CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`APPLICA TION NUMBER:
`
`22-8» 87
`
`MEDICAL REVIEW! S}
`
`

`

`
`
`MEMORANDUM
`
`DEPARTMENT OF HEALTH & HUMAN SERVICES
`Public Health Service
`
`Food and Drug Administration
`Center for Drug Evaluation and Research
`
`
`DATE:
`
`July 26, 2009.
`
`FROM:
`
`Abraham Karkowsky, M.D., Ph.D., Group Leader, Division of
`Cardiovascular and Renal Products, HFD-l 10.
`
`TO
`
`Dr. Norman Stockbridge, M.D., Ph.D., Director, Division of
`Cardiovascular and Renal Products, HFD-l 10.
`
`SUBJECT:
`
`Follow—up CDTL memo to TyvasoTM (inhaled treprostinil NDA 22—387,
`United Therapeutics Corporation).
`
`This memo is a follow up document to my CDTL review dated April 18, 2009.
`Please also refer to Dr. Stockbridge’s Division Director’s complete—response memo.
`
`Although some ofthe residual issues that were raised in the initial memos have
`been successfully resolved, other issues remain outstanding. Despite the unresolved
`issues, the current drug-device combination may be approved, subject to the post—
`marketing requirement and commitments as outlined below.
`
`With respect to the resolved issues, the sponsor has submitted the results of the
`biocompatibility information for the device components. CDRH considered the
`biocompatibility information acceptable. The residual CMC issues have also been
`resolved. The Agency recommends a 36—month shelf—life when the drug is stored at
`controlled room temperature in the foil outer packets, In-use stability data allow for a
`maximum of 24 hours of use once the LDPE ampoules are opened and the drug placed in
`the nebulizer. The drug is photosensitive and it must be protected from light by the
`storage inside the foil packages. There are still minor issues related to the carton and
`containers that should be readily resolvable and should not further delay the approval of
`this drug.
`
`The Trade-name TyvasoTM is still considered acceptable.
`
`With respect to the yet unresolved issues, the most salient ofthese is related to the
`inhalation device. CDRH requested that sponsor define the most critical tasks for the safe
`and effective use of the inhalation device. Once these critical tasks were defined, the
`sponsor was to submit a protocol and perform a human factor study, to assess whether
`subjects could safely and effectively use the device as is. The tasks to be assessed
`included the assembly of the device and preparation and administration of the inhaled
`drug.
`Instead of submitting a protocol to CDRH and incorporating their comments, the
`sponsor performed a human factor study in advance of defining the critical steps in the
`
`

`

`use of the device and in advance of concurrence by CDRH with the protocol. CDRH
`considered the study with its results, as performed by the sponsor, inadequate.
`
`Nevertheless, major problems with the ability ofthe rarified population to
`assemble, clean and administer this drug were detected by this less than acceptable study.
`
`Despite obvious inadequacies in the design of the device, the Division and United
`Therapeutics arrived at a time line to address some of the major deficiencies, allowing the
`marketing ofthe current device until that time. The underlying rationale behind allowing
`the marketing of the Tyvaso (drug-device combination) before the device is completely
`acceptable, is the lack ofa device-related safety signal in the clinical study that enrolled
`of235 subjects that were treated with either active drug or placebo using the current
`device.
`‘
`
`The sponsor has agreed to a post—marketing commitment with a deadline of one.
`year to alter the current Optineb-IR device. The alteration is limited to problems with the
`device that are already known. Thesechanges would not alter the hardware or software
`that generates the inhalation aerosol.
`
`MN but
`
`securely.
`
`The baffle plate isalso to be altered to fit more
`
`Once the new device has been re-engineered, the sponsor is then tasked with
`defining the most critical elements in the use ofthe device that are most vulnerable to
`either mitigating the benefit of drug or provoking unnecessary risk to the patient. This
`human factor study as outline in the previous CDTL review would consist of two separate
`studies. The first is an analysis of the tasks related to the care, assembly and sham
`administration of the drug with the new device. In this study the key metrics would be
`observational and determine whether the above tasks can be acceptably performed.
`
`The second study would be to incorporate a pharmacokinetic study to assess
`whether with the re-engineered device reproducible serum concentrations are generated.
`
`In addition, a CRF should be added to all ongoing study to obtain real-use
`information on the patient—related difficulties in the use of the delivery device. The
`prototype of a Device-related CRF as submitted by United Therapeutics on J uly 7, 2009,
`queries the subjects as to difficulties with the assembly care and use ofthe device and
`appears an acceptable CRF to be added to those of the ongoing studies.
`
`The sponsor also has a post—marketing requirement to determine the consequence
`of the novel route of treprostinil administration on the respiratory tract. The specific
`concern is derived from the following considerations:
`
`

