`
`FDA CENTER FOR DRUG EVALUATION AND RESEARCH
`
`HFD-1 70, Room 9B-45, 5600 Fishers Lane, Roclcville A/H) 2085 7
`
`Tel:(30l) 827-7410
`
`MEMoRAi~mUM
`
`John K. Jenkins, MD
`to:
`. Director,
`Ofice ofDrugEvaluation II
`
`Division File: NDA # 21-038
`
`. CynthiaG. McCormick,Ivfl)
`
`. Director, Division ofAnesthetics‘,/ Critical Care
`Products
`
`d Addiction Drug
`
`subject: Dexmedetomidine NDA
`
`date: November 30, 1999
`
`This memorandum summarizes for the file the basis for the approval action recommended
`by the Division of Anesthetics,'Critica'l Care," and Addiction Drug Products for NDA #21-
`038, Dexmedetomedine I-ICI for Injection, a sedative/hypnotic agent intended for use in
`the intensive care setting.
`'
`'
`i
`1'
`
`Background
`Dexmedetornidine is the dextro-enantiomer of the racemic mixture, medetomidine’ and a
`selectivea- -adrenoreceptor agonist. It has been shown in standard animal models of
`eficacy to have anxiolytic activity (0.3o2.0 ug/kg IV), analgesic activity (3-6 ug/kg IV),
`and se_dative.prop_erties (l0_-30 pg/kg IV) in a dose-related manner in mice, rats and dogs.
`Dexmedetotiiidine was developed in humans primarily for its sedative properties and was
`studied as a sedative in the intensive care setting, delivered by continuous intravenous
`infusion.
`
`It was anticipated that dexmedetomidine would provide effects similar to those of
`clonidine, also an oz-2-adrenergic agonist which has been used as an anesthetic adjuvant
`producing analgesia and sedation, and purported to decrease anesthetic requirements and
`
`‘ Medetornic_li_ne is a veterinary sedative widely available in Europe and approved in the US
`in 1997.
`i
`;
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Phannaceuticals LLC — Exhibit 1013 — Page 1
`
`
`
`improve hemodynamic stability. The theoretical basis for the use of the cx-2-adrenergic
`-agonists as adjunctive medications is that they are thought to act as neuromodulators,
`regulating central (medullary) cardiovascular or peripheral vasomotor responses such as
`those to anesthetics, thus producing an anesthetic-sparing effect. These effects were not
`specifically characterized for approval purposes, although some exploratory studies were
`undertaken during early development.
`
`A unique feature of dexmedetomidine as a sedative which was observed in phase I studies
`was its property of providing adequate sedation but with ease of alerting and without
`persisting central effects, once the patient is _aroused.
`
`Efficacy
`The Sponsor submitted two adequate and well-controliCd studies of similar design in
`support of the proposed indication for sedation. The studies were randomized, double
`blind, double-dummy parallel group multicenter trials comparing the effects of
`dexmedetomidine infusion with placebo. The trials evaluated the sedative· properties of
`dexmedetomidine and control by inference, that is, they compared the amount of rescue
`medication (midazolam in one trial and propofol in the second) required to achieve a
`specified level of sedation (by the standardized Ramsay sedation scale) between the
`placebo and treatment group from onset to extubation. There were a number of
`potentially confounding variables that were assessed as secondary outcome measures,
`particularly time to extubation and amount of morphine used for analgesia.
`
`In study W97-245, 175 patients were randomized to the placebo arm and 178 patients
`were randomized to receive dexmedetomidine by intravenous infusion at doses of0.4
`µ/kg/hr (with allowed adjustment between 0.2 ll!ld 0. 7 µg/kg/hr) following an initial bolus
`of 6 µg/kg IV. Patients were allowed to receive midazolam as needed to maintain a
`Ramsay sedation score of~- In addition, .morphine sulfate could be administered as an
`analgesic as needed. The primary outcome measure for this study was the total amount of
`rescue medication (midazolam) needed to maintain sedation as specified while intubated.
`There was a statistically significantly greater use of midazolam in patients randomized to
`placebo than to dexmedetomidine during treatment.
