CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`
`APPLICATION NUMBER:
`
`214324Orig1s000
`
`NON-CLINICAL REVIEW(S)
`
`
`
`
`
`
`

`

`
`
`
`
`
`
`
`DEPARTMENT OF HEALTH AND HUMAN SERVICES
`PUBLIC HEALTH SERVICE
`FOOD AND DRUG ADMINISTRATION
`CENTER FOR DRUG EVALUATION AND RESEARCH
`
`
`PHARMACOLOGY/TOXICOLOGY NDA/BLA REVIEW AND EVALUATION
`
`Application number:
`Supporting document:
`Applicant’s letter date:
`CDER stamp date:
`Product:
`Indication:
`
`214324
`2
`4/16/2021
`4/16/2021
`Tyvaso DPI™ (treprostinil) Inhalation Powder
`Pulmonary arterial hypertension (PAH) and
`pulmonary hypertension associated with
`interstitial lung disease (PH-ILD)
`United Therapeutics Corporation (UTC)
`Applicant:
`Division of Cardiology and Nephrology (DCN)
`Clinical Review Division:
`Pharm/Tox Division: Division of Pharm/Tox for Cardiology,
`Hematology, Endocrinology, and Nephrology
`(DPT-CHEN)
`Baichun Yang
`Reviewer:
`Xuan Chi
`Supervisor/Team Leader:
`Norman L Stockbridge
`Clinical Division Director:
`Brian Cooney
`Project Manager:
`Template Version: September 1, 2010 (Modified by DCRP: June 10, 2013)
`
`Disclaimer
`
`Except as specifically identified, all data and information discussed below and
`necessary for approval of NDA 214324 are owned by UTC or are data for which UTC
`has obtained a written right of reference. Any information or data necessary for approval
`of NDA 214324 that UTC does not own or have a written right to reference constitutes
`one of the following: (1) published literature, or (2) a prior FDA finding of safety or
`effectiveness for a listed drug, as reflected in the drug’s approved labeling. Any data or
`information described or referenced below from reviews or publicly available summaries
`of a previously approved application is for descriptive purposes only and is not relied
`upon for approval of NDA 214324.
`
`Reference ID: 4814268
`
`1
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`TABLE OF CONTENTS
`
`1
`
`EXECUTIVE SUMMARY.......................ccccccccececeseeeeeceeeeceeeceeecceeccesseceseeeeecesscesseeseeeees 3
`
`1.1.
`INTRODUCTION (AND CLINICAL RATIONALE) ......2.20.2.c.ccceceseseseececeseeeeeececeeeeeesceeeees 3
`1.2
`BRIEF DISCUSSION OF NONCLINICAL FINDINGS..........20:cceccccceeceseseseevececeeeeeeveveseeeees 3
`1.3.
`RECOMMENDATIONS..0......0.0.cccccccecescseecececescececececescecececeecececececeseecevsteceesevsvsceseesees 4
`
`1.3.1 Approvabilitycee 4
`1.3.2
`Additional Non-Clinical Recommendations ...................ee 4
`
`Labeling .......eeees 4
`1.3.3
`
`2
`
`DRUG INFORMATION ....00 0.02.0... e cece ceeeceeecceeeceeeeceeeeceseeeenceneeceseceseceesceeseceseceeeeeeees 5
`
`DRUG 000. eecccec cece cence ce eeceeecceeccueccueeceececeeceuececeeccueecseecseecceecceeeeeeecseeceueceseeceseeeees 5
`2.1
`RELEVANT INDs, NDAs, BLAS AND DMB6S..............0. 000. 00.ccccccccecceeceeceeececceeeeeecueees 5
`2.2
`DRUG FORMULATIONoo... .cccccccccecececececescececessecececeevecececscueveecevececeseecevsceceeseeees 5
`2.3
`COMMENTS ON NOVEL EXCIPIENTS ...........ccccccccccecececesescecececeecscececeesececsceceececeesceces 6
`2.4
`COMMENTS ON IMPURITIES/DEGRADANTS OF CONCERN.........cc+ecccceseceseseeeececeeeeeeees 6
`2.5
`2.6|PROPOSED CLINICAL POPULATION AND DOSING REGIMEN...........2.c0:e:0seeceeseseseseeeee 6
`2.7
`REGULATORY BACKGROUND ...0.....2.2cccccsesescecesescecececeevecececececevececeeevsvececeseeeeesceses 6
`
`3
`
`STUDIES SUBMITTED AND CROSS-REFERENCED ......................ccccceeeeeeeeeeeeeeees 6
`
`3.1
`3.2
`3.3.