`

`There were several respiratory-tree related events that were of a serious nature
`during the small clinical and open-label trial database.
`Pre-clinical observations in rats and dogs demonstrated that when treprostinil was
`administered by the inhalation route, the drug provoked respiratory tract lesions
`that were still present after a 4 week washout period.
`Treprostinil by the subcutaneous route of administration is highly irritating.
`
`The sponsor, therefore, has agreed to perform a pharmacovigilance study with at
`least an additional 1,000 patient-years of exposure as well as matched controls. The
`specific protocol as well as the most appropriate comparative group still needs to be
`defined. The intent of this requirement is to assess the frequency and ultimate
`consequence of the inhaled route of administration of Tyvaso to the respiratory system.
`
`The reviews utilized in this memo were:
`
`A follow—up memo from Monica D. Cooper Ph.D., ONDQA pie-Marketing
`Assessment Division I/Branch 1, dated July 6, 2009.
`A follow—up CDRH memo from Sugato De, Biomedical Engineer
`(ODE/DAGlD/ARDB) lead reviewer; and Ronald Kaye, Human Factors
`Specialist (ODE/DAGID/GHDB), dated June 10, 2009.
`A follow—up memo from Judy Park, PharmD, Safety Evaluator, Division of
`Medication Error Prevention and Analysis, dated June 29, 2009.
`
`

`

`Linked Applications Type/Number
`
`Submission
`
`Sponsor Name
`
`Drug Name / Subject
`
`NDA 22387
`
`ORIG 1
`
`UNITED
`THERAPEUTICS
`CORP
`
`TREPROS‘TINIL FOR
`INHALATION
`
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`
`ABRAHAM M KARKOWSKY
`07/27/2009
`
`

`

`DIVISION OF CARDIO-RENAL DRUG PRODUCTS
`
`Divisional Memo
`
`22-387 (inhaled treprostinil for pulmonary
`
`NDA:
`hypertension)
`
`Sponsor:
`
`United Therapeutics
`
`Review date: 25 April 2009
`
`N. Stockbridge, M.D., Ph.D., HFD~1 10
`Reviewer:
`Distribution: NDA 22-387
`
`HFD- 1 10/ Brurn/ Karkowsky
`
`This memo conveys the Division’s recommendation to issue a Complete Response letter
`for inhaled treprostinil for pulmonary arterial hypertension.
`Most issues have been addressed in Dr. Karkowsky’s CDTL memo (18 April 2009). I
`acknowledge revieWS by Drs. KarkOWsky (medical; 3 April 2009), Lawrence (statistics; 7
`April 2009), Kumi (clinical pharmacology and biopharmaceutics; 24 March 2009),
`Joseph (pharmacology/ toxicology; 25 March 2009), Cooper (chemistry and
`manufacturing; 24 March 2009), and Metcalfe (microbiology; 24 March 2009), and
`consults performed by the QT Interdisciplinary Review Team (30 January 2009) and Mr.
`De of CDRH (nebulizer; 1 April 2009).
`
`Treprostinil or remodulin is approved for the treatment of pulmonary arterial
`hypertension with administration by subcutaneous and intravenous routes. Both
`routes carry serious safety issues. Subcutaneous administration is very painful; in
`studies, many subjects required narcotics. The intravenous route carries the risk of
`infection.
`
`The inhaled route has been studied in a trial that established effectiveness with regard
`to 6-minute walk, but the effect is small and diminishes substantially by the end of the
`recommended interrdosing interval. Inability to dose continuously and the temporal
`variability means that the inhaled route Will not be appropriate to all who now use other
`routes of administration, but it may be adequate for some of them and certainly is a
`more convenient route on which to start treprostinil.
`
`The inhaled route moves local irritation effects to the nasopharynx and the rest of the
`respiratory tract. The available experience is up to 13 weeks; post-marketing data will
`Show how well this mode of delivery is tolerated in the long term.
`The associated OptiNeb nebulizer is not “opti”mized for this use. Patients are required
`to take three breaths on cue (else drug is delivered anyway), then power the device off
`and on, take three more breaths, power the device off and on, and take three more
`breaths. The complexities may have led to low or undetectable plasma levels of drug in
`some subjects, but the sponsor has asserted that observed low levels were appropriately
`low, because they were collected at trough.
`
`The sponsor asserts (letter dated 17 April 2009) that they can get the nebulizer re-
`engineered, a human factors study conducted with the new model, and a report
`generated for submission within one year.
`One issue critically affecting approval is that a manufacturing site
`A of the drug substance was closed in 2006 and the
`replacement site has not been approved.
`
`
`
`hgfig
`
`Treprostinz'lDiz/Memodoc
`
`'
`
`-—1——
`
`Last saved
`10:20 Saturday, April 25, 2009
`
`

