FDA CENTER FOR DRUG EVALUATION AND RESEARCH
`DIVISION or ANESTHETIC, Cnmcu. CARE, AND ADDICTION DRUG Pnooucrs
`HFD-I70, Room 9B-45_, 5600 Fishers Lane. Rockville MD 20_8§_'{
`
`Tel:(30l)443-374]
`
`MEMORANDUM
`
`November 5, E999
`
`File, NDA 21-038
`
`Bob A. Rappaport, M.D.
`Deputy Director, DACCADP
`Team Leader, Anesthetic Drug Group
`
`Supervisory Review of NDA 21-038, Dexmedetomidine I-ICl
`
`BACKGROUND:
`
`NDA 21-O38, Dcxmedetomidine HCl, was submitted _by Abbott Laboratories Inc. on
`December 18, 1998.
`Dexmedetomidine is a potent and highly selective a-2-
`adrenoreceptor agonist.
`The sponsor claims that
`their product produces titratable,
`predictable sedation in an ICU setting, from which patients are easily arousable and
`cooperative. The sponsor also claims that their product provides improved analgesia in
`the postoper_a_tivi_e
`ICU setting.
`The ct-2-adrenoreceptor agonist detomidine was
`developed for‘ use as a sedative/analgesic in horses and cattle and was registered for
`marketing in Finlandin 1983. Medetomidine, launched in 1987 in Scandinavia, was a
`more selective of»-2-adrenoreceptor agonist used as a sedative/analgesic in cats and dogs.
`It was approved for veterinary use in the US in 1997. The sedative and analgesic activity
`of medtomidine are believed to reside predominantly in its dextroenantiomer
`dexmedetomidine. The enantiomer was first synthesized by Farmos Group in Finland in
`1986. Numerous perioperative indications have been evaluated since that time. Farmos
`merged with Orion Corp.
`in 1990, and Orion licensed the injectable dosage fonn of
`dexmedetomidine for clinical use to Abbott Laboratories in 1994.
`
`Orion conducted 56 clinical
`trials of dexmedetomidine with various modes of
`administration including rapid intravenous infusion, continuous intravenous infusion,
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 1
`
`

`
`intramuscular injection, as well as transdennal and oral administration. Abbott initiated
`its own clinical development program and completed 21 studies (13 Phase I and 8 Phase
`II/III) in the US, Canada and Europe. They also completed 2 studies in Japan: a Phase I
`safety and pharmacokinetic study of rapid infusion in 9 healthy males, and a Phase II
`safety and dose response study of rapid infusion in 109 patients. The sponsor reported
`that the case report forms for these 2 studies were unavailable and they did not include
`the data in the ISS database.
`- _
`
`The clinical studies of the effectiveness and safety of this new formulation have been
`reviewed [submitted August 29, 1999] by Charles Cortinovis, M.D. Dr. Patricia Hartwell
`contributed two -addenda [submitted September 13, 1999 and October 27, 1999]
`reviewing safety data in the original application, a supplementary safety package, and the
`120-Day Safety Update. The application has also been reviewed by Jonathan "Ma, Ph.D.
`(biostatistics), Suresh Doddapaneni, Ph.D. (clinical pharmacology and biophannaceutics),
`I-larry Geyer, Ph.D. (pharmacology/toxicology), Michael Theodorakis, Ph.D. (chemistry),
`and BeLinda A. Hayes, Ph.D. (abuse liability).
`In this memo, I will briefly review the
`effectiveness and safety data summarized in the primary clinical review, as well as any
`relevant information found in the primary reviews from the other disciplines, and make
`appropriate recommendations for action on the NDA.
`
`EFFECTIVENESS:
`
`Evidence of efficacy has been submitted in two clinical studies W97-245 and W97-246.
`
`Study W97-245:
`
`This was a randomized, double blind, placebo-controlled, parallel group study conducted
`at 33 centers in Canada and Europe. The Study consisted of two parts. Part 1 was an
`open-label evaluation of dexmedetomidine in up to 4 patients per site. This portion of the
`study was designed to allow the investigators to become familiar with the observed
`clinical effects of dexmedetomidine prior to starting the double-blind portion of the study.
`Patient data from Part I was not included in the efficacy analyses.
`'
`
`In Part II of the study, adult postoperative patients who required a minimum of 6 hours of
`ventilation and sedation in the ICU setting were randomized to either dexmedetomidine
`or placebo for sedation. Within one hour of admission to the ‘ICU, patients were
`
`Dexmedetornidine
`NDA 2 I -038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 2
`
`