`
`STUDIES REVIEWED.........c.c.cccccccecescececececcececececsesevececeesevevecececseevevececeeceveveceececeees 6
`STUDIES NOT REVIEWED 20........cccscccccceceseecscececeseececececeeececececececececeseeeeceseseteees 6
`STUDIES CROSS-REFERENCED 200.0...0.c.cccecececcecscececeececececsecevecececeecevececeesevsveveveseeess 7
`
`4
`
`PHARMACOLOGY|... wi... ccccccccccceeceeceecceeccecceeccecceeceeecesceseeeecesceeecesceeeceseeseeeeceeeeeseees 7
`
`4.1
`
`PRIMARY PHARMACOLOGY..........22.:cscsesceseceesesceseceesecscececeevevscscecsececsceceseecesececeees 7
`
`5
`
`PHARMACOKINETICS..0....sssesssseecssesssseecsuecssessseessutessueessueesssesssessutessueessneesaseeses 12
`
`SPECIAL TOXICOLOGYSTUDIES. .......... 0.0... cece eeccceeeceeeeeeeeceeeeeeeeeeeceeeeeeeseeeees 15
`10
`© AND
`10.1
`®©® (2020) COMPUTATIONAL EVALUATION OF THE
`POTENTIAL BACTERIAL MUTAGENICITY OF IMPURITIES POTENTIALLY ASSOCIATED
`WITH TREPROSTINIL INHALATION POWDER.........2.:cececescececececeseececececeecscececeeeecececeseseeesees 15
`10.2
`®©® (2021) TOXICOLOGICAL PROFILE AND RISK ASSESSMENTFOR
`®©@|N INHALED TREPROSTINIL DRUG PRODUCT (TYVASO DPI
`INHALATION POWDER)..........2+s-cece+eseseececeececececsevevevececeevevsceceeveveveveveveceeseveveveceesececeteceees 18
`
`11
`
`INTEGRATED SUMMARY AND SAFETY EVALUATION.........................220000e 19
`
`Reference ID: 4814268
`
`2
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`1
`
`Executive Summary
`
`1.1.
`
`Introduction (and Clinical Rationale)
`
`Treprostinil is a prostacyclin analogue. The major pharmacologic actions of treprostinil
`are direct vasodilation of pulmonary and systemic arterial vascular beds and inhibition of
`platelet aggregation. Inhaled treprostinil therapy provides selectivity of the
`hemodynamiceffects to the lung vasculature, thus reducing systemic side effects
`comparedto other routes of administration. Following inhalation of prostacyclin analogs,
`pulmonary artery pressure decreases, and systemicarterial pressure is stable.
`
`The active pharmaceutical ingredient (API) in Tyvaso DPI" is identical to the treprostinil
`drug substance approved in Remodulin (NDA 021272) and Tyvaso (NDA 022387). It is
`also the same active moiety as the treprostinil diolamine drug substance approvedin
`Orenitram (treprostinil) Extended-Release Tablets (NDA 203496). The safety and
`efficacy of treprostinil are supported by a comprehensive set of pharmacology,
`pharmacokinetic (PK), toxicology, and clinical studies conducted for Tyvaso (treprostinil)
`Inhalation Solution (NDA 022387), Remodulin (treprostinil) Injection (NDA 021272), and
`Orenitram (treprostinil) Extended-Release Tablets (NDA 203496).
`
`The safety of the excipient fumaryl diketopiperazine (FDKP) is supported by a
`comprehensive battery of safety pharmacology and toxicology studies conducted for
`Afrezza (BLA 022472).
`
`1.2
`
`Brief Discussion of Nonclinical Findings
`
`In a rat PK study following intermittently delivery of treprostinil into the tracheal tube,
`treprostinil Tmax was <5 minutes. Treprostinil Inhalation Powder(TriP) provided much
`higher treprostinil exposure than the nebulized reference solution. There was no
`evidencethat the excipient FDKP, present in Tyvaso DPI, interfered with the absorption
`of treprostinil in the lung.
`
`In studies to assess the potential effects of a Tyvaso DPI impurity
`er)
`in vitro, the stimulating activity of
`®©at prostanoid receptors IP1,
`EP2, DP1, and EP1in tested cell lines were approximately 23-, 10-, 14-, and 25-fold|
`Pit!respectively, than treprostinil.
`’ in humanliver microsomes and hepatocytes, with T1/2 < f min. In
`©)‘little to no
`© was present at 0 minutes, suggestingthat
`
`at time zero was also observed ( ~"%).
`
`The potential bacterial mutagenicity of 7 impurities identified in Tyvaso DPI was
`evaluated by (quantitative) structure-activity relationship ((Q)SAR) using Derek Nexus
`
`Reference ID: 4814268
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`and Leadscope Model Applier. All 7 impurities (
`(b) (4) and
`o©@ were
`identified as inactive (non-mutagenic) in DEREK Nexus and negative (non-mutagenic)
`in the Leadscope Model Applier. None of the structures had any structural features
`which wereidentified as unclassified/misclassified or out of domain, respectively,in
`Derek Nexus or Leadscope.As such, all 7 impurities are considered Class 5 impurities
`and maybe treated as non-mutagenic.
`
`(b) (4)
`
`(b) (4)
`
`®®is identified as an impurity in Tyvaso DPI at a concentration
`above the qualification threshold of 1% for drug products with a maximum daily dose of
`drug substancethat is <10 mg.
`” was qualified at a concentration
`of up to 4% in the final drug product.
`®™® Was shownto have reduced
`pharmacodynamic effects compared to treprostinil, and it is expected to be
`It is unlikely that the presence of
`in Tyvaso DPI would be a safety concern for patients, if present at up to 5%.
`(4
`
`(b) (4)
`
`Basedon the results of the nonclinical pharmacology, PK, and toxicology studies
`conducted to support Tyvaso (NDA 022387), Remodulin (NDA 021272), and Orenitram
`(NDA 203496), and the assessmentof impurities present in Tyvaso DPI, it is considered
`that Tyvaso DPI has an acceptable safety profile and that there are no findings that
`preclude long-term inhalation administration in humans.
