SAFETY ANALYSIS 56
`
`SECTION 8.0
`
`'
`
`SECTION 8.1
`
`EXPOSURE 56
`
`SECTION 8.2
`
`DEMOGRAPHICS: 57
`
`SECTION 8.3
`
`DEATHS so
`
`SECTION 8.4
`
`DISCONTINUA‘TIONS 63
`
`sECTION 8.5
`
`ALL ADVERSE EVENTS 74
`
`SECTION 8.6
`SECTION 8.6.1
`76
`
`SERIOUS ADVERSE EVENTS 76 .
`Serious adverse events for ABBOTT SPONSORED STUDIES (NON JAPANESE)
`
`SECTION 8.6.2
`SECTION 8.6.3
`
`serious adverse events for ABBOTT SPONSORED STUDIES (JAPANESE) 91
`serious adverse events IN NON ABBOTT SPONSORED STUDIES 91
`
`SECTION 8.7
`SECTION 8.7.]
`SECTION 8.7.2
`
`LABORATORY FINDINGS 94
`SERUM CHEMISTRY PARAMETERS 94
`vital signs and electocardiograms 95
`
`SECTION 8.8
`
`ADVERSE EVENTS AND PRECLINICAL STUDIES 98
`
`SECTION 8.9
`
`DOSE-RESPONSE DATA 98
`
`SECTION 8.10
`
`DRUG-DRUG INTERACTIONS 100
`
`SECTION 8.11
`
`DRUG-DEMOGRAPHIC INTERACTIONS 102
`
`SECTION 8.12
`
`LONG TERM ADVERSE EFFECTS 107
`
`SECTION 8.13 WITHDRAWAL EFFECTS _107
`
`“
`
`.
`
`SECTION 8.14
`
`120 SAFETY UPDATE 107
`
`SECTION 8.15
`
`SAFETY DISCUSSION 108
`
`SECTION 9.0
`
`_
`
`, DISOUSSION 110
`
`SECTION 16:0 IL’BECOMMENDATIONS 110
`
`APPENDIX I‘lil.
`
`_
`
`
`
`

`

`SECTION 1.0 MATERIALS USED IN REVIEW
`
`Table"'l
`
`MATERIALS UTILIZED IN REVIEW
`ITEM
`DATE
`
`
`
`Volumes 1,2,3
`
`Volumes 63-725
`“
`'
`
`
`
`18 December 1998
`
`5 Aril I999
`24 Ma 1999
`
`-»
`
`
`
`
`
`
`Introductory Information
`
`Proposed Labeling
`
`
`A- olication Summar
`
`Clinical Data
`Case Re -ort Forms
`
`
`Additional Information
`Additional Information
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`SECTION 2.0 BACKGROUND
`
`SECTION 2.1
`
`INDICATION
`
`Dexmedetomidine is an intravenous alpha-2 adrenoreceptor agonist indicated for sedation
`and- analgesia in an Intensive Care Unit setting.
`
`SECTION 2.2
`
`RELATED NDA’S AND IND’S
`
`All clinical studies were conducted
`
`
`
`— No previOus NDAs are applicable.
`
`SECTION 2.3
`
`PROPOSED DIRECTIONS FOR USE
`
`Dexmedetomidine is proposedfor adults 18 years and older in an Intensive Care Setting
`who require sedation or potentiation' of analgesia for up to 24 hours. The drug will be
`administered intravenously. Dosing is initiated with a loading dose of l pig/kg over 10
`minutes followed by a maintenance dose'o£0.2-O._7 ug/kg/hour; dosing is adjusted to
`achieve the desired-‘Ievel of"sedation. The total daily dose will not‘exCeed 20 ug/kg with a
`daily steady state plasma concentration exposure of' less than 3.0 ng/ml.
`-
`
`SECTION 2.4.
`
`FOREIGN MARKETING
`
`Dexmedetomidine is not marketed anywhere in the world.
`
`

