Summary Basis of Decision for Zeldox

3.3.1 Pharmacodynamics

Ziprasidone is a member of the atypical antipsychotic class, with typical dopamine and serotonin binding (antagonist at D₂, 5-HT_2A, and 5-HT_1D, receptors; agonism at 5-HT_1A receptors). Antagonism was shown at histamine H₁ and α₁ adrenergic receptors, associated with somnolence and orthostatic hypotension, respectively.

Four Positron Emission Tomography (PET) studies were conducted in male volunteers. The studies indicated that ziprasidone occupied both the brain D₂ receptors and 5-HT₂ receptors. The receptor occupancy was high and was related to the serum levels of the drug.

3.3.2 Pharmacokinetics

Absorption

Dose-proportionality was demonstrated in the 20 to 80 mg BID (twice a day) dose range. Food greatly enhanced the bioavailability of ziprasidone.The mean absolute bioavailability of ziprasidone when administered as a 20 mg capsule with food was approximately 60%. Drug exposure values (AUC and C_max) were higher when the drug was given with food compared to administration two hours after food. The Product Monograph states that the drug should be taken with food.

Distribution

The recommended dosing of twice daily generally leads to attainment of steady state within 1-3 days.

Serum protein binding of ziprasidone was very high (>99%) in humans. Ziprasidone binds to both α₁-acid glycoprotein and serum albumin. The in vitro plasma protein binding of ziprasidone was not affected by either propranolol (a drug that exhibits high binding to α₁-acid glycoprotein or by warfarin (a drug that exhibits high binding to serum albumin). Similarly, ziprasidone did not affect the plasma protein binding of either warfarin or propranolol.

Metabolism

Ziprasidone was metabolized extensively after oral administration, with only a small amount (<5%) excreted as unchanged drug in urine and feces. Twelve metabolites were identified. Based on in vivo abundance of excretatory metabolites, the majority of metabolic clearance occurred via reduction by aldehyde oxidase, and cytochrome P450 catalyzed oxidation.

Ziprasidone was not a substrate for the cytochrome enzymes CYP1A2, CYP2C9 and CYP2C19. However, the drug inhibited the metabolism of substrates for CYP2D6 and CYP3A4.

Excretion

Approximately 20% of the ziprasidone dose was excreted in the urine and 66% of the dose was excreted in the feces. The mean terminal phase half-life after multiple dosing was between 6 and 10 hours, with a range of individual values from 3 to 18 hours.

Special Populations

There were no clinically significant differences in the pharmacokinetics of ziprasidone in adults due to gender, age or renal insufficiency.

In patients with hepatic impairment, the drug exposure (AUC) and the half-life were significantly greater compared to subjects with normal hepatic function. Lower doses should be considered for hepatic insufficiency.

Drug Interactions

Pharmacodynamic interactions:
The primary potential safety concern in pharmacodynamic interactions is QT prolongation. Ziprasidone is contraindicated in drugs known to have QT prolongation effect

Pharmacokinetic interactions:
Drug interaction studies were conducted in healthy volunteers; the doses of ziprasidone were generally lower than the clinically recommended doses. The following results were reported:

• Ziprasidone, 40 mg BID, did not affect lithium concentrations or clearance.
• Ziprasidone, 20 mg BID, did not affect plasma levels of the components of an oral contraceptive, namely ethinylestradiol and levonorgestrel.
• In the presence of carbamazepine (200 mg BID), steady-state ziprasidone AUC and C_max decreased significantly. The effect is due to the induction of CYP3A4 by carbamazepine, the enzyme being responsible for the metabolism of ziprasidone.
• Ketoconazole, a potent inhibitor of CYP3A4 increased the AUC and C_max of ziprasidone by approximately 35-40%.
• Neither cimetidine (a non-specific CYP3A4 inhibitor), nor an antacid (Maalox) affected the pharmacokinetics of ziprasidone (a single dose of 40 mg, taken as 2 x 20 mg).
• Ziprasidone did not alter the ratio of dextromethorphan/dextrorphan in extensive metabolizers, indicating that it does not inhibit the metabolism of drugs that utilize CYP2D6.