`
`administered a loading dose 6.0 pg/kg/hour over a 10 minute period, followed by a
`maintenance infusion of 0.4 pg/kg/hour.
`The infusion rate could be adjusted by
`increments of 0.1 pg/kg/hour in order to maintain a Ramsay Sedation Score‘ of 3 or
`higher. However, it was required that the rate be maintained between 0.2 and 0.7
`pg/kg/hour. Following extubation, the infusion rate was adjusted to achieve a Ramsay
`Sedation Score of 2 or above. Study drug infusion was continued for at least 6 hours after
`exrubation and, at the discretion of the investigator, up-to a maximum of 24 hours total
`study drug infusion.
`
`Rescue medications were limited to midazolam for sedation and morphine for pain. ‘After
`extubation, paracetamol was administered when clinically indicated. When the
`investigators judged that there was need for an increase in sedative medication, they were
`to first adjust the maintenance dose of dexmedetomidine. . Midazolam was-adrninistered
`as bolus doses of 0.02 mg/kg. Using the Ramsay Sedation Score,
`the patient was
`assessed prior to and 10 minutes after every rate change in study drug or administration
`of midazolam.
`If the patient required 3 bolus doses of midazolam within any 2 hour
`period, after appropriate adjustments of the study drug infusion rate, further midazolam
`was administered as a continuous infusion at 0.01 to 0.02 mg/kg/hour.
`
`'
`
`The need for analgesic administration was assessed either by direct communication with
`the patient regarding pain, or by the presence of abnormal autonomic signs such as
`sweating, tachycardia and hypertension. Morphine was administered for pain as 2-mg
`intravenous boluses.
`
`The protocol specified primary efficacy parameter was the total dose of midazolam in
`milligrams administered during the period that the patient was intubated. The efficacy
`analysis was based on the intent to Treat [ITT] population and analysis on the Evaluable
`population was also performed. A second primary efficacy endpoint was analyzed based
`on a recommendation made by the Division biostatistician, Dr. Permutt, at a development
`meeting with the sponsor. This endpoint was a comparison of the numbers of patients
`who fell into one of the following three categories of midazolam use:
`
`1. No dose
`2. Sutftiiérapeutic dose
`3. Therapeutic dose
`
`(0 mg)
`(0-4 mg)
`(>4 mg)
`
`This outcome measure was not specified in any amendment to the protocol. However,
`the analysis was undertaken prior to breaking the study blind.
`
`' 6 = asleep, no response
`5 = asleep. sluggish response to light glabellar tap or loud auditory stimulus
`4 = asleep but with brisk response to light glabellar tap or loud auditory stimulus
`3 = patient responds to commands
`'
`2 = patient cooperative. oriented, and tranquil
`! = patient anxious.'agitated. or restless
`':
`-
`
`.
`Dexmedetomidine
`NDA 21-O38
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 3
`
`

`
`Secondary eflicacy parameters listed in the protocol for this study included’:
`
`~. 1.
`
`2.
`
`Use of morphine for pain - as assessed by total dose used with
`dexmedetomidine as compared to placebo (mg/hr)
`Use of paracetamol for pain after extubation — as assessed by total dose used
`with dexmedetomidine compared to placebo (mg/hr)
`Time to extubation - measured as time..of arrival
`extubation
`
`in ICU until
`
`time of
`
`However, the secondary efficacy parameters listed in the study report were:
`
`Total dose of midazolam during study. drug administration
`Total dose of morphine during study drug administration
`. Total dose of morphine by time period
`—
`.
`‘Ramsay Sedation Score
`_
`Ratio’ of Ramsay Sedation Score of “l ” during study drug administration
`Time to extubation and weaning duration
`Nurses’ and patients’ assessment
`
`These changes in secondary outcome measures were not specified in any amendment to
`the protocol.
`
`Results:
`
`Eighty-six patients were enrolled and 85 treated in Part I of the study.
`
`In Part ll of the study, 178 patients were randomized to dexmedetomidine and 175 to
`placebo. All patients were administered study drug and comprised the ITT population.
`Two dexmedetomidine treated and 6 placebo patients were excluded from the Evaluable
`patient set.
`
`Dr. Cortinovis’ Table 4
`patient disposition:
`7'; ‘.5.
`
`[page 22 of his review], reproduced below, summarizes the
`
`’ This infonnation differs from that documented by Dr. Cortinovis in the medical officer's review and by
`Dr. Ma in the Statistician‘s review.
`It is based on documentation provided by Dr. Patricia Hartwell who
`examined the original documents at my request.
`’ The ratio is the propprtion of assessments that equal 1 divided by the total number of assessments for the
`patient.
`-
`-'
`,
`3 .
`Dexrnedetomidine
`NDA 2l-038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC — Exhibit 1019 — Page 4
`
`