`

`
`
`SECTION 3.0 CHEMISTRY
`
`
`
`Sponsor states the solution should be stored at room temperature (59 to 86 degree F).
`
`SECTION 4.0 ANIMAL PHARMACOLOGY/TOXICOLOGY
`
`The following is a summary of the pharmacology/toxiCOlogy provided by the sponsor:
`
`“Alpha-2 adrenoceptor agonists have been in clinical use since the mid 19605 with the
`introduction of clonidine, the archetypal alpha-2 adrenoceptor agonist. Although
`clonidine has been used as an antihypertensive drug, it is also used as an adjunctive
`sedative in the intensive care setting.
`'
`
`Alpha-2 adreneeeptor activation is known to result in a variety of responses from several
`organs and tissues. A prominent effect is reduction in sympathetic nervous activity.
`Activation of presyriaptic alpha-2 adrenoceptors located in sympathetic nerve endings
`inhibits the rEIEase‘bf the neurotransmitter noradrenaline. Activation of postsynaptic
`alpha-2 adrenoceptors in the CNS leads to inhibition of sympathetic activity, causing
`decreases in blood pressure and heart rate, sedation, and relief of anxiety. Activation of
`alpha-2 adrenoceptors at the spinal cord results in analgesia. Peripheral alpha-2
`adrenoceptors in blood vessels mediate vascular smooth muscle contraction.
`‘
`
`Alpha-2 adrenoceptors agonists can decrease stress induced ACI‘H release and hence
`cortisol synthesis by a direct effect on the brain. Alpha-2 adrenergic agonists can inhibit
`insulin release by actions on pancreatic islet cells and have been shown to stimulate
`
`

`

`
`
`growth hormone. Subcutaneous administration of Dexmedetornidine increased blood
`glucose, decreased insulinsecretion, and inhibited lipolysis. Alpha-2 adrenoceptor
`agonists decrease circulating norepinephrine and epinephrine by central and peripheral
`mechanisms.
`
`Dexmedetornidine is a specific alpha-2 adrenoceptor agonist as shown by both receptor
`binding and functional studies. Dexmedetornidine has_very low affinity for alpha-l
`adrenoceptors (1300 times less than for alpha-2 adrenoceptors in the rat membrane
`model) and negligible affinity for other receptors, including beta adrenergic, muscarinic,
`dopamine, serotonin, mu- and delta opiate, GABA, andbenzodiazepine receptors.
`Dexmedetornidine is a lipophilic compound which is rapidly and extensively distributed
`to tissues and rapidly eliminated.
`"
`
`Dexmedetornidine is an alpha-2 sedative in various animal species including the rat, dog,
`rabbit, and mouse producing dose dependent sedation/hypnosis when administered either
`intracerebroventricularly, subcutaneously, intraperitoneally, _or intravenously. The
`sedative effect is biphasic: lower (IO-300 ug/kg) doses cause maximal sedation and
`higher doses (2 1000 ug/kg) result in a reversal of the Dexmedetornidine sedative effect.
`The reversal of the sedative effect seen at higher doses of Dexmedetomidine is
`hypothesized to be due to the activation of alpha-l adrenoceptors by Dexmedetornidine,
`as the reversal-could also be blocked by the alpha-l antagonist prazosin. At low doses
`(< 3 ug/kg) Dexmedetornidine is anxiolytic in mice and rats. It acts synergistically with
`midazolam, diazepam, or fentanyl to induce sedation/hypnosis and has potent volatile
`anesthetic sparing properties which are mediated via central alpha-2 adrenoceptors.
`
`Dexmedetomidine administered both spinally and peripherally to rats, dogs, mice
`monkeys, and sheep produces dose dependent analgesia which is more potent than that
`caused by clonidine, ST-9l, xylazine, epinephrine, or‘norepinephrine. The analgesia
`effect in animals lasts 1 to 8 hours depending on the route of administration. The
`analgesic effects of Dexmedetomidine are mediated by alpha~2 adrenoceptors.
`
`Lower doses of Dexmedetornidine cause reduction in blood pressure and heart rate
`through a central effect whereas higher doses of Dexmedetomidine result in peripheral
`alpha-1 adrenoceptor activation and resultant higher blood pressure. The initial
`hypertensive efietof Dexmedetornidine seen with IV bolus injections is reduced with a
`slower rate of infusion. At doses causing increases in mean arterial pressure,
`Dexmedetomidinejeduces heart rate and cardiac output in a dose dependent manner.
`Dexmedetornidine'has no direct depressant effect on the myocardium except at a very
`high supra-clinical concentration. Through its sympatholytic action, Dexmedetornidine
`depresses cardiac function and contractility; the effect on cardiac function and
`contractility are dose dependent.
`
`Dexmedetomidine produces no significant effect on respiratory function except for mild
`respiratory depression at high supra-clinical doses.
`
`