`.--....:.::.
`·_>.._
`A second prospective primary analysis was undertaken at the request of the division to
`obtain a direet assessment of the sedative effects of dexmedetomidine, that is, a
`companso~·t,f the percentage of patients who were able to achieve a Ramsay sedation
`score of~ during intubation, without the use of additional rescue medication, between
`the dexmedetomidine and the placebo groups. It can be seen from the results reported in
`the table on the following page that a significantly greater number of patients in the
`dexmedetomidine group (61%) compared to the placebo group (25%) maintained a
`Ramsay sedation score of~ without any additional midazolam rescue.
`
`NDA #21..038-Dexmedetomidine HO
`
`Pagel of 10
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC – Exhibit 1013 – Page 2
`
`
`
`
`
`Midazolam use as rescue medication during intubation (HT)
`Stud W97-245
`
`PBO
`
`Dexmedetomidine
`
`p-value
`
`-
`
`N==17s
`
`- — -—N=17s
`
`Mean total dose (mg) of midazolam
`
`18.6 mg
`
`4.8 mg
`
`0.0011‘
`
`Categorized midazolam use
`# pts used
`.,
`.
`,
`
`108 (61%)
`43(2S%)
`0mg
`36(20%)
`34 (19%)
`0-4 mg
`34 (19%)
`98 (56%)
`>4 mg
`' ANOVA model with rx and ctr. “Chi-square ( afier J.Ma’s table 3.2, review, p.5)
`
`.
`<0.00l"
`
`In study W97-246, 198 patients were randomized to the placebo arm and 203 patients
`were randomized to receive dexmedetomidine by intravenous infusion at doses of 0.4
`pkg/hr (with allowed adjustment between 0.2 and 0.7 pg/kg/hr) following an initial bolus
`of 6 ug/kg IV. Patients were allowed to receive propofol as needed to maintain a
`Ramsay sedation score of 23. In addition, morphine sulfate could be administered as an
`analgesic as needed. The primary outcome measure for this study was the total amount of
`rescue medication (propofol) needed to maintain sedation as specified while intubated.
`There was a statistically significantly greater use of propofol in patients randomized to
`placebo than to dexmedetomidine during treatment.
`
`The same prospective primary analysis that was performed in study W97-245 was also
`performed in this study. It can be seen fi'om the results reported in the table below that a
`significantly greater number of patients in the déxmedetomidine group (60%) compared to
`the placebo group (24%) maintained a Ramsay sedation score of 23 without any
`additional propofol rescue.
`
`"
`
`Midazolam use as rescue medication during intubation (ITT)
`Study W97-246
`PBO
`Dexmedetomidine
`
`p-value
`
`.~:_ --.3-
`
`Mean total
`
`(mg) of propofol
`
`Categorized propofol use
`# pts used
`
`N=l98
`
`513 mg
`
`N=203
`
`72 mg
`
`,
`
`<0.000l‘
`
`122 (60%)
`47(24°/.)
`0mg
`43 (21%)
`30 (15%)
`0-50 mg
`38 (19%)
`121 (61%)
`>50 mg
`' ANOVA model with rx and ctr. “Chi-square (after J.Ma's table 3.5, review, p.9)
`
`<o.oo1*°
`
`For both studies, the time to extubation was measured and analyzed, and found to be,
`based on a very conservative approach, not significantly difierent between groups. For
`more detail,_'_Dr. Jonathan Ma's analysis p. 10-11 should be referenced.
`In addition the
`
`NDA #21-03'8~Dexmedetomidine RC1
`
`‘
`
`Page 3 of 10
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1013 — Page 3
`
`
`
`amount of morphine used for analgesia in both studies was found to be significantly
`greater in the control group. These are both important findings combined with the
`primary analysis, since they establish that the treatment group did not succeed based on
`the sedation afiorded by morphine sulfate or because of a longer time and therefore
`greater access to more medication.