`
`
`
`__
`Id
`——
`
`Received disallowed medication
`
`1303,6004, 7601
`
`‘y_Patients_ could have had more than one reason for non-evaluability
`Modified Sponsor's Table 8.la Vol. 8/I0-62-73
`
`Nine patients in the dexmedetomidine group and 10 in the placebo group were I
`discontinued fro?n the study prematurely. Each of these patients discontinued due to
`adverse events.
`
`Primary Eflicacy Analyses:
`
`1.
`
`required statistically significantly less midazolam
`Dexmedetomidine patients
`compared to the placebo treated patients in both the IT!‘ and Evaluable patient
`analyses. Dr. Cortinovis’ Table 8, page 25 of his review, summarizes these results
`and is reproduced below:
`'
`
`mms mus win
`on ORIGINAL
`
`Dexmedetoniidine
`NDA 2 I -038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 5
`
`

`
`Table 2.
`
`Total Dose of Midazolam (mg) During Intubation
`Placebo
`Dexmedetomidine
`
`Treatment Effect
`
`p-Value°
`umenuorreawauens (N) T
`-rm ns.en:4.o2
`4.23:1.-as
`xmlj
`Evaluable Patients (N)
`’
`176
`Mean 2 SEM
`4.56:tl.42
`
`l8.46t4.l4
`
`0.0014
`
`‘ p-value from ANOVA'
`SEM = Standard Error of Mean
`
`—
`
`. --
`
`Modified Sponsor’s Table 8.2a Vol. 8/ l0-62-74
`
`Statistically significant» differences were observed between the treatment groups in
`both the ITT and Evaluable analyses, with the majority of the dexmedetomidine
`treated patients requiring no midazolam compared to'th_e majority of placebo
`patients who required greater than 4 mg of rnidazolam. Dr. Cortinovis’ Table 9,‘
`page 25 of his review, summarizes these results and is reproduced below:
`
`Table 3.
`
`Total Dose Categories of Midazolam During Intubation
`Placebo
`Dexmedetomidine
`
`Treatment
`Effect
`
`p-Value'
`
`
`
`_—— ‘
`——‘T
`
`p-value from chi-square
`Modified Sponsor's Table 8.2b Vol. 8/ l0-62-75
`
`Secondary Eflicacy Analyses:
`
`Their were n¢differences in any of the analyses when performed on either the ITT or 0
`Evaluable patient data sets. The following table, based on Dr. Ma’s Table 3.3, page 6 of
`his review,_st._tnun_arizes the results for six of these analyses:
`
`Dexmedetomidine
`NDA 21-oas
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 6
`
`

`
`Table 4.
`
`Placebo
`N = 175
`Mean :9: SEM‘
`
`Dexmedctomidine
`N == 178
`Mean :1: SEM
`
`p-value
`Treatment Effect
`
`Total dose of midazolarn during drug
`administration (mg/hour)
`
`Total dose of morphine during drug
`administration (mg/hour)
`
`1.19 (0.23)
`
`0.29 (0.07)
`
`0.83 (0.07)
`
`0.47 (0.06)
`
`Total dose of morphine during 0 to 6.5
`. hours (mg)
`'
`i
`
`8.5 (0.79) --
`
`4.9 (0:56)
`
`Total dose of morphine during 6.5 to
`end (mg/hour)
`
`Ramsay Sedation Score AUC during
`drug administration
`
`0.42 (0.08)
`
`0.24 (0.05)
`
`3.3 (0.05)
`
`3.6 (0.05)
`
`Ratio of Ramsay Sedation Score of 1
`during drug administration (%)
`
`7 (0.8)
`
`3 (0.5)
`
`‘SEM = Standard Error of Mean
`
`As noted by Dr. Ma in his review, the Ramsay Sedation Score itself (AUC) was not a
`useful endpoint to consider as, for both groups, dose titration and rescue medication were
`used to maintain the patients at a specified level of sedation indicated by the Ramsay
`Score. However, a smaller ratio of Ramsay score of 1, for any particular patient, might
`indicate less anxiety during the treatment period. Statistically significant center effects
`were noted for most secondary endpoints indicating that patients in different countries
`either required and/or were administered differing amounts of sedative and analgesic
`medications.
`
`The following additional- secondary endpoints were discussed by Dr. Cortinovis in his
`review:
`.-: -.3-.
`
`Time tg extubation and weaning dg;_'atjon:
`‘
`4..
`
`No statisticallysignificant differences were noted in time to readiness for ext_ubation or
`actual extubation, when that time was measured either from ICU arrival or start of study
`drug. No statistically significant differences were found between the treatment groups for
`median duration of weaning.
`
`Nurses’ and patients’ assessment
`
`statistically significantly lower patient
`Dexmedetomidine treated patients had a
`management index’ [defined on page 30 of Dr. Cortinovis’ review] score compared with
`Dexmedetornidine
`NDA Zl-038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC — Exhibit 1019 — Page 7
`
`