`

`
`
`
`
`Other effects of Dexmedetomidine in animals include reduced seizure thresholds,
`modulation of body temperature, and induction of hypothermia. Dexmedetomidine acts in
`an additive manner with opioids in producing analgesia and counteracts opioid induced
`muscle rigidity.
`
`Animal deaths have been reported only after the administration of doses of
`Dexmedetomidine exceeding the LDSO. All adverse effects are extensions of alpha-2
`adrenergic activity.”
`
`SECTION 5.0 “SUMMARY OF HUMAN PHARMOKINETICS
`
`The pharmacokinetic and pharmacodynamic profiles of Dexmedetomidine were based on
`data from lft__studies. The sponsor has summarized the human’pharmacokinetic and
`bioavailability data as follows:
`
`“Dexmedetomidine is extensively metabolized in humans. In a radiolabeled
`Dexmedetomidine study, there was virtually no penetration of radioactivity into the
`cellular fraction. The major circulating metabolites are the N-glucuronides of
`Dexmedetomidine. Other minor metabolites include the carboxy (COOH), N-methylated
`(N-meth), the glucuronide conjugate of hydroxylated Dexmedetomidine, and additional
`unidentified minor metabolites.
`
`The pharmacokinetics of Dexmedetomidine are biphasic with rapid distribution
`(t l/2 0t 2 6 min) and a mean terminal half life of approximately 2.0 to 2.5 hours.
`Following the loading infusion, venous plasma concentrations rise rapidly. Due to the
`rapid distribution pharmacokinetics, concentrations drop quickly when the loading
`infusion stops, after which the combined effects of the loading and maintenance infusions
`hold plasma concentrations stable until the infusion is terminated. Dexmedetomidine is
`almost exclusively eliminated by metabolism; 95% of a radioactive dose is excreted as
`conjugates in the urine, and the remainder in the feces.
`
`The following were measured PK parameters in healthy human subjects:
`
`Table 2
`
`
`:Wnfieiimedetomidine Pharmacokinetic Parameters
`
`Parameter
`
`
`
`
`
`
`143.9 1155
`
`AUC (ng-h/g)
`
`Clearance (Uhour>
`
`
`
`
`
`
`
`Modified Sponsor’s Table 3 Vol 8/10-1-69
`
`
`
`

`

`8
`
`Dexmedetomidine is 93.7% bound to plasma proteins. There is no difference in binding
`due to gender, and it does not differ in normal subjects or subjects with mild, moderate,
`or severe renal impairment. Subjects with hepatic impairment have protein binding that is
`82-88% of the protein binding of normal subjects depending on the severity of the
`impairment.
`
`Mean arterial pressure, systolic blood pressure, and diastolic blood pressure showed a
`biphasic response with initial decreases at the lowest concentrations, followed by a return
`to baseline and increases over the mean baseline level when the plasma concentration of
`Dexmedetomidine was greater than 3.2 ng/ml. The mean heart rate decreased
`progressively until. the mean plasma concentration of Dexmedetomidine was 3.2-5.1
`ng/ml; at this concentration heart rate reached a plateau and did not decline further.
`Cardiac output decreased approximately 20-30% at Dexmedetomidine plasma ~
`concentrations of 2-4 ng/ml in most subjects. At concentrations above 4 ng/ml, no further
`decrease incardiac output was observed. There was no evidence of respiratory
`depression.
`
`Special Populations:
`
`Renal Failure: During a 2 day study period, six severely renally impaired subjects
`(creatinine clearance < 3 ml/min) received a single 10 minute intravenous infusion of
`Dexmedetomidine 0.6 ttg/kg; they were compared to healthy subjects. Pharmacokinetics,
`sedation, heart rate, blood pressure, oxygen saturation,'respiratory rate, and safety were
`assessed. Blood and urine samples were collected for assay of Dexmedetomidine and
`Dexmedetomidine metabolites. The collection of the samples occurred prior to closing,
`during the infusion, and at protocol specified time points up to 24 hours post-dosing.
`There were no statistically significant differences between the two groups for any of the
`pharmacokinetic parameters. There were no clinically relevant differences between
`healthy subjects and renally impaired subjects. The renal subjects were slightly more
`sedated than control subjects.
`‘
`
`Hepatic Impairment: During a 3 day study, 18 subjects with hepatic impairment received
`a single 10 minute intravenous infusion of Dexmedetomidine Doug/kg; they were
`compared to healthy subjects. The objective of the study was to evaluate the effect of
`various degreesofhepatic impairment on the pharmacokinetics of Dexmedetomidine. In
`addition, sedation, heart rate, blood pressure, oxygen saturation, respiratory rate, and
`safety were assessed. The hepatic subjects contained groups with mild, moderate, and
`severe hepatic: fail-tire. Dexmedetomidine clearance values for'subjects with mild,
`moderate, and severe hepatic impairment were only 74%, 64%, and 53%, respectively of
`those observed in the normal healthy subjects. Subjects with hepatic impairment appeared
`to have no clinically relevant different effects on hemodynamic or respiratory parameters.
`Sedation may have been greater in the hepatically impaired group. Therefore, the dose of
`Dexmedetomidine may need to be reduced in subjects with hepatic impairment.
`
`Drug Interaction: 6 studies in humans were performed to evaluate possible interactions
`between Dexmedetomidine and other drugs. 5 of these studies involved pharmacokinetic
`
`
`
`