`
`1.3
`
`Recommendations
`
`1.3.1. Approvability
`
`Approvable
`
`1.3.2 Additional Non-Clinical Recommendations
`
`None
`
`1.3.3
`
`Labeling
`
`Wesuggestfollowing changes(bold for insert and cross out for delete):
`
`(1) Under INDICATIONS AND USAGE onpage1
`
`Tyvaso DPIis a prestaeyehnmimetic
`
`©‘indicated for the treatment of
`
`(2) To second paragraph undersection 13.1
`
`bhi
`
`Oral administration of treprostinil diolamine to Tg.rasH2 miceat 0, 5, 10, and 20 mg/kg/day in males and
`0, 3, 7.5, and 15 mg/kg/day in females daily for 26 weeks did not significantly increase the incidence of
`tumors.
`
`Reference ID: 4814268
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`2
`
`Drug Information
`
`2.1.
`
`Drug
`
`CAS Registry Number: 81846-19-7
`
`Generic Name: Treprostinil
`
`Code Name: UT-15
`
`Chemical Name: [[(1R,2R,3aS,9aS)-2,3,3a,4,9,9a-hexahydro-2-hydroxy-1-[(3S)-3-
`hydroxyoctyl]-1H-benz[flinden-5-ylloxy]acetic acid
`
`Molecular Formula / Molecular Weight / Structure :
`
`OH C>3H33405
`
`OCH,CO,H
`
`Mol. Wt.: 390.52
`
`Pharmacologic Class: Prostacyclin vasodilator
`
`2.2
`
`Relevant INDs, NDAs, BLAs and DMFs
`
`NDA 022387 (Tyvaso), NDA 021272 [Remodulin® (treprostinil) Injection], NDA 203496
`[Orenitram® (treprostinil) Extended-Release Tablets], IND 134582 (Tyvaso DPI),BLA
`022472 [Afrezza® (insulin human) Inhalation Powder], and DMF
`
`2.3.
`
`Drug Formulation
`
`Tyvaso DPI is targeted to contain 10 mcgoftreprostinil per mg of powder. Tyvaso DPI
`is filled into unit-dose cartridges to contain 16 mcg, 32 mcg, 48 mcg, or 64 mcg of
`treprostinil per cartridge. The 16 mcg, 32 mcg, 48 mcg, and 64 mcg cartridges have
`nominalfill weights of 1.6 mg, 3.2 mg, 4.8 mg, and 6.4 mg of Tyvaso DPI, respectively.
`The composition of Tyvaso DPI per cartridge is shownin the table below.
`
`Component
`L6 mE
`3.2 mg
`4.8 mg
`6.4 mg
`P
`Cartridge
`Cartridge
`Cartridge
`Cartridge
`
`
`
`
`0.016 mg
`0.032 mg
`0.048 mg
`0.064 mg
`
`
`
`
`Treprostiml
`Fumaryl
`Diketopiperazine (FDKP)
`
`(b) (4)
`
`Reference ID: 4814268
`
`

`

`
`
`Reviewer: Baichun Yang
`
`NDA 214324
`
`2.4 Comments on Novel Excipients
`
`None
`
`2.5 Comments on Impurities/Degradants of Concern
`
`Acceptable
`
`2.6 Proposed Clinical Population and Dosing Regimen
`Patients with pulmonary arterial hypertension or with pulmonary hypertension
`associated with interstitial lung disease will use 4 separate treatment sessions each day
`approximately 4 hours apart, during waking hours. Initial dosage is one 16 mcg
`cartridge per treatment session. Dosage should be increased by an additional 16 mcg
`per treatment session at approximately 1- to 2-week intervals, if tolerated.
`
`2.7 Regulatory Background
`Written responses for a preNDA meeting were issued on 11/19/2020. The agent stated
`that “From a clinical perspective and pharmacology/toxicology perspective your planned
`NDA Table of Contents is acceptable”.
`
`
`
`Studies Submitted and Cross-referenced
`
` 3
`
`3.1 Studies Reviewed
`3515 EVALUATION OF COMPOUNDS AGAINST HUMAN G PROTEIN-COUPLED
`RECEPTORS
`3538 EVALUATION OF COMPOUNDS AGAINST HUMAN G PROTEIN-COUPLED
`RECEPTORS
`MKC-PC-2018-004 Determination of the in vivo pharmacokinetics of an inhaled
`Treprostinil dry powder and nebulized Treprostinil in Rat
`20UNITP1S5
`STABILITY IN HUMAN LIVER MICROSOMES STABILITY IN
`HUMAN HEPATOCYTES
` (2020) COMPUTATIONAL EVALUATION OF THE POTENTIAL
`BACTERIAL MUTAGENICITY OF Impurities Potentially Associated with
`Treprostinil Inhalation Powder
` (2021) Toxicological Profile and Risk Assessment for
`Inhaled Treprostinil Drug Product (Tyvaso DPI Inhalation Powder)
`
` in
`
`
`3.2 Studies Not Reviewed
`None
`
`
`Reference ID: 4814268
`
`6
`
`(b) (4)
`
`(b) (4)
`
`(b) (4)
`
`

`

`
`
`Reviewer: Baichun Yang
`
`NDA 214324
`
`3.3 Studies Cross-referenced
`The safety and efficacy of treprostinil is supported by a comprehensive set of
`pharmacology, pharmacokinetic (PK), toxicology, and clinical studies conducted for
`Tyvaso (treprostinil) Inhalation Solution (NDA 022387), Remodulin® (treprostinil)
`Injection (NDA 021272), and Orenitram® (treprostinil) Extended-Release Tablets (NDA
`203496). The active pharmaceutical ingredient (API) in Tyvaso DPI is identical to the
`treprostinil drug substance approved in Remodulin (NDA 021272) and Tyvaso (NDA
`022387). It is also the same active moiety as the treprostinil diolamine drug substance
`approved in Orenitram (treprostinil) Extended-Release Tablets (NDA 203496). A
`tabulated list of all nonclinical studies that are cross referenced from Tyvaso (NDA
`022387), Remodulin (NDA 021272), and Orenitram (NDA 203496) is presented
`in Module 1 Section 1.4.4 of this application.