The potential consequences of interaction with each of carbamazapine and ketaconzole are described in the Product Monograph.

3.3.3 Clinical Efficacy

Clinical Program

Zeldox (ziprasidone) was evaluated for the treatment of schizophrenia in 5 placebo-controlled studies (including 2 to 3 fixed doses of ziprasidone): 4 short-term (4-6 week) trials and one long-term (52-week) trial. In one of the studies, the effects of ziprasidone were compared to placebo and haloperidol. Additional data was requested and provided to assess the safety and efficacy of the drug in 4 longer term studies (8-28 week duration using flexible doses of ziprasidone as well as active comparator arms; e.g. risperidone, amisulpride). In some of the studies, the doses of ziprasidone extended beyond those that would be recommended in the Product Monograph (40 to 80 mg BID), namely doses of 20 and 100 BID were also evaluated.

The pre-marketing development program for ziprasidone included over 5400 patients and/or normal subjects exposed to one or more doses of ziprasidone. Over 4500 patients participated in multiple-dose effectiveness trials, and their experience corresponded to approximately 1733 patient years. A total of 1616 patients were enrolled in the nine clinical studies. Of these, 681 patients took part in short-term studies (4 to 6 weeks), 716 in longer-term studies (8 to 28 weeks) and 219 in the 1-year study. Males consitituted the majority of patients. Patients suffered from chronic or sub-chronic schizophrenia (majority) or schizoaffective disorder (minority).

Several instruments were used for assessing psychiatric signs and symptoms, among them the Brief Psychiatric Rating Scale (BPRS) and the Positive and Negative Syndrome Scale (PANSS). Assessments in regard to overall clinical state of the patient were provided by Clinical Global Impression (CGI) scores. The Scale for the Assessment of Negative Symptoms (SANS) was also employed in some clinical studies.

The rates of completion in ziprasidone-treated patients were comparable among the studies, and were also comparable between ziprasidone and groups treated with standard antipsychotic drugs. In the placebo-treated groups, slightly fewer patients completed the studies.

Efficacy Results

Baseline scores reflected disease severity and the baseline scores were comparable between studies which enrolled patients with similar conditions.

The studies have shown that:

• Ziprasidone is an effective antipsychotic drug that was shown to be statistically superior to placebo in 3 of 4 short-term, fixed-dose trials.
• Ziprasidone maintained a clinical response for up to one year in hospitalized chronic schizophrenic patients. Ziprasidone was significantly superior to placebo in both time to relapse and rate of relapse.
• In the three flexible dose studies, the final doses of ziprasidone ranged between 116 mg/day and 131 mg/day, which corresponded approximately to the 60 mg BID dose. Examination of the distribution of patients that received 40, 60 or 80 mg BID doses (recommended dose range), indicated that approximately 60% of patients were on the low and mid doses of the drug.
• In a 6-week placebo-controlled study, the 20 mg BID (40 mg/day) dose of ziprasidone demonstrated some efficacy in acutely exacerbated patients. In addition, this dose was significantly better than placebo in preventing relapse. Consequently, the Product Monograph indicates that the 20 mg BID dose may be efficacious in some patients.
• In two of the studies (6-week placebo-controlled and 12-week flexible-dose study) ziprasidone was given at a 100 mg BID (200 mg/day) dose. While the drug may be somewhat more efficacious at this dose, the incidence of adverse events is also higher. Thus, the 100 mg BID dose should not be a recommended dose. Nevertheless, the fact that a total of 216 patients received the 100 mg BID dose, contributes to the safety of the drug.

An analysis of the effect of ziprasidone on patients with clinically significant depressive symptoms (Montgomery-Asberg Depression Rating Scale, MADRS) ≥14 was conducted in two multicentre placebo-controlled studies in acute schizophrenia. A statistically significant improvement versus placebo (p < 0.05) in the MADRS was observed in the sub-set of patients with doses at the high end of the recommended range (i.e., 60 mg and 80 mg BID). However, the validity of this scale in patients with schizophrenia is not established. In addition, the baseline depression scores were low, i.e., patients with such scores would not qualify to enter a clinical trial for the treatment of depression.