`

`
`
`analysis. The agents studies were isoflurane, rnidazolam, alfentanil, propofol,
`rocuronium, and esmolol (no PK analyses were performed with esmolol). Except for
`alfentanil, no pharmacokinetic interactions were observed. PK effects by
`Dexmedetornidine on alfentanil were observed along with enhancement of alfentanil’s
`analgesic effects.
`
`Age: The influence of age on the pharmacokinctics ofDexmedetornidine in healthy
`subjects was evaluated by administration of 0.6 ug/kg Dexmedetornidine given as a 10
`minute intravenous infusion. Subjects were studied up to 24 hours following infusion.
`Age groupings varied from 18 years to > 65 years. Dexmedetomidine pharmacokinetics
`were not different between young, middle aged, and elderly subjects nor between male
`' and female subjects when differences in body weight were accounted for. Therefore the
`same Dexmedetornidine infusion is expected to produce the same Dexmedetom‘idihe
`profile, and the elimination characteristics are expected to be the same irrespective of age
`or gender.”— -
`
`SECTION 6.0 DESCRIPTION OF CLINICAL DATA
`SOURCES
`
`SECTION 6.1
`
`DEVELOPMENT
`
`Sponsor presents the following information on development of Dexmedetomidine:
`
`“me05 Group, Finland began research involving alpha-2 adrenoceptor agonists and
`antagonists in the late 19705. One of the first agents Was detomidine which has-a higher
`affinity for the alpha-2 adrenoceptor than clonidine. Detomidine was developed for use as
`a sedative/analgesic in horses and cattle and it was first registered for marketing in
`Finland in 1983. Medetomidine, a more specific and selective alpha-2 adrenoceptor
`agonist than clonidine, was launched in 1987 in Finland and the Scandinavian countries
`for use as a veterinary sedative/analgesic in dogs and cats; it was approved for use in
`animals in the US in 1997. The sedative and anesthetic pharmacological activity of
`medetomidineresides predominantly in the dextro enantiomer, Dexmedetomidine, which
`was first synthesized by Farmos group in 1986.
`
`Clinical worlEin the US began with the submission of an Investigational New Drug
`Application in 1989. Clinical investigation included many potential applications of
`Dexmedetornidine from preoperative through postOperative periods. In 1990 Farmos
`
`merged with Orion Corp. _‘
`'
`"
`'
`‘
`'
`'
`ge form of Dexmedetornidine
`for clinical use to Abbott Laboratories in 1994."
`
`SECTION 6.2_
`
`OVERVIEW OF CLINICAL STUDIES
`
`
`
`
`
`