`
`placebo treated patients. However, the actual numerical differences were notlikely to be
`clinically relevant, according to Dr. Cortinovis.
`
`Patient satisfaction survey
`
`The dexmedetomidine treated and placebo patients rated similarly their present
`experience with sedation compared to prior experiences, their comfort during the ICU
`sedation, their remembrance of pain, discomfort from breathing ‘tube, people and noise,
`and whether or not they would have the same sedative treatment in the future. However, V
`61% of dexmedetomidine treated patients compared to 52% of placebo patients rated
`their overall experience as “better than expected!’
`
`.
`
`Data Integrity;
`
`The. Division of Scientific Investigations’ [DSI] Clinical Inspection Summary for this
`application notes that one of the two pivotal study sites inspected by DSI was found to
`have protocol violations which may compromise some of the data arising out of that site.
`That site in Study W97-245 enrolled 5 out of 45 patients out of sequence; and one of
`those five subjects appeared to be a seven year old child who did not meet the inclusion
`criterion for age. DSI hasrequested clarification of the discrepancies and recommends
`that, until the matters are clarified and found to be satisfactory, we not use the data from
`those five subjects in support of the application.
`
`Dr. Thomas Permutt, biostatistics teamleader, has reviewed the data and the DSI
`
`recommendations and has concluded that removal of the data from those five patients
`would not afi”ect the outcome of the efficacy analyses.
`
`Study W97-246:
`
`This was a randomized, double blind, placebo-controlled, parallel group study conducted
`at 36 centers in Canada and Europe. The Study consisted of two parts. Part I was an
`open-label evaluation of dexmedetomidine in up to 4 patients per site. This portion of the
`study was designed to allow the investigators to become familiar with the observed
`clinical effect's‘o'f'dexrnedetomidine prior to starting the double-blind portion of the study.
`Patient data from Part I was not included in the efficacy analyses.
`
`In Part II of the study, adult postoperative patients who required a minimum of 6 hours of
`ventilation and sedation in the ICU setting were randomized to either dexmedetomidine
`or placebo for sedation. Within one hour of admission to the ICU, patients were
`administered a loading dose 6.0 pg/kg/hour over a 10 minute period, followed by a
`
`Dexmedetornidine
`NDA 21-038
`
`T Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 8
`
`