`

`Divisional memo
`
`Tyvaso (Z'reprostinil)
`
`NDA 22—387
`
`Pulmonary artery hypertension
`
`Our Complete Response letter will identify three deficiencies:
`
`1. The sponsor must resolve all remaining issues with manufacture of the drug
`substance.
`
`2. The sponsor mu st address unresolved issues with biocompatibility of nebulizer
`parts that come into human contact.
`
`3.
`
`I am willing to have the existing device on the market for a short period while the
`changes are made. There are no real engineering uncertainties that could make the
`process take longer than expected, so the sponsor will be asked to provide a
`timetable for their response, along with appropriate landmarks, and a compelling
`demonstration of their commitment to meet the timetable.
`
`Labeling still needs to be negotiated. This will proceed while the sponsor is addressing
`deficiencies.
`
`TreprostinilDiuMemo.doc
`
`‘
`
`—2—
`
`I
`
`Last saved
`10:20 Saturday, April 25, 2009
`
`

`

`This is a representation of an electronic record that was signed electronically and
`this page is the manifestation of the electronic signature.
`
`Norman Stockbridge
`4/25/2009 10:29:43 AM
`MEDICAL OFFICER.
`
`

`

`
`
`Tyvaso®, NDA 22-3 87 CDTL memo, Abraham M. Karkowsky 4/19/2009 page 1
`
`MEMORANDUM
`
`DEPARTMENT OF HEALTH & HUMAN SERVICES .
`Public Health Service
`
`Food and Drug Administration
`Center for Drug Evaluation and Research
`
`
`DATE:
`
`April 18, 2009.
`
`FROM:
`
`Abraham Karkowsky, M.D., Ph.D. Group Leader, Division of
`Cardiovascular and Renal Products, HFD-l 10.
`
`TO
`
`Dr. Norman Stockbridge, M.D., Ph.D., Director, Division of
`Cardiovascular and Renal Products, HFD—l 10.
`
`SUBJECT: Complete response memo to Tyvaso® (inhaled treprostinil NDA 22—3 87,
`United Therapeutics Corporation).
`
`'
`
`This memo outlines the current deficiencies in the application for inhaled
`treprostinil NDA 22—3 87. I therefore, recommend that a complete response letter be
`transmitted to the sponsor. The specific deficiencies for this application and the remedy
`for these deficiencies are summarized in this memo. I have also included in this memo,
`some labeling considerations.
`
`This subject of this application consists of an active drug (treprostinil sodium for
`inhalation) and an inhalation device the OptiNeb nebulizer. The two can only be
`approved as a single entity. The drug has not been assessed in controlled clinical trials
`when administered by any other nebulizer and the nebulizer is not approved as a general
`purpose nebulizer.
`
`The complexities in both the assembly and use of the device as well as the
`cumbersome nature of administering the appropriate dose (the counter only counts down
`from “3” then the patient must reset this device and repeat the process an'additional two
`times), makes it important to assess how reliably an individual can learn and perform the
`complex processes in putting together, cleaning and administering the active drug
`substance. In addition, the biopharmaceutic reviewer noted that there were several
`subjects who either had no active drug measured or had very low measurements of these
`concentrations. Whether these low concentrations of drug reflect an inability of the user
`to accurately administer the drug or whether these subjects had inordinate variability in
`the absorption of the drug is unclear.
`
`A human factor study is therefore, needed. The basis of the information to be
`collected would depend upon those processes that pose the greatest likelihood of
`provoking harm or that are most likely to diminish any benefit of the drug-delivery
`system. The human factor study would include two separate sub-studies. One study (in
`normals without the administration of medication) to assess the physical process of
`cleaning, assembling and pseudo—administering the drug with particular attention to the
`
`