`
`Dexmedetomidine is said to have been studied as adjunctive therapy insofar as rescue with
`a second agent was required in many cases to achieve the specified sedation, rather than
`increasing the infusion (and thus the dose) of dexmedetomidine as needed. Clearly it was
`the primary agent. The sponsor compared between the two randomized groups in both
`studies, the percentageof patients who received only dexmedetomidine and who required
`no rescue medication, confirming its eficacy as monotherapy in two trials;
`
`The primary review team and DLRappaport have carefully reviewed these trials. There is
`nothing to add to the Medical and Statistical analyses and I concur with their conclusions
`that these studies, while somewhat unique in their design, clearly establish that
`dexmedetomidine is an efi‘ective sedative when administered by intravenous infusion at
`doses of 0.4 1.:/kyhr (with allowed adjustment between 0.2 and 0.7 ug/kg/hr) following an
`initial bolus of 6 ug/kg IV.
`
`Safety
`Nonclinical
`
`No significant animal toxicity was described in acute studies in rats or dogs. However,
`chronic dosing of up to 28 days in dogs and rats was associated with hepatic toxicity,
`specifically enlarged livers, eosinophilic inclusions in hepatocytes, and elevated LFTs.
`These changes were not observed in the acute studies. The genesis of the hepatotoxicity
`has not been characterized as to whether it is correlated with parent compound or any
`specific metabolite. While there appears to be an adequate safety margin in dosing, the
`contribution of a difl'erent human metabolic profile may theoretically alter the toxicity of
`this compound with chronic dosing in humans. This bears further evaluation.
`
`-
`
`Dexmedetomidine had no efi'ect on ACTH-stimulated cortisol release in dogs given just a
`single dgsesof 80 pg/kg/dose S.C., but after one week of treatment with 3_ ug/kg/hr, the
`ACTH-stimulated release of cortisol was reduced by 40%. This has implications on the
`hypothalamic-pituitary-adrenal axis with prolonged ICU treatment with this agent, and
`shoultlbe fiirther elaborated concurrently with human trials evaluating the safety of long-
`term infusion.
`-
`
`The nonclinical pharmacolcinetics of dexmedetomidine are similar to humans with the
`exception of metabolism, which difi‘ers by two major metabolites. The two major
`metabolites found in human (the 2 glucuronides of imidazole nitrogen) and absent in the
`rat and dog, were never studied in animals. Because it is projected that this product will be
`used in ICU for longer than 24 hrs of infusion, the potential toxicity ofthese human
`metabolites should be evaluated. This should be done as a Phase 4 study of long-term
`
`NDA #21-03's Dcxmedetoxnidine nci
`
`‘
`
`Page 4 of 10
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1013 — Page 4
`
`
`
`infusion in an appropriate animal species, either indirectly by administration to an animal
`species that does not produce these metabolites or in an animal species which produces
`the same metabolites.
`
`Dexmedetomidine was not shown to be teratogenic in rats ‘or rabbits. However fetal
`toxicity was observed in rats, evidenced by increased postimplantation losses and reduced
`number of live pups per litter. Prenatal and postnatal eflects included reduced pup body
`weights during and after nursing and delayed motor development. Placental transfer of
`dexmedetomidine was observed in rats.
`
`Dexmedetomidine was not mutagenic in the'Ames test or the mouse lymphoma assay. It
`was shown to be clastogenic in both the in vitro human lymphocytes chromosomal
`aberration assay in the presence of metabolic activation and in in viva mouse micronucleus
`assay. .~
`-
`
`Carcinogenicity testing was considered unnecessary due to the projected short-term use of
`this product.
`
`141
`_
`_
`The safety data for this NDA was combined from two sources, ‘--
`Japanese original development program, and subsequent Abbott Laboratories data from
`the more recent development. The safety database of dexmedetornidine exposure includes
`3038 subjects, of whom 1473 were ICU patients who received the drug by continuous
`infusion. The bulk of exposure was in the range of 4-6 mg/kg and less than 16 hours. The
`dose and duration of exposure provide suficient experience to be able to assess the safety
`of this product for the proposed duration of up to 24 hours infusion.