`

`10'
`
`- conducted a total of 56 clinical trials in which a total of 1527 subjects/patients
`were exposed to Dexmedetomidine. The
`‘\—-——;—~ ‘1 studies evaluated the use of
`Dexmedetomidine in the perioperative setting and utilized various modes of
`administration includingrapid IV infusion, continuous IV infusion, and IM injection.
`Transdermal and oral administration of Dexmedetomidine were also studied in a limited
`capacity.
`
`Abbott Laboratories initiated its own clinical program to evaluate Dexmedetomidine in
`the perioperative setting using IV infusion administration. A total of 21 studies ( 13 Phase
`I and 8 Phase II/III) have been completed in the US, Canada,, and Europe in support of
`the perioperative program, during which a total of 230 subjects were closed and 767
`' patients received Dexmedetomidine. 13 of the 21 studies were Phase I trials assessing the
`pharmacokinetic/pharmacodynamic properties of Dexmedetomidine, the safety profile.
`the potential for drug interactions, and its use in special populations. The remaining 8
`studies were Phase II/III trials evaluating the use of Dexmedetomidine as an anesthetic
`adjunct in patients undergoing major surgery and electroconvulsive therapy. The effect
`on minimum alveolar concentration of an inhalation anesthetic was also studied. 4 studies
`were conducted in Europe and Canada by Abbot Labs to evaluate Dexmedetomidine’s
`potential as a sedative and analgesic agent for patients in the intensive care setting. A
`total of 631 subjects/patients were dosed/treated with Dexmedetomidine in these studies.
`
`SECTION 6.3 DEMOGRAPHICS
`
`See Section 8.2
`
`SECTION 6.4
`
`EXTENT OF EXPOSURE
`
`See Section 8.1
`
`SECTION-LP EFFICACY FINDINGS
`
`SECTION'7i1‘: OVERVIEW OF CLINICAL EFFICACY STUDIES:
`
`The Sponsor identified two Phase III studies that were conducted in support of this NDA
`application: W97-245 and W97-246. Sponsor also identified one Phase II study
`supporting the application, W97-249. Study W97-249 randomized 12 patients in one
`center and was not reviewed for findings of efficacy.
`'
`
`
`
`
`
`

`

`11
`
`SECTION 7.2
`
`SUMMARY OF STUDIES PERTINENT TO
`EFFICACY
`
`SECTION 7.2.1
`
`STUDY W97-245
`
`SECTION 7.2.1.1 5 PROTOCOL REVIEW SUMMARY
`
`TITLE: 'A Phase III,,‘Multi-Center, Randomized, Placebo-Controlled, Double-Blind
`Study Evaluating the Safety and Efficacy of Dexmedetomidine When
`Compared to Placebo, With Midazolam, in ICU Sedation in Post-Operative
`Patients
`
`OBJECTIVES:
`
`Primary:
`
`The primary objective of this two-part, Phase III study was to evaluate the
`efficacy and safety of Dexmedetomidine in patients requiring ventilation,
`sedation, and intensive care following surgery. Dexmedetomidine was to be
`administered as clinically indicated according to Ramsay sedation scores
`[see Appendix 1 for description of Ramsay sedation score]; the goal was to
`achieve Ramsay scores of 2 3, as clinically indicated. The primary efficacy
`variable for this study was the total dose (mg) of midazolam required in
`addition to the study drug to achieve adequate sedation (as clinically
`determined by the Ramsay sedation scale) during intubation.
`
`Secondary: Secondary variables included total dose of midazolam (mg) administered
`during study drug infusion, use of morphine for pain, as assessed by total
`dose used with Dexmedetomidine as compared to placebo; use of
`paracetamol for pain after extubation. as assessed by total dose used with
`Dexmedetomidine as compared to placebo, and time to extubation. as
`measured by time to arrival in ICU until time to extubation.
`
`:. as.
`
`STUDY DESIGN: .
`This was a two part study in postoperative patients requiring a minimum of 6 hours
`ventilation and sedation in ICU. Part I was open-label to allow the investigator to become
`more familiar with the observed clinical effects of Dexmedetomidine prior to starting the
`double-blind portion of the study. Part II was double blind, randomized, and placebo-
`controlled.
`.
`
`Patients were to be screened within 7 days prior to receiving study drug. Screening was to
`include a complete medical history and physical examination, laboratory assessments,
`and 12-lead electrocardiogram. Study drug administration was to be initiated as soon as
`possible after arrival in the ICU but not later than 1 hour after admission to the ICU. If
`
`
`
`