`
`The clinical and toxicology data for the excipient, fumaryl diketopiperazine (FDKP), is
`cross-referenced from MannKind Corporation’s approved drug product, Afrezza®
`(insulin human) Inhalation Powder (BLA 022472) (Letter of Authorization in Section
`1.4.2).
`
`All these nonclinical studies were previously reviewed under the referred NDAs or BLA
`and are not re-reviewed under current NDA.
`
`Pharmacology
`
` 4
`
`
`
`4.1 Primary Pharmacology
`MOA: Treprostinil is a prostacyclin analogue. The major pharmacologic actions of
`treprostinil are direct vasodilation of pulmonary and systemic arterial vascular beds and
`inhibition of platelet aggregation. Inhaled treprostinil therapy provides selectivity of the
`hemodynamic effects to the lung vasculature, thus reducing systemic side effects
`compared to other routes of administration. Following inhalation of prostacyclin analogs,
`pulmonary artery pressure decreases, and systemic arterial pressure is stable.
`
`4.1 EVALUATION OF COMPOUNDS AGAINST HUMAN G PROTEIN-COUPLED
`RECEPTORS (3515)
`
`This study (3515) tested 12 compounds, treprostinil, and 1 control agonist for 3 GPCRs
`(DP1, EP2, IP1) using cAMP assay and 1 GPCR (EP1) using calcium assay (Table 1).
`
`Reference ID: 4814268
`
`7
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`PHPCompoundMae
`
`Table 1. Cell lines, assays, control agonists, and tested compounds
`[spss[Taree[rarest|oniert[asa|Courtageis
`ines[_DPI|HERaoaT|—C1200—See
`
`|Human|Pl|CHO-KI|1206-1|cAMP|___Toprost_
`|Human|EPI|HEK293T|C120la__|Calcium_|__Tloprost__—
`
`Treprostinil
`
`Control agonists for all 4 prostanoid receptors (DP1, EP2, IP1, EP1) showed dose-
`dependentstimulation in the receptor expressing cells with expected ECso values
`(Figure 1, Figure 2). Treprostinil and some compoundsdisplayed dose-response
`agonist activity
`in the receptors tested (Figure 1, Figure 2). Activity of the impurity
`at the IP1, EP2, DP1, and EP1 receptors were approximately 23-,
`ower, respectively, than treprostinil (Figure 1, Figure 2).
`
`-,
`
`,
`
`14-,
`14-, an
`
`Reference ID: 4814268
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`Figure 1. CAMP assay in agonist mode: Dose-dependentstimulation of agonist-induced
`intracellular CAMP accumulation in human EP2, IP1, or DP1 receptor-expressing cells
`(modified from the submission)
`
`HEK293T DP1, Agonist Mode with Compounds, cAMP, Plate 1
`4000
`8
`eer
`_
`om
`S 2008]
`a
`z
`_
`~
`3
`ss
`‘2 1000
`a
`=
`i
`a
`*
`pe
`-
`7”
`
`0
`16-15-14 -13-12-11-10 -9 -8 -7 -6 -5 -4
`Log [Compound] M
`
`HEK293T EP2, Agonist Mode with Compounds, cAMP,Plate 2
`2500
`8
`a re
`>. 2000:
`-_
`=
`Sal
`3 15004
`~
`-
`8 1000:
`a
`5
`2
`=
`5 500
`=
`~
`>
`>
`~~
`
`—— @
`
`© 1000
`2°
`
`43 12-11-10 9 8 7 6 5 4
`
`CHO-K1 IP1, Agonist Mode with Compounds, cAMP,Plate 3
`
`=e Pernt
`
`8
`s
`x
`g
`3
`2
`é
`
`Reference ID: 4814268
`
`
`
`
`
`
`
`3600:
`
`FRET cAMPStandard Curves
`
`~* HBSSBuffer, Plate 1
`“@ IBMX HBSS Buffer, Plate2
`-* IBMX HBSS Buffer, Plate 3
`
`3
`ante
`=
`2000
`iB 1500
`= s000
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`Figure 2. Calcium assay in agonist mode: Dose-dependentstimulation of Calcium flux in
`human EP1 receptor-expressing cells (from the submission).
