of clozapine treatment in a state hospital. Hospital and Community Psychiatry 1992;43:700–3.

33. Meltzer HY, Burnet S, Bastani B, Ramirez LF. Effects of six months of clozapine treatment on the quality of life of chronic schizophrenic patients. Hospital and Community Psychiatry 1990;141:892–7.

34. Breier A, Buchanan RW, Irish D, Carpenter WT Jr. Clozapine treatment of outpatients with schizophrenia: outcome and long–term response patterns. Hospital and Community Psychiatry 1993;44:1145–9.

35. Coward DM. General pharmacology of clozapine. Br J Psychiatry 1992;160:5–11.

36. Carlsson M, Svensson A, Carlsson A. Synergistic interactions between muscarinic antagonists, adrenergic and NMDA antagonists with respect to locomotor stimulatory effects in monoamine–depleted mice. Naunyn Schmiedebers Arch Pharmacol 1991;343:568–73.

37. Casey DE. Behavioral effects of sertindole, risperidone, clozapine and haloperidol in cebus monkeys. Psychopharmacology 1996;124:134–40.

38. Gao X–M, Hashimoto T, Cooper TB, Tamminga CA. The dose–response characteristics of rat oral dyskinesias with chronic haloperidol or clozapine administration. J.Neural Transm 1996.

39. Bersani G, Grispini A, Marini S, Pasini A, Valducci M, Ciani N. Neuroleptic–induced extra–pyramidal side effects: clinical perspectives with ritanserin (R 55 667), a new selective 5–HT2 receptor blocking agent. Curr Ther Res 1986;40:492–9.

40. Gelders Y, VandenBussche G, Reyntijens A, Janssen P. Serotonin–S2 receptor blockers in the treatment of chronic schizophrenia. Clin Neuropharmacol 1986;9:325–7.

41. Nicolaou NM, Garcia–Munoz M, Arbuthnott GW, Eccleston D. Interactions between serotonergic and dopaminergic systems in rat brain demonstrated by small unilateral lesions of the raphe nuclei. Eur J Pharmacol 1979;57:295–305.

42. Herve D, Simon H, Blanc G, LeMoal M, Glowinski J, Tassin JP. Opposite changes in dopamine utilization in nucleus accumbens and the frontal cortex after electrolytic lesion of the median raphe in the rat. Brain Res 1981;216:422–8.

43. Rasmussen K, Aghajanian GK. Potency of antipsychotics in reversing the effects of a hallucinogenic drug on locus coeruleus neurons correlates with 5–HT, binding affinity. Neuropsychopharmacology 1988;1:101–7.

44. Claghorn JC, Honigfeld G, Abuzzahab FS, and others. The risks and benefits of clozapine versus chlorpromazine. J Clin Psychopharmacol 1987;7:377–84.

45. Meltzer HY. Clinical studies on the mechanism of action of clozapine: the dopamine–serotonin hypothesis of schizophrenia. Psychopharmacology 1989;99:l8S–27S.

46. Borison R, Pathiraja AP, Diamond BI, Meibach RC. Risperidone: clinical safety and efficacy in schizophrenia. Psychopharmacol Bull 1992;28:213–72.

47. Marder SR. Risperidone, clinical development: North American results. Clin Neuropharmacol 1992;15:92A–93A.

48. Chouinard G, Jones B, Remington G, Bloom D, Addington D, MacEwan GW, and others. A Canadian multicenter placebo–controlled study of fixed doses of risperidone and haloperidol in the treatment of chronic schizophrenic patients [erratum]. J Clin Psychopharmacol 1993;13:149.

49. Tamminga CA, Thaker GK, Moran M, Kakigi T, Gao X–M. Clozapine in tardive dyskinesia: observations from human and animal model studies. J Clin Psychiatry 1994;55:102–6.

50. Bymaster FP, Calligaro DO, Falcone JF, Marsh RD, Moore NA, Tye NC, and others. Radioreceptor binding profile of the atypical antipsychotic olanzapine. Neuropsychopharmacology 1996;14:87–96.

51. Bymaster FP, Hemrick–Luecke SK, Perry KW, Fuller RW. Neurochemical evidence for antagonism by olanzapine of dopamine, serotonin, a1–adrenergic and muscarinic receptors in vivo in rats. Psychopharmacology 1996;124:87–94.

52. Stockton ME, Rasmussen K. Electrophysiological effects of olanzapine, a novel atypical antipsychotic, on A9 and A10 dopamine neurons. Neuropsychopharmacology 1996;14:97–104.

53. Skarsfeldt T. Differential effects of repeated administration of novel antipsychotic drugs on the activity of midbrain dopamine neurons in the rat. Eur J Pharmacol 1995;281:289–94.

54. Robertson GS, Fibiger HC. Effects of olanzapine on regional c–fos expression in rat forebrain. Neuropsychopharmacology 1996;14:105–10.

55. Moore NA, Tye NC, Axton MS, Risius FC. The behavioral pharmacology of olanzapine, a novel "atypical" antipsychotic agent. J Pharmacol Exper Ther 1992;262:545–51.

56. Skarsfeldt T. Differential effect of antipsychotics on place navigation of rats in the Morris water maze: a comparative study between novel and reference anti–psychotics. Psychopharmacology 1996;124:126–33.

57. Beasley CM Jr, Sanger T, Satterlee W, Tollefson G, Tran P, Hamilton S. Olanzapine HGAP Study Group. Psychopharmacology 1996;124:159–67.

