Open-Label Pilot Study to Examine the Value of Substituting Quetiapine for Benzodiazepines

1.1 Background Numerous patients having depression or anxiety appear to use benzodiazepines chronically and respond incompletely to conventional antidepressants. These patients are more likely to request or incur frequent changes of medication because of non-response, incomplete response or intolerance. The hypothesis of this study is that symptoms of anxiety, depression and insomnia; and indices of psychosocial function will all improve, while BZ use will decrease significantly during a twelve-week trial period of substituting quetiapine for benzodiazepines.

1.2 Rationale for this study The scientific rationale for this derives from a combination of clinical observation and a review of the literature. Anecdotal clinical experience with numerous patients having depression or anxiety suggested to me that patients with chronic benzodiazepine use were less likely to respond or remit with conventional antidepressants and more likely to request or incur frequent changes of medication because of non-response, incomplete response or intolerance. Although most of these patients were not drug abusers and the doses they took were within ranges recommended in the package inserts for these drugs (e.g., 5-20mg/ day of diazepam or equivalent), they were "dependent" in the sense that they took the BZ daily or almost daily, would not give it up in spite of "good responses" (although often with residual symptoms of anxiety or insomnia) to antidepressants or mood stabilizers, and were determined to get their prescription refills on time and often upset if they could not. I hypothesized that chronic benzodiazepine use created circadian or ultradian cycles of sedation, which might be experienced as fatigue or amotivation, and withdrawal, which might be experienced as anxiety or insomnia. Patients often tinkered with the doses and timing of the doses in an attempt to smooth out these cycles. The current study will determine whether symptoms of anxiety, depression and insomnia; and indices of psychosocial function improve, while BZ use decreases during a twelve-week trial period of substituting quetiapine for benzodiazepines. This would be a proof of concept study in a unique subpopulation with MDD or GAD.

1. STUDY OBJECTIVES 2.1 Primary objective For MDD:The primary efficacy measures will be change from baseline to endpoint in the MADRAS and the change in mean daily benzodiazepine dose in diazepam equivalents during the past week.

For GAD: The primary efficacy measures will be change from baseline to endpoint in the HAM-A and the change in mean daily benzodiazepine dose in diazepam equivalents during the past week.

2.2 Secondary objectives Secondary efficacy measures will include response and remission rates and self-rated scales of symptoms and psychosocial function (the IDS-SR for depressive symptoms, the SAS-SR for social/vocational function and the Q-LES-Q for quality of life).

1. STUDY PLAN AND PROCEDURES 3.1 Overall study design and study plan Subjects will be followed to assure a stable regimen of antidepressant medications and BZ for at least 8 weeks. Patients will then be treated with flexible doses of quetiapine-ER up to 300mg/day for 8 weeks. Dosing will begin with Seroquel 50 mg. at bedtime and will escalate weekly to Seroquel 100mg., Seroquel-XR 200mg. and Seroquel-XR 300 mg depending on clinical response and side effects. BZ doses will be gradually tapered at the equivalent of 5mg/day/week of diazepam equivalents where possible. This is not a forced titration schedule.

3.2 Rationale for study design, doses and control groups The design is an open-label, proof of concept study. The dosing is based on my prior clinical experience with this patient population. There are no control groups.

3.4 Treatments 3.4.1 Identity of investigational product and comparators Seroquel-XR 50 mg., Seroquel-XR100mg., Seroquel-XR 200mg. and Seroquel-XR 300 mg.

3.4.2 Doses and treatment regimens Dosing will begin with Seroquel-XR 50 mg. at bedtime and will escalate weekly to Seroquel-XR 100mg., Seroquel-XR 200mg. and Seroquel-XR 300 mg depending on clinical response and side effects.

3.5 Pre-study, concomitant and post-study treatment(s)

Prohibited medication during the study:

• Potent cytochrome P450 inhibitors (including but not limited to ketoconazole, itraconazole, fluconazole, erythromycin, clarithromycin, troleandomycin, indinavir, nelfinavir, ritonavir, fluvoxamine and saquinavir)
• potent cytochrome P450 inducers (including but not limited to phenytoin, carbamazepine, barbiturates, rifampin, St. John's Wort, and glucocorticoids)
• Any psychotropic besides quetiapine, antidepressant or benzodiazepine.

Other medication, which is considered necessary for the subject's safety and well-being, may be given at the discretion of the investigator(s). The administration of all medication (including investigational products) must be recorded in the appropriately.

1. OUTCOME AND VARIABLES 4.1 Primary Variable For MDD: The primary efficacy measures will be change from baseline to endpoint in the MADRAS and the change in mean daily benzodiazepine dose in diazepam equivalents during the past week.