`
`HEK293T EP1, Agonist Mode with Compounds,Calcium, Plate 4
`6000
`
`
`
`Ratio665/620x10,000
`
`e
`
`§
`
`-@ lloprost
`== Treprostinil
`
`(b) (4)
`
`4000
`
`2000
`
`0
`-15 -14 -13 -12 -11-10 -9
`
`-8
`
`-7 -6 -5 -4
`
`Log [lloprost] M
`
` Eieettaettt
`
`enaea
`lloprost
`2.327e-010
`
`>a0
`
`| = 00 | 5.490e-008(b) (4)
`
`Treprostinil
`
`AVG Buffer
`
`EVALUATION OF COMPOUNDS AGAINST HUMAN G PROTEIN-COUPLED
`4.2
`RECEPTORS(3538)
`
`®®treprostinil, and 1 control agonist for 3
`This study (3538) tested
`GPCRs (DP1, EP2, IP1) using cAMP assay and 1 GPCR (EP'1) using calcium assay.
`Cell lines, assays, and control agonists arelist in Table 1.
`
`Control agonists for all 4 prostanoid receptors (DP1, EP2, IP1, EP1) showed dose-
`dependentstimulation in the receptor expressing cells with expected ECs0 values
`(Figure 3, Figure 4). Both treprostinil and
`®™®displayed dose-
`(b) (4) at
`response agonist activity in the receptors tested, and activity of
`the IP1, EP2, DP1, and EP1 receptors were approximately 23-, 10-, 14-, and 25-fold
`
`lower, respectively/(Figure 3, Figure 4). Similar to the results in study 3515, activity of
`® at the IP1, EP2, DP1, and EP1 receptors were lowerthanthat of
`treprostinil (Figure 3, Figure 4).
`
`Reference ID: 4814268
`
`10
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`Figure 3. CAMP assayin agonist mode: Dose-dependent stimulation of agonist-induced
`intracellular CAMP accumulation in human EP2, IP1, or DP1 receptor-expressingcells
`(modified from the submission)
`
`HEK293T EP2, Agonist Mode with Compounds, cAMP
`4000
`
`v
`
`©
`
`-@ lloprost
`== Treprostinil
`a
`
`v AVG Buffer
`
`(b)(4)
`
`3
`So
`+
`So
`% 3000
`°
`3wo
`$ 2000
`
`1000
`-14 -13 -12 -11-10 -9 8 -7 6 -5 -4 -3
`
`2&
`
`HEK293T DP1, Agonist Mode with Compounds, cAMP
`
`¥
`
`-~ PGD2
`«= Treprostinil
`"
`= AVG Buffer
`
`) )
`
`
`
`Ratio665/620x10,000=nywoa32882
`
`0
`~16 -15-14-13-12-11-10 -9 -8 -7 -6 5 4
`Log [Compound] M
`
`
`TrgECSTHiSope|ECEO [PGD2____——~+it 8.679|-1.405|2.096010
`
`Log [Compound] M
`PC LOgECSO} HillSlope|ECO
`LT -0.6378|1.215¢-006
`Treprostinil -0.9616|2.352¢e-009-8.629
`
`
`Treprostinil
`-9.483
`-1.606
`3.292e-010
`(b) (4)
`(b) (4)
`
`FRET cAMP Standard Curves
`
`
`
`-® HBSS Buffer
`—™ |BMX HBSSBuffer
`
`5000
`
`8=
`
`4000
`& 3000
`
`20004
`: 4000
`
`N2g
`
`CHO-K1 IP1, Agonist Mode with Compounds, cAMP
`5000
`
`v
`
`8
`© 4000
`2
`& 3000
`
`© |loprost
`“= Treprostinil
`si,
`=~ AVG Buffer
`
`
`
`(b) (4)
`
`-15 -14 -13 -12 -11-10 -9
`
`-8 -7 6 -5 -4
`
`
`
`Treprostinil
` IBMX HBSS Buffer|-6.981
`
`Log [Compound] M
`Log [CAMP] M
`[SSCSAISTOET CSO
`rah
`AU
`HiliSiop
`pWoprost_————SS~=*dS 1021.095|6.195801
`-f
`0350
`9740
`5
`3340-008
`HBSS Buffe
`1.0462-007
`0.7368
`
`Reference ID: 4814268
`
`11
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`Figure 4. Calcium assay in agonist mode: Dose-dependentstimulation of Calcium flux in
`human EP1 receptor-expressing cells (from the submission).
`
`HEK293T EP1, Agonist Mode with Compounds, Calcium
`6000
`
`6s
`
`=® lloprost
`-—@ Treprostinil
`me
`—— AVG Buffer
`
`(b)(4)
`
`3
`S
`Se
`4000
`o
`©wo
`® 2000
`
`°=
`
`so a
`
`v
`
`0
`-14 -13 -12 -11 -10 -9
`
`-8
`
`-7
`
`-6
`
`-5
`
`-4
`
`Log [Compound] M
`TogEC50|HillSlope|EC50
`6.868|1.101|1.354e-009
`Treprostinil
`
`7.000e-008(b) (4)
`
`5
`
`Pharmacokinetics
`
`Determination of the in vivo pharmacokinetics of an inhaled Treprostinil dry
`5.1.
