OptiNose US, Inc.

, 1020 Stony Hill Road, Suite 300, Yardley, PA 19067

May 31, 2021

Division of Dockets Management (HFA-305)

Food and Drug Administration
Department of Health and Human Services
5630 Fishers Lane, Room 1061
Rockville, Maryland 20852

CITIZEN PETITION

OptiNose US, Inc. (individually or together with its affiliates, “OptiNose”) submits this
petition under 21 C.F.R. § 10.30 and section 505 of the Federal Food, Drug, and Cosmetic Act
(FDCA) and its implementing regulations to request that the Commissioner of Food and Drugs
take the actions set forth below in Section A with respect to abbreviated new drug applications
(ANDAs) and section 505(b)(2) applications that reference or rely upon OptiNose’s XHANCE®
(fluticasone propionate) nasal spray.

A. Actions Requested

OptiNose requests that the Food and Drug Administration (“FDA” or the “Agency”) take
the following actions:

 FDA should not approve an ANDA referencing XHANCE unless the applicant
demonstrates bioequivalence and therapeutic equivalence through a non-inferiority
clinical endpoint study of the type described herein in addition to in vitro studies,
pharmacokinetic study data, qualitative and quantitative sameness of excipients, and
an adequate showing of device equivalence;

 FDA should issue a draft product-specific guidance with recommendations that are
consistent with these requests; and

 FDA should not assign an “A” rating for a section 505(b)(2) product unless the
applicant demonstrates bioequivalence and therapeutic equivalence through a non-
inferiority clinical endpoint study.

B. Statement of Grounds

I. Executive Summary

OptiNose US, Inc. holds NDA 209022 for XHANCE™ (fluticasone propionate) nasal
spray (“XHANCE”) for the treatment of nasal polyps in patients 18 years of age or older.
XHANCE is a complex drug-device combination product that delivers fluticasone propionate, a
locally acting corticosteroid, through a unique breath-powered™ exhalation delivery system
device (“EDS”).

The XHANCE EDS™ differs from the device constituent in other approved nasal spray
products. The XHANCE EDS is specifically designed to deposit the fluticasone propionate

Main +1 267 364 3500 Fax +1 267 395 2119 www.optinose.com

suspension in the most superior and posterior regions of the nose, particularly regions where
nasal polyps occur. The design of the XHANCE EDS interacts with the patient’s anatomy and
physiology, which include elements that are both static and dynamic (in that they change during
inhalation and exhalation, as well as in response to external conditions and endogenous
stimuli), to produce a different distribution of drug in the nasal cavity than a conventional nasal
spray.1 In this way, both the device and the interaction of the device with the patient’s anatomy
and physiology to achieve drug distribution, are central to the safety and efficacy of XHANCE.
Further, the device is designed to create device-mediated potential therapeutic effects
complementary to, or independent of, the drug-mediated effect, including those related to
carbon dioxide, nitric oxide, change in pH, or positive pressure.

The XHANCE EDS introduces multiple new variables into the intranasal delivery process
that influence drug deposition and clinical effect. Existing in vitro methods, and other
alternatives to a comparative clinical endpoint study, are insufficient to fully assess whether a
different exhalation delivery system used in a generic product would function, and would
interact with the patient’s anatomy and physiology, in a way that would provide equivalent drug
deposition at the site of nasal polyps and, consequently, would be expected to have the same
safety and efficacy as XHANCE.

For these reasons, OptiNose submits this citizen petition to request that FDA take the
following actions:

First, FDA should require an ANDA referencing XHANCE to include data from the type
of non-inferiority clinical endpoint study described herein, in addition to in vitro studies,
pharmacokinetic study data, qualitative and quantitative sameness of excipients, and an
adequate showing of device equivalence. A non-inferiority clinical endpoint study, in particular,
is an essential element of the showing of bioequivalence and therapeutic equivalence. There are
no alternative methods available to either accurately measure or predict equivalent in vivo drug
deposition or confirm the absence of safety and efficacy differences that may result from drug
deposition that differs from the deposition that results from the design of the XHANCE EDS and
its interaction with the patient’s static and dynamic anatomy and physiology during drug
delivery.

Second, FDA should issue a draft product-specific guidance with recommendations that
are consistent with these requests.