`
`There was also limited exposure (78 patients) who received infusion longer than 24 hours
`with the longest infiision lasting between 30-40 hours in 2 patients.
`
`The deaths and serious adverse events reported were not unexpected for the ICU
`population under study in this NDA either in quality or in quantity.
`
`In the plzlicelio-controlled infiision studies in Phase 2-3, the only commonly reported
`adverse events observed in more than 1% of patients treated with dexmedetomidine and
`occurring with a fiequency more than 2-fold that of the placebo were predictably
`hypotension (22%) , hypertension (12%) , and bradycardia (5%).
`
`NDA #21-038rDe:unedetomidinc HCJ
`
`_
`
`Page 5 of 10
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1013 — Page 5
`
`
`
`Summary of Treatment-Emergent Adverse Events Occurring in >l% of Dexmedetomidine patients in
`Phase II/III Continuous Infusion ICU Sedation Studies’
`Randomized dexmedetomidinc
`
`Adverse Event
`
`Placebo
`
`I
`
`N=379
`(N=387)
`16 (4%)
`84 (22%)‘
`Hypotension
`24 (6%)
`47 02%)‘
`Hypertension
`6(2%)
`20 (5%)‘
`Br-adycardia
`4 (1%) .
`I3 (3%)
`Mouth Dry
`20 (5%)
`16 (4%)
`.__
`__
`Nausea
`‘Statistically significant difference between randomized dexmedetomidine and placebo patients ps0.05
`Data source 2.2.5.5
`
`- .
`
`- —
`
`Abnormal laboratory findings, which might have been,anticipated fi'om the preclinical
`studies, such as elevated LI-‘Ts and glycosuria, were not borne out in laboratory testing.
`
`There are no safety data in pediatric patients. The sponsor will be required to study this
`product in children fiom birth to 16 years of age as a Phase 4 commitment.
`
`Approximately 500 patients over 65 years of age have been studied in this NDA An
`additional analysis of patients over 75 years has been requested of the sponsor with
`comparison of adverse events by age, separating the elderly by >65 to 75 and >75 years of
`age. This will be undertaken in an eflbtt to assess whether dosage adjustment may be
`needed in the very elderly patients based on anticipated PD differences associated with
`sedative agents.
`
`Abuse Potential
`
`~
`
`.
`
`Dexmedetomidine might be expected, based on its clinical pharmacological efl'ects and its
`similarity to clonidine’, to have some abuse liability. Indeed animal studies indicate that
`there are some reinforcing properties. Reinforcing behavior in primates was elicited by
`demedetonudme 1.0 ug/kg/dose >sa1ine and equivalent to saline at 0.0625 ug/kg/dose.
`At a dose of 0.25 pg/kg/dose dexrnedetomidine produced reinforcing behavior comparable
`to pentazocine (CW). Dexmedetomidine also has been shown to attenuate morphine
`withdrawalfsuggestive but not conclusive evidence for dependence liability. A mild
`withdrawal syndrome has been described in rodents after 7 days of treatment.
`
`Extensive receptor binding studies using standard radioligands were presented in the
`NDA, demonstrating very high afinity for the ct-adrenergic receptors andrnoderate
`aflinity for the serotonergic receptors. Binding at the opiate receptors was negligible.
`Comparative binding to relevant controlled substances was not provided.
`
`'
`
`’ After Sponsor’s Table 21 ISS 8/10-239-65
`3 Clonidine is not currently controlled in the CSA. There have been reports of abuse with
`clonidine, mostly of reports of opiate addicts using clonidine to suppress withdrawal
`symptoms rather than for its psychotropic eflects.
`
`NDA #21-O3-&De:nnedetomidine act
`
`-
`
`Page6ofl0
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1013 — Page 6
`
`
`
`On balance, the available studies suggest an abuse potential lower than some products
`controlled in schedule IV or as lowas some not controlled at all. I do not agree with the
`controlled substances evaluation team that this product should not be scheduled due to
`lack of information, but rather that the available information suggests a rather low
`potential for abuse. Furthermore the clinical setting in which it will be used, limited to
`hospital intensive care units, reduces that potential.‘ Continued vigilance is indicated,
`nevertheless, for any actual diversion and abuse that might occur in the post approval
`setting, so that appropriate measures can be taken to control this substance if needed.