`

`Tyvaso®, NDA 22-387 CDTL memo, Abraham M. Karkowsky 4/19/2009 page 2
`
`process that are most critical for the safe and effective use of the drug system. The second
`study would assess whether the drug itself is effectively inhaled (based on serum
`concentrations) and whether under optimum learning conditions, more reliable exposure
`to treprostinil is obtained.
`
`
`Several biocompatibility tests are still incomplete. These include the implantation
`tests for the medicine cup
`and sealing ring materials _ «\ mouse lymphoma
`and chromosomal aberration tests for the.
`,-.-..-
`mouthpiece material. Furthermore,
`the tests which the sponsor has previously completely still need to be submitted as a
`completed study report. These deficiencies should be corrected, and the reports submitted
`prior to approval.
`
`13(4)
`
`There are additional issues that Will be included into the complete response. The
`drug substance information for NDA 22—387 is referenced to the subcutaneous
`
`formulation of treprostinil (NDA 21—272). The sponsor submitted a supplement for a new
`treprostinil drug substance manufacturing facility and"
`process ‘
`
`_
`:. The previously approved site for that process was closed in 2006. Since
`treprostinil for inhalation cannot be currently manufactured until the referenced NDA
`supplement is approved, treprostinil for inhalation cannot currently be approved.
`
`M4)
`
`The clinical reviewer (me) noted that the effect of treprostinil in increasing walk-
`distance was small. The benefit was about half that of the other approved prostacyclin for
`inhalation'(Ventavis®), in a broadly equivalent population. There is no reason to believe
`(no controlled data) that the inhaled route of administration would supplant either the
`subcutaneous or the intravenous route for the administration of treprostinil. There is also
`no reason to believe that there would be added effects on adding the inhaled drug to the
`drug when administered by the SC or IV route.
`
`Since the parentral (SC or IV) routes of drug delivery is more flexible than when
`delivered by the inhaled route, I can’t recommend substituting the inhaled route for
`treprostinil as an alternative to long term treatment with the parentral routes.
`When treprostinil is administered by the parentral route concentrations are fairly stable
`(input is constant). For the inhaled route, however, concentrations in plasma are not
`constant and likely wane at the site of action (the pulmonary vascular sites) during the
`interdosing interval. Consequently, it is likely that with the inhalation route the benefit is
`asymmetric during the dosing interval with greater effects early on and lesser effect just
`immediately prior to the next dose. For these reasons, the lack of flexibility in dose and
`the inconstant effect during the dosing interval,
`the inhaled treprostinil would not
`necessarily be an alternative to the subcutaneous route of administration.
`
`There is only a modest benefit of inhaled treprostinil, limited to a benefit on peak
`6MWD (six-minute walk distance). This benefit, however, wanes at the interdosing
`interval, with the loss of approximately 30% of the effect at trough at lZ-weeks compared
`to the peak effect at lZ-weeks. Furthermore, the persistence of benefit to patients in
`excess of 12—weeks studied in this development program is unclear.
`
`

