Ziprasidone is a newer drug intended for the treatment of the symptoms of schizophrenia. This new drug may have an added benefit of being able to help with some of the difficulties in problem solving and memory that many patients with schizophrenia experience. The present study wants to look at ziprasidone and two other drugs frequently used to treat the symptoms of schizophrenia (risperidone and olanzapine) to see if problem solving and memory get better with ziprasidone treatment. Moreover, we will look at symptoms and how they change with treatment.

Typical neuroleptics (i.e., haloperidol, chlorpromazine) are effective at ameliorating the positive psychotic symptoms of schizophrenia but are less efficacious in the treatment of negative symptoms, and there is limited evidence to support their ability to attenuate the cognitive dysfunction observed in schizophrenia (Meltzer et al. 1999). The primary mechanism through which typical neuroleptics achieve their effect is through dopamine (DA) blockade, but recent data suggest that DA blockade may be associated with diminished cognitive improvement despite effective clinical treatment. For example, in a recent molecular genetics study we have shown that subjects with greater DA availability display better cognitive performance on a task of executive functioning (Malhotra et al. in press). Thus, DA blockade may interfere with potential cognitive improvements associated with antipsychotic drug treatment.

Atypical antipsychotics have a higher 5-hydroxytryptamine-2 (5-HT2) to D2 receptor binding ratio than typical agents, and therefore may be more effective in the treatment of cognitive impairments.Unfortunately, there is limited data on the cognitive properties of the new atypical agent, ziprasidone. In addition to having a high 5-HT2 to D2 receptor binding ratio, like the other atypicals, ziprasidone also has weak anticholinergic effects and minimal activity at muscarinic (M1), histaminergic (H1) and alpha1-adrenergic receptors (Casey, 2001) which may also beneficially influence cognitive performance (Byerly et al., 2001). Therefore, ziprasidone may have a unique ability to improve the cognitive performance of patients with schizophrenia.

Evidence from neuropsychology (Gold, Carpenter, Randolph, Goldberg, & Weinberger, 1997), brain imaging (Buchsbaum et al., 1992) and electrophysiology (Shelley et al., 1996) all converge to implicate impaired working memory (WM) function in schizophrenia. As such, the neural substrates that subsume WM, the temporal course of information flow through this system, and importantly whether ziprasidone intervention can aid in normalization of function, are critical issues in schizophrenia research. In the present study, we propose to integrate:

1. A novel cognitive electrophysiological assessment specifically designed to detect subtle differences in the stages of information processing where WM deficits become manifest.
2. A state of the art computerized neuropsychological battery that assesses WM and other cognitive domains.
3. Positron emission tomography (PET) of dopamine D2 receptor occupancy.

These methods will provide a means to specifically characterize the effects of ziprasidone on cognitive performance and dopamine blockade in patients with schizophrenia. The primary hypotheses to be tested are 1) that ziprasidone treatment will be associated with improvements in WM and, 2) WM performance will be associated with D2 occupancy in ziprasidone treated patients.

Data will be collected in the context of an open label, randomized clinical trial comparing the efficacy of ziprasidone to the atypical agents, olanzapine and risperidone. This trial will compare the effects of ziprasidone with risperidone or olanzapine on positive and negative psychotic symptoms, mood, and side effects, as well as provide the first comprehensive data on the effects of these drugs on information processing, working memory and dopamine D2 receptor occupancy. These pilot data will allow us to test the hypotheses that 1) ziprasidone will be associated with improvements in information processing and working memory. 2) ziprasidone will be associated with improvements in psychotic symptoms and mood. 3) ziprasidone associated improvements in cognition and behavioral symptoms will be at least as significant as those associated with treatment with olanzapine or risperidone. Finally, we will examine the relative D2 occupancies of these drugs using PET

Subjects will be randomized to drug treatment with ziprasidone, olanzapine or risperidone, such that 30 subjects will receive ziprasidone, and 30 will receive either risperidone or olanzapine. Patients who enter the study on risperidone or with a history of risperidone treatment within the past 6 months will be randomized to either ziprasidone or olanzapine. Patients entering on olanzapine or with a history of olanzapine treatment within the past 6 months will be randomized to either ziprasidone or risperidone. Patients on other medications, with no history of olanzapine, risperidone or ziprasidone treatment, will be will be randomized to any of the three drugs. Patients treated with ziprasidone, at any time in the past, will be excluded. The final groups will consist of 30 subjects receiving ziprasidone, and 30 subjects receiving risperidone or olanzapine depending on their treatment history. Target dose for ziprasidone will be 160 mg/d with this dosage achieved within two weeks of initiation of drug treatment. Target dose for olanzapine will be 20 mg/d with this dosage achieved within two weeks of initiation of drug treatment. Target dose for risperidone will be 4 mg/d with this dosage achieved within two weeks of initiation of drug treatment. Dosage will be fixed at the target dose for the remainder of the trial. Patients who cannot tolerate the target dose will continue in the study, if feasible, and maintained at a lower dose. Extra-pyramidal side effects, if any, will be treated with benztropine as needed. Concomitant medications will not be permitted.