`
`The infusion rate could be adjusted by
`maintenance infusion of 0.4 pg/kg/hour.
`increments of 0.1 pg/kg/hour in order to maintain a Ramsay Sedation Score‘ of 3 or
`higher. However,
`it was required that the rate be maintained between 0.2 and 0.7
`pg/kg/hour. Following extubation, the infusion rate was adjusted to achieve a Ramsay
`Sedation Score of 2 or above. Study drug infusion was continued for at least 6 hours after
`extubation and, at the discretion of the investigator, up to maximum of 24 hours total
`study drug infusion. _-
`_
`
`Rescue medications were limited to propofol for sedation and morphine for pain. After
`extubation, paracetamol was administered when clinically indicated. When the ‘
`investigators judgedthat there was need for an increase in sedative medication, they were
`to first adjust the maintenance dose of dexmedetomidine. Propofol was administered as
`bolus doses of 0.02 mg/kg. Using the Ramsay Sedation Score, the patient was assessed
`prior to and 10 minutes after every rate change in study drug or administration of
`propofol. ffthe patient required 3 bolus doses of propofol within any 2 hour period, after
`appropriate adj ustrnents of the study drug infusion rate, further propofol was administered
`as a continuous infusion at 0.5 to 4.0 mg/kg/hour.
`
`The need for analgesic administration was assessed either by direct communication with
`the patient regarding pain, or by the presence of abnormal autonomic signs such as
`sweating, tachycardia and hypertension. Morphine was administered for pain as 2-mg
`intravenous boluses.
`
`in
`The protocol specified primary efficacy parameter was the total dose of propofol
`milligrams administered during the period that the patient was intubated. The efficacy
`analysis was based on the Intent to Treat [ITT] population and analysis on the Evaluable
`population was also performed. A second primary efficacy endpoint was analyzed based
`on a recommendation made by the Division biostatistician, Dr. Permutt, at a development
`meeting with the sponsor. This endpoint was a comparison of the numbers of patients
`who fell into one of the following three categories of propofol use:
`
`1. No dose
`2. Subtherapeutic dose
`3.‘-ifherapeutic dose
`
`(0 mg)
`(0-50 mg)
`(>50 mg)
`
`This outcome measure was not specified in any amendment to the protocol. However,
`the analysis was tihdertaken prior to breaking the smdy blind.
`
`‘ 6 = asleep. no response
`5 = asleep, sluggish response to light glabellar tap or loud auditory stimulus
`4 = asleep but with brisk response to light glabellar tap or loud auditory stimulus
`3 = patient responds to commands
`-
`2 = patient cooperatiye. oriented. and tranquil
`I = patient anxious."agitated. or restless
`3.
`
`;
`Dexmedetomidine
`NDA 21-038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 9
`
`

`
`IO
`
`Secondary efficacy parameters listed in the protocol for this study included':
`
`-1. Use of morphine for pain -
`as assessed by
`dexmedetomidine as compared to placebo (mg/hr)
`2. Use of paracetamol for pain after extubation - as assessed by total dose used
`with dexmedetomidine compared to placebo (mg/hr)
`3. Time to ex_tubation - measured as time of arriY.al in ICU until time of extubation
`
`total dose used with
`
`However, the secondary efficacy parameters listed in the study report were:
`
`I. Total dose·of propofol during study d11,1g administration
`2. Total dose of morphine during study drug administration
`3. Total dose of morphine by time period
`...
`4. Ramsay Sedation Score
`5. Ritio6 of Ramsay Sedation Scpre of "l" during study drug administration
`6. Time to extubation and weaning duration
`7. Nurses' and patients' assessment
`
`These changes in secondary outcome measures were not specified in any amendment to
`the protocol.
`
`Results:
`
`Ninety-three patients were enrolled and 92 treated in Part I of the study.
`
`In Part II of the study, 203 patients were randomized to dexmedetomidine and 198 to
`placebo. All patients were administered study drug and comprised the ITT .population.
`Three dexmedetomidine treated and 7 placebo patients were excluded from the Evaluable
`patient set.
`
`Dr. Cortinovis' Table 18
`patient disposition:
`
`[page 44 of his review], reproduced below, summarizes the
`
`APPEARS THIS WAY
`ON ORIGINAL
`
`~This information differs from that documented by Dr. Cortinovis in the medical officer's review and by
`Dr. Ma in the Statistician's review. It is based on documentation provided by Dr. Patricia Hartwell who
`examined the original documents at my request.
`6 The ratio is the pro_p_onion of assessments that equal I divided by the total number of assessments for the
`patient.
`
`Dexmedetomidine
`NDA21-038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC – Exhibit 1019 – Page 10
`
`

`
`5.
`
`
`Intent to Treat Patients (All Treated)
`'
`W
`Non-Evaluable patients
`'
`Q
`Evaluable Patients
`Reasons for Non-Evaluability (Patient Numbers) —
`Insufficient lntubation
`’
`_ ‘
`Received disallowed medication
`~ "
`Enrolled twice
`
`I-
`
`Modified Sponsor's Table 8.1a Vol. 8/ 10-86-73
`
`Fourteen patients in the dexmedetomidine groupland 8 in the placebo group were
`discontinued from the study prematurely. Most of these patients discontinued due to
`adverse events._
`- -
`
`Primary Eflicacy Analyses:
`
`1. Dexmedetomidine patients._required statistically significantly less propofol
`compared to the placebo treated patients in both the ITT and Evaluable patient
`analyses. Dr. Cortinovis’ Table 22, page 47 of his review, summarizes these
`results and is reproduced, with modifications, below:
`
`APPEARS THIS WAY
`ON ORIGINAL
`
`Dexmedetomidine
`NDA 2l~038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 11
`
`