Third, FDA should require a section 505(b)(2) applicant to demonstrate bioequivalence

and therapeutic equivalence through a non-inferiority clinical endpoint study as described
herein in order to obtain an “A” rating in Approved Products with Therapeutic Equivalence
Evaluations, more commonly known as the Orange Book.

1This static and dynamic variation in anatomy and physiology is described below in section B.III.A.1.a.2–
3, and the interaction with the XHANCE EDS is described below in section B.III.A.1.d.

2
II. Background

A. Factual Background

FDA approved the XHANCE NDA on September 18, 2017.2 XHANCE uses a novel device
for intranasal administration of a suspension formulation containing fluticasone propionate.
The device constituent of XHANCE is a breath-powered exhalation delivery system (“EDS”) and
differs from devices that are part of any other approved nasal spray product. As illustrated in
Figures 1A and 1B below, exhalation delivery systems like the XHANCE EDS produce deposition
profiles that are dramatically different than those associated with conventional nasal sprays.

Figure 1A - Gamma Image of Liquid Deposition in Nasal Cavity of Healthy Subject

Superimposed on Corresponding Sagittal Magnetic Resonance Imaging Section3

Figure 1B – Liquid Deposition in Silicone Cast with Simulated Nasal Polyp4

A = Conventional nasal spray B= Exhalation delivery system

2FDA, NDA 209022 Approval Letter (Sept. 18, 2017),

https://www.accessdata.fda.gov/drugsatfda_docs/appletter/2017/209022Orig1s000ltr.pdf. (attached
hereto as Exhibit 1).
3P.G. Djupesland. Nasal Drug Delivery Devices: Characteristics and Performance in a Clinical Perspective
– A Review. Drug Deliv. Transl. Res. 2013 Feb.; 3(1):42–62 (attached hereto as Exhibit 2).

3
 The XHANCE EDS―as illustrated below in Figure 2 ―consists of a vial containing a
suspension of fluticasone propionate, a metering spray pump, a complex shaped nosepiece that
directly interacts with the user’s nostril and nasal cavity, and a movable mouthpiece that directly
interacts with the mouth and oral cavity.5 The device also includes a valving system to reduce
patient coordination burden and to create an “air burst” effect, a grip designed to enable both
one- and two-handed use and both ipsilateral and contralateral administration, and a specific
angle and distance between the mouthpiece and nosepiece.

Figure 2

To administer XHANCE, the user inserts the sealing nosepiece into one nostril in a
manner that creates an airtight seal with the flexible external nasal tissues (nares and nasal
vestibule) and then inserts the mouthpiece between the lips while maintaining the nasal seal.
When fully inserted, the multi-dimensionally asymmetric nosepiece seals with the external nasal
tissue to allow pressure transfer from the mouth while also acting as a mechanical stent to open
the narrow nasal valve, and particularly the superior segments of the nasal valve. After taking a
deep breath, the user blows into the mouthpiece and static positive pressure is created by the
closed internal mechanical valve in order to preserve lung vital capacity, reduce the need for
coordination of actuation timing, and build up a “burst” potential for airflow.

Blowing through the mouth against resistance―both initial static resistance and post-
actuation dynamic resistance―causes the soft palate in the back of the throat to elevate and
remain sealed closed against the nasopharyngeal wall/roof, isolating the oral cavity from the
nasal cavities with an airtight seal. When the device is actuated by pressing the vial, the orally
generated air pressure is released in a “burst” of airflow timed to start just prior to the release of
drug spray aerosol from the device applicator. Figure 3 below illustrates this process.

4P. G. Djupesland et al. New Exhalation Delivery Systems (EDS) Enhance Topical Steroid Delivery in
Chronic Rhinosinusitis With Nasal Polyps (CRSwNP), presented at Am. Academy of Allergy, Asthma &
Immunology Annual Meeting (Mar. 2017) (attached hereto as Exhibit 3).
5See FDA, NDA 209022 Summary Review Memo, at 2-3 (Sept. 18, 2017),
https://www.accessdata.fda.gov/drugsatfda_docs/nda/2017/209022Orig1s000SumR_rev.pdf (attached
hereto as Exhibit 4).