3.3.4 Clinical Safety

The most common adverse drug reactions (ADRs) associated with the use of ziprasidone in the short-term placebo studies (up to 6 weeks), and observed at a rate at least 2x that of placebo were somnolence (14%), extrapyramidal symptoms (14%), and respiratory tract infection (8%). The Product Monograph includes the ADRs that occurred in the eight pivotal studies (described in section 3.3.3 Clinical Efficacy), as well as ADRs from the post-marketing reports.

The following ADRs are of particular importance for ziprasidone (and are in addition to the standard class concerns for atypical anti-psychotics, such as tardive dyskinesia, neuroleptic malignant syndrome, hyperprolactinemia, priapism).

Rash
Rash occurred at a rate of 4% with ziprasidone vs 3% for placebo, and was found to have an apparent dose relationship. Rash was the most common ADR associated with dropout, including 7 dropouts for ziprasidone patients, as compared to 0 for placebo.

Weight Change
The data provided were inadequate to support a regulatory conclusion that ziprasidone has an advantage over other atypical antipsychotic drugs with regard to weight gain and/or hyperglycemia, or lipid regulation, with the exception of short-term data supporting an advantage over Zyprexa.

QTc prolongation
Review of the pivotal trials concluded that ziprasidone had the potential to prolong QT intervals, and that there were insufficient numbers of patients available in the ziprasidone clinical database for an appropriate evaluation of safety. In response, the sponsor submitted a dedicated QTc study. On review, several deficiencies were identified.

• Both the clinical safety database and the multi-comparator QTc prolongation indicated that administration of ziprasidone within the therapeutic dose range caused a potentially clinically significant plasma level-related prolongation of mean QTc intervals as compared to other atypical antipsychotic drugs. It is well established that a number of drugs which prolong the QTc interval are associated with an increased occurrence of both torsades de pointes and sudden unexplained death. Although torsades de pointes had not been observed in the pre-marketing trials in association with the use of Zeldox at the recommended dose, experience was currently too limited to rule out an increased cardiac risk.
• There are no consistent, reliable clinical warning signs of QT prolongation, nor is routine ECG monitoring regarded as an effective detection procedure; thus, this cardiac risk cannot be managed by early detection.

In response to the deficiencies identified above, additional QTc-related data were submitted: consisting of Dose Response Studies and three years of post-marketing reports from the US and parts of the EU. Data from the Dose Response Studies suggested that there may be a plateau of mean QTc prolongation at twice the maximum recommended dose; the data from the studies and 3 years of post-marketing reports were insufficient to allow a conclusion that the known QTc effect did not translate to increased risk of clinically significant cardiac-related harm compared to other anti-psychotics. At this point, an appeal process was denied by the Office of Science in Health Canada for reasons that additional data was required to clarify the magnitude of risk and to explore the purpoted advantages of ziprasidone.

A four-member appeal panel consisting of external experts was convened to review the data. They concluded that there was sufficient evidence that ziprasidone "demonstrated distinct advantages over the atypical antipsychotics not only in terms of weight gain but Type 2 diabetes and hypertriglyceridemia as well". They further concluded that "the indicated benefits of this compound in terms of weight gain/metabolic disturbance are much more compelling than the more nebulous implications of its risk of QTc prolongation". The panel noted that there was an indication in the post-marketing data that there was the possibility of an increased cardiovascular risk over other antipsychotics but that this risk could not be estimated relative to the other antipsychotics at that time. To help answer this question, they suggested an analysis of the five-year adverse drug reaction (ADR) data and the creation of a registry to create more data.

The sponsor submitted a review of 5-year post-marketing spontaneous data from the FDA Adverse Events Reporting System (AERS) database conducted using a set of heart-related search terms as well as Periodic Safety Update Report information. Small elevations in spontaneous reporting rates of cardiac related adverse events were observed for ziprasidone compared to two other atypical antipsychotics. The observation did not preclude approval, but its inclusion and acknowledgement in the Product Monograph was requested. The text was incorporated as requested with sponsor's reservations about the known limitations of post-marketing spontaneous report data. With this disclosure, along with text as per Health Canada's QT/QTc Guidelines in the Contraindications and Warnings sections of the Product Monograph, Zeldox was granted a Notice of Compliance in August 2007.