`
`Table 6.
`
`Total Dose of Propofol (mg) During Intubation
`Placebo
`Dexmedetomidine
`
`Treatment Effect
`
`p-Value'
`
`—I
`
`
`
`
`
`' p-value from ANOVA
`SEM = Standard Error of Mean
`
`Modified Sponsor's Table 8.2a Vol. 8/I0-86-73
`
`.'
`
`‘
`
`2. Statistically significant differences were observed between the treatment groups
`in both the ITT and Evaluable analyses, with the majority of the
`'
`'
`dexmedetomidine treated patients requiring no propofol compared to the
`majority of placebo patients who required greater than 50 mg of propofol. Dr.
`Cortinovis’ Table 24, page 48 of his review, summarizes these results and is
`reproduced, with modifications, below:
`
`Table 7.
`
`Total Dose Categories of Propofol During Intubation
`Placebo
`Dexmedetomidine
`
`intent-to-Treat Patients (N)
`
`> 0mg to 50 mg
`>50 mg
`Evaluable Patients (N)
`
`>0 mg to 50 mg
`
`IE£:E_
`47(24%)
`l22(60%)
`30(l5%)
`43(2l°/o)
`l2l(6l%)
`38(l9°/o)
`
`46(24°/o)
`
`e
`
`l20(60%)
`
`30(l6%)
`H5(60"/9)
`
`42(2l°/o)
`38(i9°/o)
`
`Treatment
`Effect
`
`p-Value‘
`<0.00l
`
`' p-value from chi-square
`Modified Sponsor's Table 8.2b Vol. 8/ I0-86-74
`
`Secondary Efficacy Analyses:
`
`Their were no differences in any of the analyses when performed on either the ITT or
`Evaluable pa__tient‘~-_data sets. The following table, based on Dr. Ma's Table 3.6, page 10 of
`his review, summarizes the results for six of these analyses:
`
`Dexmedetomidine
`NDA 2 l-O38
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC — Exhibit 1019 — Page 12
`
`

`
`Placebo
`N = 198
`Mean 1 SEN!"
`
`Dexmedetomidine
`N = 203
`Mean 1 SEM
`
`p-value
`Treatment Effect
`
`Total dose of midazolam during drug
`administration (mg/hour)
`
`Total dose of morphine during drug
`administration (mg/hour)
`
`Total dose of morphine during 0 to 6.5
`hours (mg)
`"
`
`Total dose of morphine during 6.5 to
`end (mg/hour)
`
`Ramsay Sedation Score AUC during
`drug administration
`
`Ratio of Ramsay Sedation Score of 1
`during drug administration (%)
`
`‘SEM = Standard Error of Mean
`
`39 (4.1)
`
`5.3 (1.2)
`
`0.89 (0.07)
`
`0.43 (0.05)
`
`8.5 (0.64) '-
`
`4.] (0.47)
`
`0.55 (0.07)
`
`0.l6 (0.03)
`
`3.] (0.04)
`
`3.4 (0.04)
`
`7 (0.7)
`
`4 (0.5)
`
`As noted by Dr. Ma in his review, the Ramsay Sedation Score itself (AUC) was not a
`useful endpoint to consider as, for both groups, dose titration and rescue medication were
`used to maintain the patients at a specified level of sedation indicated by the Ramsay
`Score. However, a smaller ratio of Ramsay score of 1, for any particular patient, might
`indicate less anxiety during the treatment period. Statistically significant center effects
`were noted for most secondary endpoints indicating that patients in different countries
`either required and/or were administered differing amounts of sedative and analgesic
`medications.
`
`The following additional secondary endpoints were discussed by Dr. Cortinovis in his
`review:
`
`Time to extubation and weaning dggatign:
`
`No statistically significant differences were noted in time to readiness for extubation or
`actual extubation, when that time was measured either from ICU arrival or start of study
`drug. No statistically significant differences were found between the treatment groups for
`median duration of weaning.
`
`Nurses‘ and patients‘ assessment
`
`statistically significantly lower patient
`Dexmedetomidine treated patients had a
`management index‘ [defined on page 52 of Dr.__Cortinovis’ review] score compared with
`‘:
`-
`Dexmedetomidine
`NDA 2i-038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC — Exhibit 1019 — Page 13
`
`