4
 Figure 3: Closed-Palate Positive Pressure Bi-directional Delivery

Schematic A: View from Left Side Schematic B: View from Top

The XHANCE EDS allows air pressure and volume to be channeled from the mouthpiece
to the sealing nosepiece, while controlling pressure or volume loss through the device before and
during actuation. The release of air and pressure occurs in a synchronized manner to allow for
the breath to entrain the droplets. Under these conditions, considering both direct device
activity and dynamic intra-facial physiology created by use of the device, drug aerosol is
delivered to the superior and posterior regions of the deep nasal cavity not only beyond the nasal
valve, but above and behind bony structures separating the target region of interest for nasal
polyps from the main respiratory channel along the floor of the nasal cavity.

XHANCE EDS delivery results in a majority of the initially deposited metered spray
being deposited beyond the head of the inferior turbinate and a substantial portion of the
initially deposited dose being deposited in the upper posterior region of the nasal cavity beyond
the head of the inferior turbinate and above the inferior meatus, including in the middle meatus,
where the osteomeatal complex is located6. This drug deposition profile is markedly different
than a conventional nasal spray and disproportionately places topically acting steroid in the
region of the nasal cavity where nasal polyps typically originate, and importantly where
paranasal sinuses ventilate and drain, as discussed more fully below in sections B.III.A.1.c-d.

Further, as discussed more fully below in section B.III.A.1.d., the XHANCE EDS is
designed to create device-mediated potential therapeutic effects complementary to, or

6 Djupesland 2017, supra note 4, Exhibit 3; J. V. Bosso et al. Efficient Topical Steroid Delivery to Grades 1

and 2 Nasal Polyps in the Ostiomeatal Complex: Deposition Patterns With Exhalation Delivery System
and Conventional Intranasal Spray Visualized With Fluorescein in Silicone Cast. (2020), presented at Am.
College of Allergy, Asthma & Immunology Annual Meeting (2020) (attached hereto as Exhibit 5).

5
independent of, the drug-mediated effect, including those related to carbon dioxide, nitric oxide,
change in pH, or positive pressure.7

B. Legal and Regulatory Background

1. Abbreviated New Drug Applications

The Drug Price Competition and Patent Term Restoration Act of 1984 (the Hatch-
Waxman Amendments) amended the FDCA to, among other things, add section 505(j), which
established an abbreviated approval pathway for generic drugs. Under section 505(j) of the
FDCA, in order for an ANDA to be approved, the applicant must show, among other things, that
the proposed generic product is the “same as” the reference listed drug (“RLD”) in certain
respects and is “bioequivalent to” the RLD.8 Among other requirements, the ANDA applicant
must provide sufficient information to show that the proposed generic drug product contains
condition(s) of use that previously were approved for the RLD as well as (in general) the same
active ingredient(s), route of administration, dosage form, strength, and labeling as the RLD.9

7 XHANCE Prescribing Information § 12.1 (Apr. 2021) (“[S]tudies suggest that carbon dioxide, which is
present in the exhaled breath delivered into the nose through the device, may influence inflammatory
mediator activity and neuropeptide activity, possibly through mechanisms of action that also include
removal of nitric oxide, change in pH, or positive pressure. The direct relationship of these findings to
long-term symptom relief is not known.”) (attached hereto as Exhibit 6); P. G. Djupesland et al. Novel
Nasal Exhalation Delivery Systems (EDS) May Produce Beneficial Activity Independent of Delivered Drug
in Inflammatory Nasal Diseases and Migraine via Exhaled CO2 and Mucosal pH Changes, presented at
62nd Am. Rhinologic Soc’y Annual Meeting (Sept. 2016) (attached hereto as Exhibit 7); World Intellectual
Property Organization, International Publication No. WO 2014/155192 (attached hereto as Exhibit 8).
The following additional references support the potential device-mediated effect although these studies
are not related to XHANCE directly:
T.B. Casale et al. Intranasal Noninhaled Carbon Dioxide for the Symptomatic Treatment of Seasonal
Allergic Rhinitis. J Allergy Clin. Immunol. 2008; (12)1:105–109 (attached hereto as Exhibit 9).
C. Vause et al. Effect of Carbon Dioxide on Calcitonin Gene Related Peptide Secretion from Trigeminal
Neurons. Headache. 2007; 47:1385–1397 (attached hereto as Exhibit 10).
D. Shusterman, P.C.Avila. Real Time Monitoring of Nasal Mucosal pH During Carbon Dioxide
Stimulation Implications for Stimulus Dynamics. Chem. Sense. 2003; 28:595–601 (attached hereto as
Exhibit 11).
A.Z. Tzabazis et al. Trigeminal Antihyperalgesic Effect of Intranasal Carbon Dioxide. Life Sci. 2010 July 3;
87(1-2):36–41 (attached hereto as Exhibit 12).
M. Surel et al. Comparison of the Use of the Valsalva Maneuver and the Eutectic Mixture of Local
Anesthetics (EMLA®) to Relieve Venipuncture Pain: a Randomized Controlled Trial. J. Anesth. (2012)
(attached hereto as Exhibit 13).
8 FDCA §§ 505(j)(2)(A)(ii) & (iv); 21 C.F.R. §§ 314.94(a)(7), 314.127(a)(6)(i).
9 FDCA § 505(j)(2)(A).