`powderand nebulized Treprostinil in Rat (MKC-PC-2018-004)
`
`This study (MKC-PC-2018-004) sponsored by MannKind Corporation, Danbury, CT was
`donein
`®®(The sponsorship underthe
`associated IND-134582 was changed from MannkKind Corporation to United
`Therapeutics Inc. on Oct 18, 2018). This study evaluated the pharmacokinetics (PK) of
`two dry powderformulations of treprostinil and a reference compound nebulized
`treprostinil.
`
`Male Sprague Dawleyrats (body weight 200-250g) were anesthetized with isoflurane,
`then were intubated and Treprostinil Inhalation Powder(TriP) was delivered
`intermittently into the tracheal tube (Table 2). After dosing, blood samples were
`collected at specific time points over a period of 6 hours. Nebulized treprostinil was
`used as a reference compound. The selected dosesreflect a range believed to exceed
`the therapeutic dose in humans.
`
`Reference ID: 4814268
`
`12
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`Table 2. The study design for treprostinil administration (from the submission)
`Target
`Target dose
`amount of
`of
`.
`inhalation
`Treprostinil Treprostinil
`powder per
`per rat*
`rat (Ing)
`(mg/kg)
`
`content
`
`1.02 (wt%)
`
`
`1.02 (wt)
`
`0.39 (wt%)
`
`
`0.39 (wt%)
`
`0.041**
`
`
`
`
`
`*Based on 250 g rat.
`
`.iacae.was0.041 ak Nomoreio25%onkewasa
`
`
`The lower dose of TriP formulation #2 (target 0.006 mg/rat,) could not be detected in
`plasma. Forall detectable test compounds, the maximal plasma concentration was
`detected 5 minutes post-dosing(first time point of blood collection). TriP provided much
`higher treprostinil exposure than the nebulized reference solution (Figure 5, Table 3).It
`wasstated that there was no evidencethat the excipient FDKP, presentin TriP
`formulations, interfered with the absorption of treprostinil into the lung.
`
`Figure 5. Concentrations of Treprostinil in Rat Plasma Samples (from the submission)
`= 35000
`Ea
`2 30000
`©
`3 25000
`&
`3 20000
`8 15000
`
`—@—THP1 (0.041mg/kg)
`—h=TrIP2 (0.019m¢g/kg)
`—H—TriP1 (0.018mg/kg)
`—><= NebulizedTreprostinil (0.041mg/kg)
`
`
`
`10000
`
`5000
`
`0
`
`o E” &
`
`= s
`
`2o
`
`‘a
`
`50
`
`100
`
`200
`150
`Time (min)
`
`250
`
`300
`
`350
`
`Reference ID: 4814268
`
`13
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`L/min
`
`,
`
`035
`
`0.082
`
`Table 3. Non-compartmental analysis of plasma data after insufflation of TriP1/2 or
`nebulization of treprostinil (from the submission)
`Group 1|Group 2 Group 5
`
`Sean:
`0.041
`0.018
`0.019
`0.006
`0.041*
`
`
`
`
`
`rate
`8.38
`;
`F
`min
`Half-life
`Tmax
`min
`ey
`5
`5
`BLOQ
`5
`
`
`
`
`
`Cmax | 303753|7364.6| pg/ml | 114493 | BLOQ 1330.7
`
`
`
`
`
`Tate ditag | pgimL*min 792886.7|317547.3|296854.6 | BLOQ|22542.1
`exposure
`CLIF <444 | mL/minkg | <52 <57 | <64 | BLOQ
`
`
`
`
`
`
`
`*The total nebulized dose was 0.041 mg/kg. No more than 25%(0.010 mg/kg)
`was deposited. CL/F was calculated based on a dose of 0.010 mg/kg.
`
`STABILITY INHUMAN LIVER MICROSOMESSTABILITY IN HUMAN
`5.2
`HEPATOCYTES(20UNITP1S5)
`
`This study (2ZOUNITP1$5)wasto determinethe stability of
`humanliver microsomes and human hepatocytes.
`
`o@in
`
`®® was added into the reaction mixture (containing mixed-gender
`humanliver microsomes) or mixed-gender humancryopreserved hepatocytes
`suspension (1.5 x 106 cells/mL) at a final concentration of 1 WM (in duplicate). The
`mixture wasincubated in a shaking waterbath at 37°C. Aliquots (200 uL) were
`withdrawnat 0, 10, 20, 30, and 60 minutesfor liver microsome reaction and 0, 15, 30,
`60, and 120 minutes for hepatocyte assay, and analyzed using LC-MS/MS.Positive
`controls, testosterone (1 UM) and 7-hydroxycoumarin (7-HC) (100 uM), were
`performedin parallel.
`
`(b) (4)
`
`» in human liver microsomes
`
`(Table 4) and hepatocytes (Table 5), with T1/2 < @ min.
`
`Reference ID: 4814268
`
`14
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
` Table4. Stabilityofss in Human Liver
`
`Microsomes(modifi
`
`om the submission
`
`
`
`a S
`
`Test Article
`
`Percent Remaining (AVG, n=2)
`
`Treprostinil Concentration (uM) (AVG, n=2)
`
`Dased pecies| min|10min|20min|30min|60min|
`
`Human‘Human||
`
`4 Whenthe calculated half-life is < the first non-zero timepoint, the half-life is listed as POith the calculated half-
`life also listed in parentheses.