`

`_
`
`12
`
`possible, study drug was to be started prior to the patient’s awakening in the ICU or
`requiring any other medication for sedation. If a patient required sedation post-surgery
`and prior to the start of study drug, midazolam (l-mg bolus) could be given as required.
`Study drug infusion was to be continued for 6 hours post extubation. The investigator
`_
`could have continued the infusion at his/her discretion for a maximum of 24 hours.
`Patients were to be observed for a 24-hour period after the end of the study drug infusion.
`
`Part I of the study was to include up to 4 patients per site. Patients were to be
`administered a loading dose of 6.0 mcg/kg/h of Dexmedetomidine over a 10 minute
`period, followed by a maintenance infusion of 0.4 mcg/kg/h. Following the initial
`.:
`maintenance infusion, the rate was to be adjustedif clinically necessary, in increments of
`0.1 mcg/kg/h or higher. The infusion rate was to be maintained between a range of 0.2 to
`0.7 mcg/kg/h to achieve and maintain a Ramsay sedation score of 3 or higher (as '
`clinically appropriate for the patient’s needs). Following exiubation, the infusion rate .
`could havebeen adjusted to achieve a Ramsay sedation score of 2 and above (as
`clinically appropriate).
`'
`
`In Part II of the study (double blind, randomized, placebo controlled), approximately 300
`patients were to be randomized to one of two treatment groups: Dexmedetomidine or
`placebo with additional doses of midazolam for sedation administered as clinically
`indicated. Patients were administered a two stage infusion consisting of a 10 minute
`loading dose of 6.0 mcg/kg/h of Dexmedetomidine or placebo followed by a maintenance
`infusion of 0.4 mcg/kg/h. Following the initial maintenance infusion, the rate could be
`adjusted in increments of 0.1 mcg/kg/h or higher, and was to be maintained in the range
`of 0.2 to 0.7 mcg/kg/h as clinically deemed necessary to achieve and maintain a Ramsay
`sedation score of 3 or higher as clinically appropriate. Following extubation, the infusion
`rate was to be adjusted to achieve a Ramsay sedation score of 2 or above as clinically
`appropriate.
`"
`
`INCLUSION CRITERIA:
`To be included in the study, patients were to have satisfied all of the following inclusion
`criteria:
`
`0
`
`Signed and dated the Informed Consent after the study had been fully explained or
`had a legally acceptable representative sign and date the Informed Consent.
`0 Required sedation for ventilation and intensive care for a minimum of 6 hours
`following surgery
`0 Male or female; age 18 and over (in Austria, age 19 or older)
`0
`If female and of child bearing potential, was not pregnant (confirmed by negative
`pregnancy test) and not lactating.
`'
`-
`
`EXCLUSION CRITERIA:
`
`Patients were not eligible for the study if they met any of the following criteria:
`0 Had serious central nervous system trauma.
`0 Had undergone or required intracranial surgery during current hospitalization.
`
`

`

`13
`
`0 Required the use of neuromuscular blocking agents during the study period, except
`for the insertion of the endotracheal tube.
`0 Required epidural or spinal analgesia during the ICU stay.
`0
`In whom opiates or benzodiazepines were contraindicated or had known or suspected
`serious allergy to any medication that might have been administered during the course
`of the study.
`0 Was grossly obese (estimated body weight was greater than 50% above ideal body
`weight)
`0 Was currently hospitalized for drug overdose
`r -In whom alpha-2 antagonists or alpha-2 agonists were contraindicated
`0‘ Was currently being treated or had been treated within the last 30 days with alpha-2
`agonists or antagonists
`_
`.
`0 Had participated in a trial with any experimental drug within 30 days prior to
`admission to the ICU
`' ‘
`
`0 Was terminally ill, whose life duration expectancy was no more than or around 24
`hours.
`
`Was considered unable to undergo any procedure required by the protocol
`Had demonstrated tolerance to standard sedating medications
`Had previously received Dexmedetomidine
`Had unstable or uncontrolled diabetes
`Had excessive bleeding which was likely to require resurgery
`. Had received midazolam for maintenance of anesthesia
`Has clinically significant arrhythmia or any other cardiac condition or factor which,
`in the investigator’s opinion, might have increased the risk to the patient or precluded
`obtaining satisfactory data.
`'
`
`REMOVAL OF PATIENTS FROM THERAPY ASSESSMENT:
`A patient was to be withdrawn from the study immediately if any of the following
`occurred:
`'
`
`0 Due to an adverse event, the investigator decided that discontinuation was in the
`patient’s best interest.
`0 The patient requested withdrawal from the study.
`0
`Patients requiring reoperation
`0 A changeeceurred in the patient’s status such that exclusion criteria became part of
`the patient profile.
`
`Patients who: withdrew from the study were not to be replaced. Those patients
`withdrawn fi'om the study due to an adverse event were to have all events documented
`and followed to a satisfactory resolution. Patients who were withdrawn from the study
`for any reason during study drug administration were required to have all final
`evaluation procedures completed.
`
`DOSING SCHEDULE:
`
`Dexmedetomidine or placebo was to be administered as a two-stage infusion: a 10
`minute loading dose followed by a maintenance infusion using standard syringe pump
`
`