`

`Tyvaso®, NDA 22—387 CDTL memo, Abraham M. Karkowsky 4/19/2009 page 3
`
`The down—side of this drug is mostly related to site—related irritation. This
`irritation is not surprising given the near universal pain at the infusion site when
`treprostinil is administered by the subcutaneous route. When treprostinil is administered
`by the inhalation route, drug delivery is irritative to the nasopharynx, oropharynx and
`lungs. Animal studies demonstrate that after l3—weeks of treatment with a 4 week
`washout there were residual lesions in the oropharynx, and lung in both species and the
`heart in rats.
`
`There does not appear to be an adequate database to currently assess the relative
`respiratory and cardiac hazard of treprostinil by the inhaled route, particularly when
`compared to modest benefit in walk distance. Adverse events in the treated group during
`the double blind and open label phase included those related to the oro- and naso-pharynx
`and respiratory tree including; cough, throat irritation, pharyngeal pain, epistaxis,
`hemoptysis and wheezing. These events were greater among treated than placebo
`patients.
`
`Serious adverse events during the open—label portion of the study included
`pneumonia (in 8 subjects). There were three serious episodes ofhemoptysis noted during
`the double-blind and open—label experience. One was lethal and two required invasive
`maneuvers to stem bleeding.
`
`It should however, be noted that similar hemoptysis events were noted when
`treprostinil was administered parentrally.
`
`Given the modest database and small benefit, as well as the pro-clinical studies
`indicating irreversible lesions in the oropharynx and respiratory tree, additional data
`should be gathered perhaps as a post—marketing commitment.
`
`There are several limitations to the use of this drug, which should be included
`within labeling. The current application contains a single placebo—controlled study that
`demonstrated that the inhalation of treprostinil increases walk distance relative to placebo
`in WHO type I patients with functional class NYHA III who are on stable doses of
`bosentan or sildenafil. The benefit is more obvious when the concomitant medication was
`bosentan. Dosing is limited to the 9 puffs (breaths), approximately 54 lag/dosing four
`times a day, with no dosing at night. There is insufficient experience with higher doses to
`warrant including this recommendation in the label.
`
`The effect on repolarization (QT effect) is small and dissipates rapidly, as the
`concentration of treprostinil wanes. These effects should also be included within labeling.
`
`The drug has been granted “orphan” status and as such, pediatric studies are
`waived.
`
`Since I did the original medical review, I will refer you to that review for additional
`information. In addition, the following reviews were consulted in the construction of this
`memo.
`
`

`

`Tyvaso®, NDA 22-387 CDTL memo, Abraham M. Karkowsky 4/19/2009 page 4
`
`o
`
`Pharmacology/Toxicology review by Dr. Xavier Joseph D.V.M., dated March 25,
`2009.
`
`0
`
`0 Clinical Pharmacology and Biopharrnaceutic Review by Robert O Kumi, Ph.D.,
`dated march 24, 2009.
`Statistical review by John Lawrence, Ph.D., dated April 7, 2009.
`A single DSI audit dated January 8, 2009 by Tejashn' Purohit—Sheth, regarding the
`inspection of Dr. Robert C. Bourge, M.D., University of Alabama.
`0 Thorough QT study-team review by Drs. Atul Bhattaram, Qianyu Dang, Joanne
`Zhang, Suchitra Balakrishnan and Christine Garnett dated January 30, 2009.
`Chemistry review by Monica D. Cooper, Ph.D., dated March 23, 2009.
`Proprietary name review by Judy Park, PharrnD., dated, February 18, 2009.
`Microbiology review by John W. Metcalfe, Ph.D., dated March 23, 2009.
`Regulatory device consult review by Sugato De, Biomedical Engineer, CDRH
`dated April 1, 2009. The CDRH consult incorporated comments on the human
`factor study review by Ron Kaye (ODE/DAGID/ GHDB).
`
`Proprietary name:
`The proprietary name Tyvaso® is acceptable. The name was not found to be
`easily amenable to provoke mediation errors.
`
`Chemistry:
`The drug substance is currently approved for subcutaneous and intravenous
`administration (NDA 21,272). That application, however, has a currently pending
`supplement for a treprostinil drug substance manufacturing site (NDA 21-272/SCM-010).
`The old manufacturing site ———————--——~ was closed in 2006. Consequently, the
`drug substance for this application cannot be acceptably produced and until the
`supplement for NDA 21-272 is approved.
`
`33(4)
`
`The chemist recommended a 36-month shelf life if stored at controlled
`temperature (room temperature) and protected from light (in the foil pouches as
`proposed).
`
`Microbiology:
`The product is sterilized Mum-\W‘: prior to final seal.
`There were no microbiology deficiencies.
`
`Device:
`
`M4)
`
`The device performs acceptably. The output of the nebulizer during four tests of
`
`three cycle assessments averaged
`- close to the ' 7/ breath cycle which is the
`labeled output of the nebulizer. The distribution of the particle sizes averaged ~ ,
`which the CDRH reviewers considered acceptable.
`
`Of note, the output of the nebulizer was diminished when the amount of drug in
`the nebulizer cup was less than the --—~ dispensed in a single ampoule (it was slightly
`
`[3(4)
`
`