For GAD: The primary efficacy measures will be change from baseline to endpoint in the HAM-A and the change in mean daily benzodiazepine dose in diazepam equivalents during the past week.

4.2 Secondary Variables Secondary efficacy measures will include response and remission rates and self-rated scales of symptoms and psychosocial function (the IDS-SR for depressive symptoms, the SAS-SR for social/vocational function and the Q-LES-Q for quality of life).

4.3 Efficacy Variable(s) For MDD: The primary efficacy measures will be change from baseline to endpoint in the MADRAS and the change in mean daily benzodiazepine dose in diazepam equivalents during the past week.

For GAD: The primary efficacy measures will be change from baseline to endpoint in the HAM-A and the change in mean daily benzodiazepine dose in diazepam equivalents during the past week.

An overdose is not regarded as an AE. All symptoms associated with the overdose should however be reported as AEs. For further information regarding overdose, see section 11.2.

Pregnancy in itself is not regarded as an AE unless there is a suspicion that an investigational product may have interfered with the effectiveness of a contraceptive medication. For further details regarding pregnancy, see section 11.3.

4.4.1.3 Reporting of serious adverse events Investigators and other site personnel must inform the FDA, via a Medwatch form, of any unexpected and possibly Study Drug-related serious adverse events (SAEs) according to the FDA reporting requirement timelines. The Investigator must also inform appropriate AstraZeneca representatives of any SAE that occurs in the course of the study within one day (ie, immediately but no later than the end of the next business day) of when he or she becomes aware of it. The investigator must also report follow-up information on SAEs within one day. A copy of the MedWatch report must be faxed to AstraZeneca at the time the event is reported to the FDA. Reporting to the FDA is recommended regardless of whether an IND is required for the study.

Additionally, the Investigator shall provide AstraZeneca with a report of all other SAEs that did not qualify for expedited reporting to the FDA (e.g., considered to be expected and/or not related) on at least a quarterly basis in order for AstraZeneca to meet its regulatory reporting obligations.

If the Study is blinded, the Institution shall provide AstraZeneca with the code break information with the initial SAE report or a copy of the randomization schedule at the start of the Study, so that AstraZeneca may un-blind SAE reports to the extent necessary to meet global regulatory reporting requirements.

A complete written SAE report must be sent with a cover page indicating the following:

Drug Name (Seroquel); this is an Investigator Sponsored Study (ISS) Research (Protocol) and AstraZeneca Tracking Number [IRUSQUET0483] Principal Investigator's IND number assigned by the FDA (if applicable) Principal Investigator's full name and address Unblinding information (if applicable) Send by way of fax to Seroquel ISS Safety Representative (302) 885-3043

If a non-serious AE becomes serious, this and other relevant follow-up information must also be provided to AstraZeneca and the FDA within 1 day as described above. All SAEs have to be reported to AstraZeneca, whether or not considered causally related to the investigational product. All SAEs will be documented. The investigator is responsible for informing the IRB and/or the Regulatory Authority of the SAE as per local requirements.

4.4.2 Other safety measurements and variables Safety assessments will include vital signs at each visit and hematology, chemistries, pregnancy testing and urine drug screening prior to and at the end of treatment.

4.5 Volume of blood sampling and handling of biological samples The total volume of blood that will be drawn from each subject in this study is as follows: 30 cc 5. DATA MANAGEMENT We will develop standardized paper forms for collecting all data needed on study subjects including eligibility, demographic and other baseline data, sequential clinical assessments, dosing information, side effects, and outcome measures. all data required to explore the study hypotheses will be collected and entered into a database in the appropriate format for data analysis.

1. STATISTICAL METHODS AND DETERMINATION OF SAMPLE SIZE 6.1 Description of analysis populations All analyses will be conducted on the intention-to-treat population. 6.2 Method of statistical analysis The Wilcoxon signed-rank test will be applied to the mean scores available at weeks 0 vs. endpoint for the primary outcome measures: MADRS, HAM-A \& daily BZ dose, and for the secondary outcome measures IDS-SR, SAS-SR and Q-LES-Q respectively to test the questions of whether there is a difference from pre- to post-quetiapine treatment. Separate tests will be conducted for the two diagnostic groups (MDD and GAD), but there is no intent to compare the two.

The proportions of a priori-defined responders and remitters during the study will be reported in descriptive statistics.

6.3 Determination of sample size NA The N of 20 per group is dictated by budget and feasibility considerations. This is a pilot study with no control or comparison groups. Under a 2-tailed α value of .05, we will have the power to detect a large effect size.