6
 2. Section 505(b)(2) Applications

A section 505(b)(2) application relies on “investigations . . . [that] were not conducted by

or for the applicant and for which the applicant has not obtained a right of reference or use from
the person by or for whom the investigations were conducted.”10 Like a full NDA, a section
505(b)(2) application must demonstrate that the proposed product is safe and effective. But
FDA has interpreted section 505(b)(2) to permit an applicant to rely upon the following two
sources of information to support this showing: (1) published scientific literature; and/or
(2) FDA’s finding of safety and effectiveness for a previously-approved drug product.11 FDA
permits an applicant to rely upon a prior finding of safety and effectiveness for a listed drug
“only to the extent the product seeking approval and the listed drug are the same” and only “to
the extent such reliance is scientifically justified.”12

The applicant must submit data to support the “safety and effectiveness of any
differences between the listed drug and the drug proposed in the section 505(b)(2)
application.”13 These data are known as “bridging” data.14 The type and quantum of required
bridging data “will vary from case to case” but may include clinical or animal data, “as
appropriate.”15 In draft guidance, FDA recommends that a section 505(b)(2) application should
include “[a] Bioavailability/Bioequivalence (BA/BE) study comparing the proposed product to
the listed drug.”16

3. Bioequivalence Recommendations for Locally Acting Nasal Spray Drug

Products

As noted, an ANDA applicant must show that its proposed generic drug product is
bioequivalent to the RLD,17 meaning that “the rate and extent of absorption of the drug do not
show a significant difference from the rate and extent of absorption of the [RLD] when

10 Id. § 505(b)(2).
11FDA, Draft Guidance for Industry, Applications Covered by Section 505(b)(2), at 2–3 (Oct. 1999)
(“505(b)(2) Draft Guidance”), https://www.fda.gov/media/72419/download (attached hereto as Exhibit
14).
12Letter from Janet Woodcock, M.D., FDA, to Kathleen M. Sanzo, Morgan, Lewis & Bockius, LLP, et al.,
re: Docket Nos. 2001P-0323/CP1 & C5, 2002P-0447/CP1, & 2003P-0408/CP1, at 2, 3, 12 (Oct. 14, 2003)
(“505(b)(2) Citizen Petition Response”); see also id. at 4 (“[S]uch reliance will be appropriate only to the
extent that the proposed product in the 505(b)(2) application shares characteristics . . . in common with
the listed drug.”) (attached hereto as Exhibit 15).
13 Id. at 4; see also 21 C.F.R. § 314.54(a).
14 505(b)(2) Draft Guidance, supra note 11, at 8 (Exhibit 14).
15 505(b)(2) Citizen Petition Response, supra note 12, at 4 (Exhibit 15).
16 505(b)(2) Draft Guidance, supra note 11, at 8 (Exhibit 13). Bioequivalence generally means that “the

rate and extent of absorption of the drug do not show a significant difference from the rate and extent of
absorption of the listed drug when administered at the same molar dose of the therapeutic ingredient
under similar experimental conditions in either a single dose or multiple doses.” FDCA § 505(j)(8)(B)(i);
see also 21 C.F.R. § 320.23(b).
17 FDCA § 505(j)(2)(A)(iv).

7
administered at the same molar dose of the therapeutic ingredient under similar experimental
conditions in either a single dose or multiple doses.”18 FDA’s regulations provide that
bioequivalence is “the absence of a significant difference in the rate and extent to which the
active ingredient or active moiety . . . becomes available at the site of drug action when
administered at the same molar dose under similar conditions in an appropriately designed
study.”19 FDA has explained that “[f]or systemically acting drug products, the rate and extent of
systemic absorption of the drug is usually the most sensitive, accurate and reliable indicator of
the rate and extent to which the active ingredient becomes available at the site of drug action.”20
For such systemically acting drug products, a comparison of drug and/or metabolite
concentrations in an accessible biological fluid, such as blood, generally forms the basis for a
determination of bioequivalence.