`

`Tyvaso®, NDA 22-387 CDTL memo, Abraham M. Karkowsky 4/l 9/2009 page 5
`
`diminished when only W" was in the cup). The diminished delivery with lower cup
`volumes should be included in the package insert.
`
`{3‘4}
`
`Several biocompatibility tests are still incomplete. These were described above.
`
`Also as noted above, the human factor study including the prioritization of user
`tasks (that is the tasks most susceptible to compromising either the safe or effective use
`of the combination drug—device product), is still pending. Given the complex processes in
`the assembly and administration of the inhaled drug, such a study is required prior to the
`approval of the application.
`
`_
`Pharmacology/toxicology:
`The current submission bridges the information from the use of treprostinil by
`either the SC or IV route of administration to the inhalation route as proposed here. One
`short-term and two longer-term studies one in rat and one in dog were performed to
`assess the specific toxicity of the new route of treprostinil administration. The key
`observations are the effects of aerosolized treprostinil on the naso—pharyngeal, respiratory
`tracts and myocardium (in rats).
`
`In the one short-term study when nebulized treprostinil sodium, at concentration
`of 50 pg/L, was administered to male rats for 2, 3 and 4 hours with estimated exposure of
`300, 416 and 569 pig/kg via nose—only inhalation, there was a significant decrease in
`respiratory rates and minute volumes (baseline controlled). The change reverted 24 hours
`after completion of the exposure.
`
`Longer term exposure (13 weeks) in rats exposed to aerosol by nose only at the
`following exposures 7.1 ug/L for 20 min/day, 44 pg/L for 30 min/day and 40.3 ug/L for
`225 minutes/day [total exposure of 7 (low), 67 (mid) and 464 (high) ug/kg/dayg] control
`animals were not exposed to any inhalation therapy.
`
`Lesions in the respiratory tract included: squamous metaplasia in the larynx (all
`doses); hemorrhage and macrophage accumulation in the lungs (all doses);
`hyperplasia/hypertrophy of goblet cells in the nasal cavity (all doses);
`degeneration/regeneration of the respiratory epithelium in the nasal cavity (mid and high
`doses), respiratory epithelial ulceration (high dose) and finally olfactory epithelial
`degeneration in the nasal cavity (high doses). Also noted was myocardial
`degeneration/fibrosis (all doses). There were degenerative changes observed in the testes
`and adrenal adrenal glands. Of these changes, only the changes in the epididymal lesions
`appeared reversible.
`
`In dogs 13-weeks exposure by the oro-nasal route of inhalation at exposures of
`0.025 mg/L for 15 minutes and 0.224 mg/L for either 6 or 30 minutes daily for 13 weeks,
`with estimated exposure of 107 (low), 322 (mid) and 1558 (high) ug/kg/day. Microscopic
`related lesions in the respiratory tract in nasal cavity and larynx in the mid and high doses
`groups included focal or multifocal respiratory epithelial degeneration/regeneration in the
`nasal cavity (mid and high dose). Other lesions included goblet cell
`
`