`There have been to date no reports of diversion or abuse of medetomidine approved in
`1997.
`"
`
`'. _
`Biapharmaceutics
`The ADME of dexmedetomidine has been fairly well studied, but some unanswered
`questions remain that may be very relevant to long term infusion. For example, it is has
`been demonstrated that there is almost no accumulation of parent drug, following IV bolus
`administration, and that there is nearly complete biotransoformation. The fate ofthe
`metabolites, however, has not been well characterized. Biotransofomiation includes direct
`N-glucuronidation (two major metabolites, total of 34%) and CYP 2A6-mediated
`metabolism (three additional metabolites, 14%), and N-methylation (three metabolites,
`18%). There are additional urinary metabolites that have not been identified yet.
`Dexmedetornidine is about 94% protein‘-bound.
`'
`
`Evaluation of dexrnedetomidine in patients with renal failure demonstrated no change in
`dexmedetornidine PK with severe renal failure following a single dose, but there is no
`information about the possible accumulation of metabolites when dexmedetomidine is
`infused continuously, particularly for. long periods of time. The bulk of elimination of
`metabolites is thought to be renal. Therefo_re, this information should be obtained in Phase
`4 in anticipation of more prolonged infusion in patients with renal insuficiency.
`
`Hepatic impairment afl'ected the PK of dexmedetomidine as expected, and the appropriate
`adjustments for patients with mild, moderate and severe hepatic impairment will be
`included
`package insert.
`
`There was no efi‘ect of age on the pharmacokinetics of dexmedetomidine, although only
`20 elderly volunteers, ranging {Tom 66 to 83 years (mean, 72) were evaluated. The
`possibility of pharmacodynamic difl'erences increasing with increasing age were not
`examined, but should be looked at more closely in Phase 4, as sedative/hypnotics have a
`tendency to result in more significant safety problems (hypotension, confiision, respiratory
`depression) in the elderly. Dexmedetomidine has not been evaluated in the pediatric
`population.
`
`NDA #21-038.-‘pexrnedetornidine Hcr
`
`.
`
`Page 7 of to
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Phannaceuticals LLC — Exhibit 1013 — Page 7
`
`
`
`Interaction studies with a spectrum of anesthetics in viva such as alfentail, midazolam,
`._ propofol and isoflurane did not indicate interactions when added to dexmedetomidine or
`to alfentail, ntidazolarn, propofol or rocuronium when dexrnedetomidine was added.
`
`Chemistry and Manufacturing
`Dexmedetornidine is the dextro-enantiomer of medetomidine (4-[l-(2,3-
`dimethylphenyl)ethyl]-1H-imidazole hydrochloride) and it is manufactured by separation
`of the isomers fiom the racemic mixture. Preparation and characterization of the drug
`substance, levels of impurities including optical purity (levo-enantiomer limited to s 1%)
`have all been judged acceptable. Stability data on the bulk drug substance and regulatory
`specifications were also deemed acceptable.’
`
`/————______.
`
`,
`
`The drug product is a sterile aqueous solution of dexmedetomidine for intravenous
`infiision upon further dilution. The formulation consists of dexmedetomidine HCI (the
`active ingredient) and sodium chloride and water for injection. The drug product is
`prepared using standard methods, has undergone stability testing (undiluted) under ICH
`storage conditions generating data to support a 2-year shelf life, and has been shown to be
`stable in light. Sterility of the drug product is achieved through aseptic fill and terminal
`sterilization by autoclave. The process and data have been reviewed by rrticrobiology and
`found to be acceptable.
`
`The drug product is prepared for use by diluting it with sterile 0.9% sodium chloride
`solution for injection after which it is stable for'24 hours.
`‘
`
`Compatibility data are provided with commonly used IV solutions, drugs (vasoactive
`agents, muscle relaxants, sedatives, narcotics and plasma substitute), tubing, and syringes
`commonly used for administration of IV drugs. It was observed that dexmedetornidine
`has the potential for adsorption onto certain types of natural rubber. This will be noted in
`the package insert, advising use with synthetic components or coated natural rubber "
`components?