Drug products that act locally, however, do not reach their site of action through
absorption into the bloodstream. For locally acting drug products, the statute permits FDA to
“establish alternative, scientifically valid methods to show bioequivalence if the alternative
methods are expected to detect a significant difference between the drug and the [RLD] in safety
and therapeutic effect.”21 The statute also provides that FDA may “assess bioavailability by
scientifically valid measurements intended to reflect the rate and extent to which the active
ingredient or therapeutic ingredient becomes available at the site of drug action.”22

In April 2003, FDA published a draft guidance on appropriate methods to measure

bioavailability and/or establish bioequivalence in support of NDAs or ANDAs for locally acting
drugs in nasal aerosols (metered-dose inhalers (MDIs)) and nasal sprays (metered-dose spray
pumps).23 This draft guidance acknowledges that “BA and BE assessments for locally acting
nasal aerosols and sprays are complicated because delivery to the sites of action does not occur
primarily after systemic absorption.”24 The droplets and/or drug particles deposited topically
become available at local sites of action when absorbed locally, and thus the location of drug
deposition is an important component influencing drug activity.25

In this draft guidance, FDA further acknowledges that medication activity at local sites
within the nasal cavity is a function of multiple factors, including droplet and drug particle sizes
and size distributions, drug dissolution, absorption across mucosal barriers to nasal receptors,

18 Id. § 505(j)(8)(B)(i).
19 21 C.F.R. § 314.3(b).
20Letter from Janet Woodcock, M.D., FDA, to Damon Burrows, Allergan, Inc., Docket No. FDA-2014-P-
0304, at 8 (Nov. 20, 2014) (attached hereto as Exhibit 14).
21 FDCA § 505(j)(8)(C).
22 Id. § 505(j)(8)(A)(ii); 21 C.F.R. § 320.23.
23FDA, Draft Guidance for Industry, Bioavailability and Bioequivalence Studies for Nasal Aerosols and
Nasal Sprays for Local Action (April 2003)(“Nasal Spray Draft Guidance”),
https://www.fda.gov/media/70867/download (attached hereto as Exhibit 17).
24 Id. at 3.
25 Id.

8
and rate of removal from the nose. 26 For these reasons, FDA recommended that the
bioequivalence of locally acting nasal drug products be demonstrated through a battery of
evidence, commonly described as a “weight-of-evidence” approach, including comparative
evaluation of the formulation and device, equivalent in vitro performance (demonstrated
through comparative in vitro studies), equivalent systemic exposure (demonstrated through
comparative pharmacokinetic studies), as well as equivalent local delivery (demonstrated
through comparative clinical endpoint studies).27

4. FDA’s Draft Product Specific Guidance on Fluticasone Propionate

FDA has issued a general draft product specific guidance (“PSG”) document for
fluticasone propionate (metered spray, 0.05 mg/spray) products referencing FLONASE, as well
as a separate draft product-specific guidance document providing recommendations that are
specific to proposed OTC products.28 FLONASE has a conventional nasal spray delivery device
constituent, unlike the XHANCE EDS.

Consistent with the Nasal Spray Draft Guidance, FDA’s general draft PSG for products
referencing FLONASE recommends that bioequivalence be demonstrated through a
combination of in vitro studies, a pharmacokinetic study, and a comparative clinical endpoint
study.29 Additionally, FDA recommends that (1) the test formulation be qualitatively and
quantitatively the same as the reference formulation, (2) prospective applicants refer to FDA’s
guidance on the identification and assessment of any differences in the design of the user
interface for a proposed generic drug-device combination product, and (3) prospective
applicants consider the external operating principles and external critical design attributes of
the reference product, the size and shape of the reference product, and the number of doses in
the reference product.30 The draft PSG provides for comparative particle size distribution
(“PSD”) data (note that in this context, PSD refers to the size of particles in the liquid
suspension formulation contained in the vial and not to the size of droplets delivered into the
nose or their distribution in the nose under conditions of actual use31) as an alternative to the
comparative clinical endpoint bioequivalence study “[i]f drug PSD in the [test] and [reference]