`> Intrinsic clearance (CLint) was calculated based on CLint = k/P. wherekis the elimination rate constantandPis the
`protein concentration in the incubation.
`
`tability results should be interpreted with caution for these experiments.
`
`Table 5. Stabilityofl? [NINN in Human
`hepatocytes (modified from the submission)
`
`Half-life’
`(min)
`
`Test Article
`
`Treprostinil Concentration (uM) (AVG, n=2)
`
`Srnnasin|intin]i *Whenthecalculatedhalf-aeis<thefirstnon-zerotimepoint,thehalf-lifeislistedasPOwitnthecalculatedhalf-
`
`life also listed im parentheses.
`* Intrinsic clearance (CLint) was calculated based on CLint = k/P, where k is the elimination rate constant andPis the
`cell concentration in the incubation.
`
`10
`
`Special Toxicology Studies
`
`
`
`10.1NTIALBACTERIALMUToon) COMPUTATIONAL EVALUATION OF THE
`
`AGENICITY OFImpurities Potentially Associated
`POTE
`with Treprostinil Inhalation Powder
`
`Thisis an in-silico assessmentofpurities in a” prepared by
`
`The potential bacterial mutagenicity
`of seven impurities
`
`
`and
`
`associated with treprostinil inhalation powder, a dry powder formulation of the
`active pharmaceutical ingredient (API) treprostinil and the excipient fumaryl
`diketopiperazine (FDKP), was evaluated by (quantitative) structure-activity relationship
`
`Reference ID: 4814268
`
`15
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`[(Q)SAR] using the Derek Nexus [Nexusv.2.2.1 (Build 91, January 2018), DEREK
`Nexusv. 6.0.1] and Leadscope Model Applier (v. 2.4.2-1) systems.
`
`Chemical structures were entered as .molfiles into ChemIDplus
`(https://chem.nim.nih.gov/chemidplus/ ) to ascertain if a Chemical Abstracts Service
`Registry Number (CAS RN) wasassigned. If available, this identifier would be used to
`searchfor any available results of experimental bacterial mutagenicity assessment. In
`the case of the seven impuritiesA
`"and
`© identified by
`United Therapeutics, no matching CAS RNwasidentified for any of the substances.
`
`The chemical structures of each of the impurities as well as that of Treprostinil were
`submitted to DEREK Nexus and Leadscope Model Applier for evaluation of potential
`bacterial mutagenicity. All seven impurities (
`’ and
`om were
`identified as inactive (i.e., non-mutagenic) in DEREK Nexus and negative (non-
`mutagenic) in the Leadscope Model Applier. DEREK Nexus and Leadscopedid not
`identify any of the structural features of these substances as unclassified / misclassified
`or out of domain, respectively. Furthermore, evaluation of the structure of treprostinil
`suggested that this substance was not mutagenic, andthis result was confirmed by
`Amesassaytest data. Each of the treprostinil impurities assessedin this report bears
`structural resemblanceto treprostinil and, thus, the Ames assaydata for treprostinil help
`to substantiate the in-silico mutagenicity predictions for the seven impurities.
`
`Basedonthein-silico assessmentssand expert judgement! and per the ICH M7
`guideline?,
`and
`are considered as Class 5 impurities and
`may be treated as non-mutagenic. Treprostinil, an API, was considered non-mutagenic
`based ontest results in a GLP 5-test strain Ames assay’.
`
`(b) (4)
`
`(b) (4)
`
`The chemicalstructures usedfor the predictions, as well as that of the API, and results
`of in silico assessment are shownin Table 6.
`
`1 Powley, M.W.(2015). (Q)SAR Assessments of Potentially Mutagenic Impurities: a
`Regulatory Perspective on the Utility of Expert Knowledge and Data Submission. Regul
`Toxicol Pharm, 71: 295-300.
`2 ICH — International Conference on Harmonization of Technical Requirements for
`Registration of Pharmaceuticals for Human Use. (2017). ICH Harmonized Tripartite
`Guideline: Assessment and Control of DNA Reactive (Mutagenic) Impurities in
`Pharmaceuticals to Limit Potential Carcinogenic Risk M7 (R1). Step 4 — 31 March 2017.
`3 [UT] United Therapeutics Corporation. (2013). Pharmacology Review(s). NDA 203496.
`Approval Date 20 December 2013. Submitted to US FDA Center for Drug Evaluation
`and Research (CDER). [online]. Available at:
`https:/Awww.accessdata.fda.gov/drugsatfda_docs/nda/2013/2034960rig1sO00TOC.cfm .
`
`Reference ID: 4814268
`
`16
`
`

`

`Treprostinil
`(CAS RN 81846-19-7)
`
`Salmonella
`and
`in E. CoA
`
`Salmonella
`and
`
`mutagenicity
`based on
`
`experimental
`data (UT, 2013)
`
`NDA 214324
`
`Reviewer: Baichun Yang
`
`Table 6. Summary ofin silico assessment(from the submission)
`
`in E. CoA Negative for
`
`Reference ID: 4814268
`
`17
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`10.2
`
`=
`®®(2021) Toxicological Profile and Risk Assessmentfor
`in Inhaled Treprostinil Drug Product (Tyvaso DPI Inhalation Powder)
`
`This assessmentfor treprostinil impurity
`
`(b) (4)
`
`wasprepared by
`
`(b) (4)
`
`®©is identified as an impurity in Tyvaso DPI at a concentration
`abovethe qualification threshold of 1% for drug products with a maximum daily dose of
`drug substancethat is <10 mg (For Tyvaso DPI Inhalation Powder, the
`maximum daily dose being administered is 16 yg 4 times daily for a total daily dose of
`64 ug treprostinil and may increase by 16 ug/time every 1-2 weeks). A hazard
`characterization was conductedfor
`®©®to qualify that impurity to a
`concentration of not more than 2%.