`
`A suitable trade name has not yet been selected for the drug product to which the Agency
`agrees.
`.
`
`Data Integrity
`All questions related to data integrity were resolved during the course of review and
`inspection, including questions about some unreported deaths, randomization errors and
`protocol violations that were not reported. DSI inspections were conducted, and aside
`fiom some reports of careless errors in recordkeeping there was no evidence to suggest
`
`NDA #21-03_8_Dexmedetornidine BC!
`
`_
`
`Page a of 10
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Phannaceuticals LLC — Exhibit 1013 — Page 8
`
`
`
`that the data on which the conclusions and recommendations for this NDA will be based
`have significant problems.
`'
`
`_
`Comments:
`There is adequate evidence to support the eficacy and safety of dexmedetomidine to
`approve it for ICU sedation by continuous infiision for 24 hours. It is anticipated that
`there will be increasing demand for more prolonged use of this product once it is
`approved. In addition to collecting additional safety data on prolonged use, there should
`be a better characterization of the activity, toxicity and fate of the metabolites. _
`
`Additional datanshould be obtained for safe use at the extremes of age—pediatric dosing,
`pharmacokinetics and safety should be obtained. Geriatric pharmacodynamic/safety data
`in the very_elderly >75 years should also be generated-or existing data analyzed.
`
`Once the metabolic profile is better established with multiple dosing, its safety should be
`evaluated in patients with renal failure.
`-
`
`Surveillance for possible diversion and abuse can be done through the existing mechanisms
`such as Medwatclg SAM}-ISA’s DAWN database, and DEA reports.
`'
`
`Phase 4 Cammitmenu
`
`The focus of the dexmedetomidine development plan was short-term ICU sedation in
`adults. It is quite clear that this product will not have use limited to this population, and
`therefore the following phase 4 commitments will be requested of the sponsor in an efi‘ort
`to obtain safety data in more extended ICU infusion, in pediatric patients and in the
`elderly.
`
`Nonclinicgl studies
`1. A two-week study in dogs with a'2-week recovery phase should evaluate general
`toxicology of prolonged infusion of dexmedetomidine and the efl‘ect of chronic
`infiision on HPA axis.
`
`2. A second study should evaluate changes in drug metabolism following two
`weeks of infiision.
`
`3:""A;‘third study should evaluate the potential toxicity of human major metabolites
`which are absent in rats and dogs.
`'
`
`A
`gglinical gtudies
`1. Pediatrics: Studies to obtain an indication for sedation in pediatric patients fiom
`birth to 16 years of age in the ICU setting. The development plan should include
`pharmacokinetics and safety in pediatric patients fi'om birth to _l6 years, and
`eficacy data designed at determining appropriate dosage regimens.
`. Geriatrics: Further studies are needed to evaluate the safety v. difierential
`toxicity of dexmedetomidine in very elderly patients, as has been described with
`other sedative/hypnotic drug products.
`3. Longer-term infusion studies should include safety and pharmacokinetics.
`
`NDA #21-03.8Dexmedetomidine EC!
`
`’
`
`Page 9 of 10
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Phannaceuticals LLC — Exhibit 1013 — Page 9
`
`
`
`4. Renal Impairment: Additional data are needed to examine the potential
`accumulation of dexmedetomidine metabolites upon continuous infusion in
`patients with renal impairment.
`
`It is expected that a reasonable timeline for submission of the protocols might be
`approximately.6 months from approval; and co~~~etion of these studies, approximately 2
`years.
`
`Recommended Action: Approval of dexmedetomidine HCl as an adjunctive _J1ledication
`for ICU sedation.
`
`. ..
`
`APPEARS THIS WAY
`ON ORIGINAL
`
`-..
`
`.........
`
`- -.:
`
`-
`NOA #21~38 Dexmedetomidine HO
`
`Page 10 oflO
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC – Exhibit 1013 – Page 10