`
`l4
`
`placebo treated patients. However, the actual numerical difierences were not likely to be
`clinically relevant, according to Dr. Coninovis.
`
`Patient satisfaction survey
`
`rated similarly their present
`The dexmedetomidine treated and placebo patients
`experience with sedation compared to prior experiences, their comfort during the ICU
`. sedation, their remembrance of pain, discomfort from breathing-tube, people and noise,
`and whether or not they would have the same sedative ueatment in the future. However,
`70% of dexmedetomidine treated patients compared to 60% of placebo patients rated '
`their overall experience 5 “better than cxpected._”
`
`Subgroup Analyses of Efficag:
`
`Dr. Ma has reviewed the sponsor’s subgroup analyses and reports a few exceptions to the
`overall findings of significant efiicacy of dexmedetomidine. These include:
`
`1.. Of the 43 patients in both groups at the five German centers in Study 245,
`only 1 (5%) in the treated group required 0 mg of midazolam during the
`intubation period. At the other centers more than 50% of the treated patients
`required no midazolam.
`.
`. For the 20 patients at the single Austrian center in Studies 245 and 246,
`similar amounts of midazolam and propofol were required by the placebo and
`dexmedetomidine treated groups.
`. When analyzed by type of surgery, similar amounts of midazolam were
`required by the 34 patients undergoing head and neck surgery in. Study 245
`(p=0.96). Statistically significant differences were found for the two treatment
`groups
`for patients undergoing cardiac surgery,
`laparotomy and other
`surgeries in that study and all types of surgery in Study 246. However, the p-
`value for head and neck surgery for Study 246 was 0.052.
`
`SAFETY:
`
`‘-4:
`
`--—~_.—\/——» ita from the ISS
`The originalflND4§ submission-excluded the
`database. Tiiis fact was discovered by Dr. Coninovis after the submission had been filed.
`In a teleconference in late May of 1999, the sponsor claimed that they had not included
`this data because it came from studies performed to assess different indications than the
`one that is the subject of this application. The sponsor was informed that it would be
`necessary for them to compile, analyze and submit this missing data. The sponsor
`informed us that it would take a minimum of two months to complete the assignment and
`an early August submission was agreed upon. The new data was submitted on August
`16, 1999. This submission was found to be incomplete, with missing case report forms
`..[ERF’s], CR1-"s ‘from the .' -
`- which had not been translated, and missing
`
`I‘?
`
`Dexmedetornidine
`NDA 21-038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC — Exhibit 1019 — Page 14
`
`

`
`I5
`
`the sponsor
`During a follow-up teleconference,
`form tabulations.
`case report
`acknowledged the missing data and stated that they had determined some of the data to be
`not useful due to unavailability of CRF's or the omission of data from CR1-‘ ’s . They
`were told to immediately provide as much, of the data as possible and written
`explanations for any data which would not be submitted. These final portions of the
`safety database have been submitted.piecemeal since that time. In her first addendum to
`the medical review, Dr. Hartwell has evaluated this late data in full and carefully
`delineates
`the various parts, based on GCP suitability, availability of primary
`documentation, and overall importance to the safety profile of dexmedetomidine.
`
`In an attempt to incorporate the recently submitted data into the exposure database, Dr.
`Hartwell created two tables [see pages 5 and 6 of her first addendum] which summarize
`the number of studies and the number of patients included in the supplemental ISS,
`broken down by GCP suitability and by those with "available CR1-“s.
`She then
`incorporated the patients from the supplemental submissions into a table [page 6 of her
`first addendum] of all exposed patients, updating Dr. Cortinovis' Table 32 [page 57 of his
`review]. At this time, based on the information available from the sponsor, it appears that
`a total of 3338, subjects have been exposed to dexmedetomidine in clinical studies.
`However, the sponsor has categorized 11 Phase I studies (109 subjects) and 4 Phase II/III
`studies (146 subjects) as containing inadequate information; and this data was not
`included in the supplemental ISS. Thus, the overall ISS database includes 3083 subjects
`exposed to dexmedetomidine.
`«
`
`Extent of exposure by dose is summarized in the table below, based on Dr. Cortinovis’
`Tables 7 and 22 [pages 24 and 47, respectively, of his review] and Dr. Hartwell’s Tables
`4 and 5 [page 7 of herfirst addendum]:
`..
`
`APPEARS THIS WAY
`ON ORIGINAL
`
`Dexmedetomidine
`NDA 21-038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 15
`
`