58. Beasley CM Jr, Tollefson G, Tran P, Satterlee W, Sanger T, Hamilton S. Olanzapine versus placebo and haloperidol: acute phase results of the North American double–blind olanzapine trial. Neuropsychopharmacology 1996;14:111–23.

59. Keck PE, McElroy SL. Olanzapine: a novel antipsychotic medication. The Journal of New Developments in Clinical Medicine 1996;14:63–78.

60. Tollefson GD, Sanger TM. Negative symptoms: a path–analytic approach to the double–blind, placebo– and haloperidol–controlled clinical experience with olanzapine. Am J Psychiatry. Forthcoming.

61. Gao X–M, Sakai K, Tamminga CA. Chronic olanzapine or sertindole treatment results in reduced oral dyskinesias in rats compared to haloperidol [abstract]. Neuroscience Abstracts 1997 22:1087.

62. Sanchez C, Arnt J, Dragsted N, Hyttel J, Lembol HL, Meier E, and others. Neurochemical and in vivo pharmacological profile of sertindole, a limbic–selective neuroleptic compound. Drug Development Research 1991;22:239–50.

63. Bøgeso KP, Andersen K, Arnt J, Frederiksen K, Hyttel J, Perregaard J, and others. Design of atypical antipsychotic drugs. In: Fog R, Gerlach J, Hemmingsen R, editors. Schizophrenia: an integrated view. Copenhagen: Munksgaard; 1995. p 361–75.

64. Jackson DM, Johansson C, Lindgren L. Dopamine receptor antagonism blocks amphetamine and phencyclidine–induced motor stimulation in rats. Pharmacol Biochem Behav 1994;48:465–71.

65. Fink–Jensen A, Kristensen P. Effects of typical and atypical neuroleptics on Fos protein expression in the rat forebrain. Neurosci Lett 1994;182:115–8.

66. VanKammen D, McEvoy J, Targum S, Kardatzki D, Sebree TB. A randomized, controlled, dose–ranging trial of sertindole in patients with schizophrenia. Psychopharmacology 1996;124:168–75.

67. Zimbroff DL, Kane JM, Tamminga CA, Daniel DG, Mack RJ, Morris DD, and others. A controlled, dose–response study of sertindole and haloperidol in schizophrenia. Am J Psychiatry 1997. Forthcoming.

68. Hale A, vander Burght M, Wehnert A, Sloth–Nielsen M. A European dose–range study comparing the efficacy, tolerability, and safety of four doses of sertindole and one dose of haloperidol in schizophrenic patients. Presented at the American College of Neuropsychopharmacology Annual Meeting; 1996 December; San Juan, Puerto Rico.

69. Krystal J, D'Souza DC, Holgate KL, Staser JA, Silber CJ, Mack RJ. Sertindole: a multi–center, one year, haloperidol controlled trial assessing the long–term safety, efficacy and quality of life in stable schizophrenic patients. Presented at the American College of Neuropsychopharmacology Annual Meeting; 1996 December; San Juan, Puerto Rico.

70. Granneman GR, Wozniak P, Ereshefsky L, Silber C, Mack R. Population pharmacokinetics of sertindole during long–term treatment of patients with schizophrenia. Presented at the American College of Neuropsychopharmacology Annual Meeting; 1996 December; San Juan, Puerto Rico.

71. Tandon R, Silber C, Mack R. The action of sertindole on negative symptoms in schizophrenia. American College of Neuropsychopharmacolgy Annual Meeting; 1996 Dec; San Juan, Puerto Rico.

72. Swann A, Daniel D, Silber C, Mack R. Long term cardiovascular safety of sertindole. Presented at the American College of Neuropsychopharmacology Annual Meeting; 1996 December; San Juan, Puerto Rico.

73. Goldstein JM, Arvanitis LA. ICI 204,636 (seroquel): a dibenzothiazepine atypical antipsychotic; review of preclinical pharmacology and highlights of phase II clinical trials. CNS Drug Reviews 1995;1:50–73.

74. Borison RL, Arvanitis LA, Miller BG, US Seroquel Study Group. ICI204636, an atypical antipsychotic: efficacy and safety in a multicenter, placebo–controlled trial in patients with schizophrenia. Clin Psychopharmacol1996;16:158–69.

75. Hirsch SR, Link CGG, Goldstein JM, Arvanitis LA. ICI 204,636: a new atypical antipsychotic drug. Br J Psychiatry 1996;168:45–56.

76. Lahti AC, Lahti R, Tamminga CA. The new antipsychotics and experimental therapeutics. In: Breier A, editor. Advances in neuroleptic drugs. Washington (DC):American Psychiatric Press. Forthcoming.

Résumé

Objectif :Dans cet aperçu, on présente les aspects de la pharmacologie clinique et préclinique des nouveaux antipsychotiques en insistant sur les caractéristiques précliniques des médicaments qui pourraient laisser présager des actions cliniques souhaitables.

Méthode : La littérature pertinente et les données publiées ont fait l'objet d'analyses et de résumés.

Résultats : Les neuroleptiques classiques sont très efficaces pour traiter les symptômes positifs de psychose, mais ils peuvent causer des effets secondaires moteurs graves.  La clozapine ne provoque aucun effet secondaire moteur, et elle entraîne une action antipsychotique unique en son genre.  La rispéridone produit également de légers effets secondaires, mais seulement à des doses relativement faibles.  On a lancé l'olanzapine et la sertindole depuis peu.  Ces deux médicaments ont une puissante action antipsychotique, et ils réduisent beaucoup les effets secondaires moteurs; chacun d'eux présente également des avantages éventuels par rapport aux symptômes négatifs.