`

`Tyvaso®, NDA 22-387 CDTL memo, Abraham M. Karkowsky 4/19/2009 page 6
`
`hyperplasia/hypertrophy (high dose), ulceration in the squamous and respiratory
`epithelium (high dose) and degeneration/necrosis in the squamous epithelium (high
`dose); degeneration in the larynx (mid and high doses). Lung hemorrhage was observed
`in 1 of 6 low dose dogs but this incidence was within the historical experience.
`
`The Cmax in venous blood, of the low doses of both dogs and rats are less than a
`log-unit higher than generated by the humans by the inhalation route.
`
`Biopharmaceutics:
`The current information for treprostinil inhaled relies heavily on the labeling of
`the subcutaneously administered route.
`
`Treprostinil when administered by the inhaled route is rapidly detected in the
`plasma. The Tmax of the treprostinil is between 7—15 minutes after the end of the
`inhalation. The concentrations rapidly dissipate with a Tug of approximately 45 minutes
`to 1 hour. In a cross-over study comparing the AUC of a single dose of 3 or 6 breaths to a
`dose of 15 ng/kg/min administered intravenously for 60 minutes, the absolute
`bioavailability based on venous blood was determined to be between 62-74%.
`
`Treprostinil by the oral route (treprostinil ethanolamine) was studied in five drug-
`drug interaction studies. There were no interactions with either bosentan or sildenafil.
`When administered with gemfibrozil a CYPZCS inhibitor, exposure and Cmax of
`treprostinil was approximately doubled. There was a small 14% change in exposure when
`co—administered with fluconazole (a CYP2C9 inhibitor). In the presence ofrifampin,
`treprostinil’s AUC decreases by 30%.
`
`It is unlikely that any of the drug interactions would alter the effectiveness of
`treprostinil, since exposure at the active site occurs in advance of systemic exposure.
`Increase or decrease in systemic related adverse events such as vasodilation, however,
`may be provoked by the concomitant use of drugs that either increase or decrease serum
`levels of treprostinil.
`
`QT study:
`With respect to the thorough QT study, there were small increases in QTcI
`intervals‘(as well as other rate-corrected QT interval measurements) after inhalation of 14
`breaths, at the initial assessment points (approximately 5 minutes) with the effect
`dissipating fairly rapidly. The inhaled dose of treprostinil was only 55% higher than the
`recommended dose. The predicted values of QT measurements at Cmax are shown below.
`Of note, the venous plasma concentrations generated by the inhalation route of
`administration generates concentrations of approximately 12 ng/ml. Concentrations
`generated by the subcutaneous route of administration can be approximately a log unit
`higher than the concentrations generated by the inhalation route.
`
`