`O®on
`In cell-based assays comparedto treprostinil, the activity of
`the prostacyclin receptor (IP), prostaglandin E2 (PGE2) receptor 2 (EP2), prostaglandin
`D2 (PGD2) receptor 1 (DP1), and PGE2 receptor 1 (EP1) receptors was approximately
`23-, 10-, 14-, and 25-fold lower, respectively.
`In microsomes,
`waspresent at 0 minutes, suggesting that
`oe
`during the 5-minute pre-incubation period.
`
`little to no
`
`wie
`
`®®is not predicted to be mutagenic. Noothernonclinical toxicology
`studies have been conducted with
`®©®either neat or presentin
`treprostinil drug products.
`
`(b) (4)
`
`(b) (4)
`
`Impurity
`“™ is
`expected to have reduced pharmacological activitycompared to treprostinil and 10be© If
`any
`*** remains intact following absorption in the lung, it is not
`expected to interfere with the normal CYP-mediated metabolism of treprostinil and is
`unlikely to cause unforthcomingtoxicity.
`
`™® »oresent
`Following patient use of Tyvaso DPI, the small quantity of
`is unlikely to
`in the drug productortreprostinil released from
`substantially impact the pharmacologicaleffects of treprostinil. Also, because the daily
`dose of inhaled treprostinil is individually titrated to tolerability and effect for each
`patient,it is unlikely that the presence of
`®in Tyvaso DPI would be
`a safety concernfor patients.
`
`(b) (4)
`
`Using structurally similar surrogate treprostinil, a Permitted Daily Exposure (PDE)of
`” was estimated. Briefly, adjusting|{}-fold for inter-individual
`variability andfold to extrapolate from a therapeutically efficacious dose to a No-
`
`Reference ID: 4814268
`
`18
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`Observed-Effect-Level (NOEL), which is consistent with ICH guidelines and current
`industry standards for calculating a PDE, there are no additional safety concernsif
`”is present at approximately 3% of the dose of Tyvaso DPI.
`Dose (mg/day)
`(b)(4)
`1 Ma =
`
`x Dose (mg/day)
`
`11
`
`Integrated Summary and Safety Evaluation
`
`Brief Background / Introduction
`
`The API in Tyvaso DPI is identical to the treprostinil drug substance approvedin
`Remodulin (NDA 021272) and Tyvaso (NDA 022387). It is also the same active moiety
`as the treprostinil diolamine drug substance approvedin Orenitram (treprostinil)
`Extended-Release Tablets (NDA 203496). The safety and efficacy of treprostinil are
`supported by a comprehensive set of pharmacology, pharmacokinetic (PK), toxicology,
`and clinical studies conducted for Tyvaso(treprostinil) Inhalation Solution (NDA
`022387), Remodulin (treprostinil) Injection (NDA 021272), and Orenitram (treprostinil)
`Extended-Release Tablets (NDA 203496).
`
`The safety of the excipient FDKP is supported by a comprehensive battery of safety
`pharmacology and toxicology studies conducted for Afrezza (BLA 022472).
`
`No additional pharmacology studies norin vivo nonclinical toxicology studies were
`conducted to support the NDAfor Tyvaso DPI.
`
`Pharmacology
`
`No additional nonclinical pharmacology studies were conducted to support the NDAfor
`Tyvaso DPI. Following intermittently delivery of treprostinil into the rat tracheal tube,
`treprostinil Tmax was <5 minutes. Treprostinil Inhalation Powder(TriP) provided much
`higher treprostinil exposure than the nebulized reference solution. There was no
`evidencethat the excipient FDKP, present in Tyvaso DPI, interfered with the absorption
`of treprostinil in the lung.
`
`The stimulating activity of the impurity
`®™@ at the IP1, EP2, DP1, and
`EP1 receptors in tested cell lines were approximately 23-, 10-, 14-, and 25-foldlower,
`respectively, than treprostinil.
`® was
`in humanliver microsomes and hepatocytes, with T1/2 <°@min.In
`microsomes,little to no
`™®Was present at 0 minutes, suggestingthat
`© In hepatocytes,
`at time zero was also observed ( ~"%).
`
`wre
`
`(b) (4)
`
`Toxicology
`
`Reference ID: 4814268
`
`19
`
`

`

`NDA 214324
`
`Reviewer: Baichun Yang
`
`No additional in vivo nonclinical toxicology studies were conducted to support the NDA
`for Tyvaso DPI. The potential bacterial mutagenicity of 7 impurities identified in Tyvaso
`DPI was evaluated by (quantitative) structure-activity relationship ((Q)SAR) usingDerek
`Nexus and Leadscope Model Applier. All 7 impurities (® snd
`(b)(4)
`wereidentified as inactive (