`

`l4
`
`and IV administration sets. Study drug was never to be administered directly‘into the
`pulmonary artery. Study drug was to be initiated as soon as possible after arrival in the
`ICU but no later than 1 hour after admission to ICU. If possible, study drug was to be
`started prior to the patient’s awakening in ICU or requiring any other medication for
`sedation. If a patient required sedation post surgery and prior to start of study drug,
`additional doses of rnidazolam for sedation could have been given as required. Drug
`administration consisted of a 10 minute loading dose of 6.0 mcg/kg/h followed by a
`maintenance infusion of 0.4 mcg/kg/h. The 6 mcg/kg/h loading dose was chosen to
`' achieve a Dexmedetomidine plasma concentration of approximately 1.5 rig/mg as a
`result of experience gained in Phase I studies and resultant PK modeling. Clinical
`.
`effects of sedation should have been observed within 15 minutes of the start of the study
`' drug. The infusion rate could have been adjusted in increments of 0.1 mcg/kg/h or
`higher, and should have been maintained in the range of 0.2:0.7 mcg/kg/h as clinically
`deemed necessary to achieve and maintain a Ramsay sedation score of at least 3 as
`clinically appropriate. Following extubation, the infusion rate could have been adjusted
`to achieve a Ramsay sedation score of 2 and above as clinically appropriate.
`
`During study drug administration, rescue medication was limited to rnidazolam for
`sedation and morphine for pain as required. After extubation, paracetamol use was to be
`permitted as clinically indicated. During the 10 minute loading dose 1 mg bolus dose of
`midazolam was allowed if necessary. The maintenance dose of Dexmedetomidine or
`placebo was to be adjusted prior to any administration of additional rnidazolam. Study
`drug infusion was to be continued for 6 hours post extubation. The investigator may
`have continued the infusion at his/her discretion for a total of 24 hours total drug
`infusion.
`
`In the ICU, rnidazolam for sedation was to be administered following an increase in
`study drug infusion. Prior to the administration of rnidazolam, patients were to be
`assessed for sedation using the Ramsay sedation scale. The Ramsay assessment was to
`be performed prior to and 10 minutes after every rate change in study drug
`administration or administration of any rnidazolam. Pain was assessed either by direct
`corrununication with the patient or by autonomic signs (sweating, tachycardia,
`hypertension). Sponsor recommended that initial doses of rnidazolam be administered as
`a bolus in doses of 0.02 mg/kg. If, in addition to increasing the infusion rate of study
`drug, the patientreceived 3 boluses of rnidazolam within any 2 hour (during study drug
`infusion), further midazolam, if necessary, may have been administered at a continuous
`infusion rate of 0_.Ol-0.0Z mg/kg/h. Morphine may have been administered for pain in
`increments'o’f 2-mg IV boluses. Prior to the administration of morphine, the patient was
`to be assessed for pain.
`'
`_
`
`Standard ICU monitoring protocols were to be employed. All patients were to be
`ventilated to maintain PaC02 and Pa02 tensions as determined by the investigator.
`Minimum ventilatory support was to be utilized. Patients were to be weaned from the
`ventilator and extubated only if the investigator deemed it appropriate and after meeting
`the following criteria:
`0
`Patient was aWake or arousable, neurologically intact, cooperative, and comfortable.
`
`
`
`