`

`Tyvaso®, NDA 22—387 CDTL memo, Abraham M. Karkowsky 4/19/2009 page 7
`
`Table 1: Modeled effects of inhaled treprostinil on measurements of repolarization at Cmax for the
`
`14 puff dose. Slope‘is unit change per pg/ml treprostinil concentration. .
`One—sided Upper
`95%
`Predicted '
`Standard
`Overall
`Confidence
`QTc
`Error
`Slope of
`Model
`Hound of
`at Average
`of Plasma
`Plasma
`QT
`fit
`Predicted QTc
`Cm
`p-value
`Concentration Concentration
`Parameter
`<.0001
`8.0418
`6.8977
`0.0000
`0.0040
`0.0004
`QTcl
`<.0001
`8.4696
`7.3455
`0.0000
`0.0040
`0.0004
`QTCF
`<.0001
`14.7745
`13.3221
`0.0000
`0.0065
`0.0006
`QTCB
`[1] Linear Mixed Moan 11’ [ll [of rlnnge from basellm (not phrabo—con'ecl) versus the plasma mncenlrallon as a [hard effect wllh subjed
`included in the model as a midorn effect.
`121 Upper Bound - upper Dire-sided 95% linear nixed model based Confidence llmlt,
`
`-
`
`Medical/Statistical;
`The application consisted of a single placebo-controlled, single regimen study in
`patients who were WHO class I and nearly all NYHA class III subjects who were on
`stable does of either bosentan or later sildenafil as an alternative.
`
`The randomized treatments were either placebo or treprostinil administered by the
`OptiNeb® nebulizer. The initial dose was 3 puffs (with each puff dispensing 6 ug of
`drug). Doses were repeated 4 times during waking hours at approximately 0, 4, 8 and 12
`hours. If the initial first administration was tolerated, the dose was to be gradually up-
`titrated to a target dose of 9 puffs QID.
`
`Subjects were stratified based on study center.
`
`6MWD were performed at screening (visit 1), first dose (visit 2), week 6 (visit 3)
`and immediately prior to the last dosing (visit 4) and after 4 hours following the last dose
`(visit 5). The timing of the 6MWD was to be 3 and 5 hours after a bosentan and 30—120
`minutes after a sildenafil dose (these are approximately peak effects of those therapies).
`On visits 2, 3 and 4 the 6MWT was carried out to capture peak inhalation effects (10—60
`minutes post inhalation) at visits 1 and 5 the effects will be measured at trough
`(approximately 4 hours post inhalation).
`
`The primary endpoint of the study was 6MWT at peak inhalation effect on week 12
`(visit 4). For subjects with missing information values (i.e., ranks) will be imputed based
`on whether the subject discontinued for worsening of disease or for non-disease related
`events. For those who discontinued for adverse events the algorithm employed a LOCF
`or last rank carried forward analysis. For those who discontinued due to worsening
`disease, a worst value or worst rank was imputed.
`
`Secondary end points were:
`0 Time to worseningl.
`
`
`’ Time to death, transplantation, hospitalization for PAH or transition to IV or SC prostacyclin therapy.
`
`

`

`Tyvaso®, NDA 22—387 CDTL memo, Abraham M. Karkowsky 4/19/2009 page 8
`
`The Borg Dyspnea score.
`Change in NYHA functional class as assessed by the investigator.
`Change in walk distance walked at the week 12 trough walking distance.
`Change in distance walked at the week 6 peak walking distance.
`QOL as of week 12 (Minnesota Living with Heart Failure, MLWHF
`questionnaire).
`-
`Signs and symptoms of PAH
`Troponin T and proBNP levels
`Change in walk distance at day 2 (peak walking distance after the initial dose)
`The PK of treprostinil.
`
`For the secondary endpoints, alpha was allocated sequentially. Should the hierarchy
`ofparameters no longer attain significance, the residual analyses then become
`exploratory in nature.
`-
`
`There was a difference in walk distance comparing placebo at peak
`
`
`
`The first secondary endpoint, the time to clinical worsening was not different
`between treatments. Other secondary endpoints are therefore only descriptive in nature.
`
`With respect to the effect at the interdosing interval at week 12, there was a 13
`meter difference comparing the treprostinil to placebo. This effect is approximately 2/3 of
`that observed at peak.
`
`Safety:
`
`Dining the 12-week placebo controlled study, the majority of adverse events
`could be attributed to either to the route of administration, adverse events ofprostacyclin
`use (vasodilatation and joint pain), and events associated with the pulmonary artery
`' hypertension disease process. The. adverse events during the double blind phase are
`shown below:
`
`

`

`Tyvaso®, NDA 22-387 CDTL memo, Abraham M. Karkowsky 4/19/2009 page 9
`
`
`Table 3: Adverse events and % occurrin_ durin_ the Double-blind ortion of TRIUMPH 001
`Adverse event W. Placebo
`=120
`Patient with an event*
`101 88%
`100 83%
`62 54%
`35 29%
`47 41%
`27 23%
`
`
`
`
`
`II—
`
`
`
`-
`Headache
`
`
`'— Lower res . irato
`
`Throat itation
`. Ph. nealain
`Diarrhea
`
`
`Chest discomfort
`
`10 8%
`16 14%
`[Hm—I.-
`
`flea—m—
`
`tract infection
`
`Urinary tract infection
`
`5 (4%)
`
`4:. ‘20\
`
`1 (< 1%)
`
`Pulmon-
`
`.O
`
`h-ertension
`
`[1%-
`
`*Sum of adverse events gre