`
`1 3'
`‘
`-Q
`
`Continuous
`Infusion
`
`Rapid
`Infusion
`
`IM
`Administration
`
`Phase 1 Studies:
`Mean Total Dose
`N
`
`(pg/kg)1 SD
`Mean Total Duration
`V N
`(hr) 1 SD
`Minimum
`
`'
`
`_
`‘
`
`4
`
`Mean Total Dose
`
`N
`(pg/kg)1SD
`Mean Total Duration
`N
`
`(hr) 1 SD
`Minimum '
`Maximum
`
`'-
`
`. .
`
`174
`
`i
`
`a
`
`"
`‘ is
`
`?--»-»
`
`-
`35
`
`‘~- -
`
`3.52 1 3.97
`
`0.90 1 .63
`
`3.86 1 3.33
`
`I74
`7.65 1 8.26
`
`l67
`0.13 1 0.05
`
`-
`
`'
`I0
`0.02 1 0.00
`
`.
`
`3'
`
`” -
`
`1475
`
`10.10 1 6.54
`0.02
`
`267
`010
`
`106
`
`0.05 1 0.03
`0.02
`
`.'.-"'v.:‘..;::
`
`‘"1.
`''
`
`662
`1.91: 0.75
`
`Mean Total Duration
`N
`.-.-; -.s'.
`
`(hr) 1 SD
`
`14.7 1 4.51
`
`N/A
`
`N/A
`
`for all patient/subject listings in this table does not equal the total N for the safety
`NB. The total
`databue. The‘ total N here falls between the total N for patients exposed and the total N for patients in the
`database (those with adequate and available data). based on availability of dose and duration infonnation
`from the different study sites.
`
`During her review of the safety data, Dr. Hartwell requested more specific information on
`exposure by close and duration from the sponsor. Dose by duration exposure data for all
`treated patients in the Phase II/III continuous infusion studies is summarized in Dr.
`Hartwell‘s Table-1 from page 2 of her second addendum:
`
`‘:
`-
`
`Dexmedetotnidine
`NDA 2 l-038
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Phannaceuticals LLC — Exhibit 1019 — Page 16
`
`

`
`Table 10. Extent of Exposure — Frequency of Duration by Dose
`Phase II/III Continuous Infusion Studies - All-Treated Patients
`
`(ll?!)
`
`—-
`
`E-C1111-:11-I111‘K-I1
`$111-—-KT-I--K11
`—KaZ-21.111-1-1
`Z—Zl—j;1III1III1ZiX
`ZHT-11111111111
`CKEZ—I—Z1jIIEII1I1ZE£C
`IZEZZZZKZITZZIK
`I—E—1XjX1-111111211
`C—C1:11EZf11-1111111
`IZX1—1ECXIX1l1ZTT11TZ1
`n-It-IDIII-T-—1
`HKZXTICIZZXZZ
`Ctjitjjijijjii-1
`EH11‘-1-I:-1:11-‘Kl:
`—KZZjZ:KZ1111KI111-T
`TIKZZIIIIZIZITXIKIX
`
`From Sponsor‘: Table of Duration by Dose. Supplementto NDA. I0-27-99. Exhibit 4
`NB. The total N corresponds to the total number of patients with available total duration data as in my Table 9. above. However,
`there was a discrepancy in the total N between the sponsor's ISS Supplement (submitted August I6. I999). Appendix C, Table
`2.l.2. p. 68 and Supplemental Information submitted on October 27, 1999 (p. 7). This discrepancy of 2 patients appears to be a
`simple error in addition.
`
`Dose by duration exposure data for all treated patients in the Phase I continuous infusion
`studies is summarized in Dr. Hattwell’s Table I from page 1 of her second addendum:
`
`Table 11. Extent of Exposure — Frequency of Duration by Dose
`Phase I Continuous Infusion Studies — All-Treated Patients
`
`Ej
`
`i
`TE-—-f-if-mi
`-1--.j-u-1
`-f--€31-E-j-j-D
`T-ili-f-1-f-f-T -11
`1111311111111
`111:1--:1?--ijj-1
`T-Z-n-13--inn
`n-n-1-1-in