`

`
`
`15
`
`0
`
`0
`
`Patient had an Fi02 value 5 0.4, PEEP < 5 cm H20, and pressure support 5 10cm
`H20.
`‘
`
`Patient had the following lung mechanics: minute ventilation expired > 4 Urnin but
`< 15 Umin, tidal volume > 5 mng and spontaneous respiratory rate <25/rnin.
`
`The following drugs were not to be allowed during study drug infusion:
`o
`Sedating agents Other than midazolam; analgesic agents other than morphine (after
`extubation, use of paracetamol was permitted as clinically indicated).
`0 Neuromuscular blocking agents except for the insertion of the endotracheal tube.
`0" 'Epidural or spinal analgesic agents.
`.
`. 6 Any drugs contraindicated with the use of Dexmedetomidine, midazolam. or
`morphine.
`-
`o Alpha-2 agonist/antagonist.
`
`SECTION 7.2.1.2
`
`STATISTICAL ANALYSIS
`
`Only patients from the randomized, double-blind, placebo-controlled part of the study
`(Part II) were to be included in the efficacy analyses. A patient was required to satisfy the
`following evaluability criteria in order to be included in the evaluable subset:
`o The patient received study drug for at least 6 hours,.unless the patient was
`I prematurely discontinued by the investigator due to an adverse event.
`0 The patient received none of the following medications during study drug
`administration: sedating agents other than midazolam, analgesic agents other than
`morphine or paracetamol, neuromuscular blocking agents except for insertion of the
`endotracheal tube, epidural or spinal analgesic/anesthetic agents, any drugs .
`contraindicated with the use of midazolam, Dexmedetomidine, or morphine, or other
`prohibited medications
`.
`o The patient received only morphine or paracetamol for 'pain management.
`0 The patient was intubated for at least 6 hours.
`
`A patient was included in the intent-to-treat subset if he or she was randomized and
`required intensivegare and sedation following surgery. Patients in Part I and H of the
`study were included in the safety subset if he/she received any study drug.
`
`EFFICACY ANAEYSES:
`The primary efficacy analysis was based on the intent-to-treat subset of patients._A
`second set of efficacy analyses was completed on the evaluable subset.
`
`Primary Efficacy Analysis
`
`The primary efficacy variable in this study was the total dose (mg) of midazolam during
`intubation received as rescue medication for sedation during the period of study drug
`administration. The total dose was summarized by the number of patients in the following
`three total dose categories: no midazolam (0 mg); a subtherapeutic dose over time (>0 mg
`
`
`
`

`

`16
`
`to 4 mg); and a therapeutic dose (>4 mg). Differences in the distributions of the
`proportion of patients in each category between the dexmedetomidine treatment group
`and theplacebo treatment group were tested with a chi-square statistic. Center differences
`were also explored. The total dose was also summarized by N, mean, standard error of
`the mean (SEM), minimum, median, and maximum The treatment groups were
`compared using an analysis of variance (ANOVA) with treatment, center and treatment-
`by-center interactionincluded in the model.
`
`.
`
`[Reviewer Note: The final primary eflicacy analysis submitted in this application is '
`difi’erentfrom the sponsor’s proposal in the original protocol. None of the amendments to
`this study reflect the analysis that was performed. ‘At a meeting with the sponsor at the
`conclusion of the Phase II studies, Dr. Thomas Permutt (the_reviewing statistician)
`suggested that the capability of Dexmedetomidine to provide sedation would be more
`convincinglydemonstrated by an analysis ofhow many patients needed any rescue
`medication rather than by measuring the‘amount of rescue medication utilized by both
`placebo and Dexmedetomidine patient groups. Consequently, the sponsor was '
`encouraged to incorporate calculations of the number ofpatients receiving any amount
`of midazolam in the primary efficacy analysis. ]
`
`Secondary Efficacy Analyses
`
`The following are secondary efficacy variables in the study:
`
`0 Total dose of midazolam during study drug administration:
`
`The total dose of midazolam (mg) administered during study drug infusion was
`calculated. The total dose was divided by the length of infusion to determine the total
`dosing rate during infusion and was expressed as mg/h. The length of infusion was
`defined as the difference between the time ofthe start of study drug and the end of
`study drug infusion. The total dosing rate was summarized by N, mean, SEM,
`median, minimum. and maximum. The treatment groups were compared using an
`ANOVA with treatment, center, and treatment-by-center interaction included in the
`model.
`
`"L;
`
`.3.
`
`o
`
`The total dose of morphine during study drug administration:
`
`Analysis oftot'al dose of morphine during study drug administration was run on three
`populations of patients. The first population consisted of patients who did not receive
`any midazolam during intubation. The second population consisted of patients with a
`subtherapeutic total dose over time of midazolam during intubatio