Patients will be re-assessed every month following the initiation of treatment for the 3 months following baseline. Subjects will be re-assessed with the information processing assessment, neurocognitive battery , behavioral and side effect ratings, and information about drug dosage and compliance with treatment at each visit. Moreover, subjects will be asked to participate in a PET study of dopamine D2 occupancy at the third month visit . Trained raters blind to patient's drug condition will conduct behavioral and side effect ratings.

The cognitive electrophysiological assessment will employ a parametrically designed A-X Continuous Performance Test (AX-CPT) task with increasing levels of difficulty proven to elicit traditional behavioral measures of WM function such as reaction time (RT) and accuracy (Bates et al, 2000). This paradigm provides a means to evaluate the efficacy of ziprasidone treatment in remitting cognitive dysfunction in patients with schizophrenia. However, accuracy and RT only index the final motor response and do not capture information about antecedent stages of information processing. Visual evoked response potentials (ERP's) will be collected while subjects perform the AX-CPT tasks allowing for assessment of early sensory registration of stimuli (N1), and the time course of subsequent cognitive analysis (P3) (Bates et al, 2000). This methodology will facilitate assessment of the stage of information processing where schizophrenia deficits in WM become manifest. Topographical analysis will assess whether schizophrenia patients display amplitude attenuation over scalp sites correlating to prefrontal cortex while performing the WM tasks, and if functioning improves over the course of ziprasidone intervention.

Visual ERP's will be employed for the following reasons. Auditory P3 attenuation has proven not to change with administration of typical neuroleptics in longitudinal designs (Pfefferbaum et al., 1989). Visual P3, however, may provide a more sensitive measure for detecting changes in illness severity as it is though to be related more to clinical state (Duncan, 1988). The proposed study could possibly reveal improved functioning that has gone undetected in studies employing auditory ERPs. The proposed study would be the first to assess the longitudinal effects of ziprasidone treatment on WM functioning using parametrically altered WM tasks while concurrently obtaining visual ERP's.

The neurocognitive battery will employ tasks of WM, executive functioning, memory, motor function and verbal fluency. This paradigm provides a means to evaluate the effects of ziprasidone treatment in improving cognitive function across multiple domains in patients with schizophrenia.

PET imaging with [carbon-11 (11C)]-raclopride provides an in vivo measure of dopamine D2 receptor occupancy during ziprasidone treatment and will be used to assess the relationship between ziprasidone's D2 occupancy and measures of WM and cognition. It is hypothesized that moderate D2 occupancy is associated with clinical improvement and the degree of occupancy will be correlated with WM performance. The [11C]-raclopride studies will be performed in a subset of 30 patients after twelve weeks of ziprazidone treatment. As [11C]-raclopride studies of the D2 receptor have consistently shown comparable levels of binding in controls and untreated patients with schizophrenia (e.g. Farde et al.1990), the treated patients can be compared to the untreated controls. We are using this approach because it may not be feasible to scan the patients in the unmedicated state. It is extremely difficult to justify a two to four week drug free period that would be needed to assess baseline D2 receptor availability.

The PET studies will be performed at the PET Center at the North Shore University Hospital. On the day of the PET scan an intravenous line will be placed in an antecubital vein for radiotracer administration, and to draw a plasma drug level at the time of scanning. The subject will be positioned in the General Electric (GE) Advance scanner. A fifteen-minute transmission scan will be obtained. Then, 15 millicuries (mCi) of [11C]-raclopride will be injected. Scanning will begin immediately after radiotracer injection and will last for 60 minutes.

Radiotracer: [11C]-raclopride is a relatively selective radiotracer for the dopamine (D2/D3) receptor and is a commonly used radiotracer in normal controls and psychiatric patients (e.g. Smith et al., 1995, Volkow et al., 1994, Farde et al., 1990).

An magnetic resonance (MR) scan will be performed to rule out structural brain pathology, for image registration with the PET scan and correction for the effects of cerebral atrophy. The MR scans will be performed with a GE Signa 1.5 Tesla scanner.

Inclusion Criteria:

• Diagnosis of schizophrenia, schizoaffective disorder or schizophreniform disorder
• Ages 18 - 50
• Ability to provide written informed consent
• Brief Psychiatric Rating Scale (BPRS) total score > 40 and Clinical Global Impressions (CGI) > 4 (moderately ill) and/or intolerant to current antipsychotic drug treatment

Exclusion Criteria:

• History of prior treatment with ziprasidone
• History of medical condition that contraindicates ziprasidone treatment
• Treatment with depot antipsychotic medication in past 3 months
• Current diagnosis of alcohol or psychoactive substance dependence
• Impaired ability to provide written informed consent

Baseline characteristics for all participants prior to randomization are reported because treatment assignment information is not available. That information was not preserved when the orginally planned analyses were not conducted because the enrollment was insufficient for statistically valid results.

This is a small sample.