26 Id.
Id. See also Q. Liu et al. Scientific Considerations for the Review and Approval of First Generic
27

Mometasone Furoate Nasal Suspension Spray in the United States from the Bioequivalence Perspective.
AAPS J. 2019 Jan. 7; 21(2):14 (attached hereto as Exhibit 18).
28FDA, Draft Guidance on Fluticasone Propionate Spray, Metered; Nasal (revised June 2020)
(“FLONASE Draft Guidance”), https://www.accessdata.fda.gov/drugsatfda_docs/psg/PSG_020121.pdf
(attached hereto as Exhibit 19); FDA, Draft Guidance on Fluticasone Propionate Metered, Spray; Nasal
(OTC; May 2021) (“OTC FLONASE Draft Guidance”),
https://www.accessdata.fda.gov/drugsatfda_docs/psg/PSG_205434.pdf (attached hereto as Exhibit 20).
29 FLONASE Draft Guidance, supra note 28, at 1–8 (Exhibit 19).
30 Id. at 8–9.
31 See id. at 8; Liu 2019, supra note 27 (Exhibit 18).

9
products can be accurately measured using a validated analytical method such as morphology-
directed Raman spectroscopy or any other advanced methodology.”32

5. Therapeutic Equivalence Determinations

FDA considers a proposed product to be therapeutically equivalent to the RLD or, in the
case of a section 505(b)(2) application, the listed drug relied upon, only if the proposed product
“can be expected to have the same clinical effect and safety profile [as the RLD or listed drug
relied upon] when administered to patients under the conditions specified in the labeling.”33 To
be identified as therapeutic equivalents, drug products additionally must be pharmaceutical
equivalents (i.e., they “are drug products in identical dosage forms and route(s) of
administration that contain identical amounts of the identical active drug ingredient . . . and
meet the identical compendial or other applicable standard of identity, strength, quality, and
purity”) and must be bioequivalent.34 Drug products that are therapeutic equivalents are
identified in FDA’s Orange Book with an “A” rating.35

III. Discussion

A. A non-inferiority clinical endpoint study and additional data are necessary to

demonstrate bioequivalence and therapeutic equivalence to XHANCE.

XHANCE’s breath-powered exhalation delivery system (“EDS”) is unlike the device

constituent in any other approved nasal spray product. The unique drug deposition achieved by
the EDS, including through the interaction of the device with the patient’s static and dynamic
anatomy and physiology, is central to the safety and efficacy of XHANCE. For this reason, a
“weight of the evidence” approach that includes the type of non-inferiority clinical endpoint
study described below, among other types of data, is necessary to ensure that a proposed generic
product using a different device constituent is bioequivalent and therapeutically equivalent to
XHANCE.

1. The XHANCE EDS is central to the product’s safety and effectiveness.

a) The XHANCE EDS accommodates the anatomy, complexity and

dynamics, and inter-patient variability of the nasal cavity and mid-
facial structures, including the mouth, involved in use of the
product.

The XHANCE EDS was designed to deliver fluticasone propionate to the most superior
and posterior regions of the nasal cavity―where nasal polyps occur―taking into account the
anatomy, complexity and dynamics, and inter-patient variability of the nasal cavity and other
relevant structures.

32 FLONASE Draft Guidance, supra note 28, at 8 (Exhibit 19).

33 21 C.F.R. § 314.3(b).
34 Id.
35FDA, Approved Drug Products with Therapeutic Equivalence Evaluations (the “Orange Book”) (41st
ed. 2021), at xiii (attached hereto as Exhibit 21).

10
 (1) Anatomy of the nasal cavity

The nose is divided by a midline septum from the anterior towards the posterior into two
convoluted nasal passages that are joined into a common pathway behind the most posterior
part of the septum in the nasopharynx. Each cavity starts with the vestibule just inside the
nostrils where it is lined with non-ciliated squamous epithelium. The vestibule is supported by
cartilage and transitions into a narrow anterior triangular dynamic segment of the nasal
anatomy called the nasal valve. The nasal valve region includes the narrowest opening segment
of the nose, and of the entire respiratory tract, and therefore is a primary flow-limiting segment
of the nasal cavity.36 It is not a vertically oriented structure, but “tipped forward” and includes a
region that extends posteriorly to approximately the head of the inferior turbinate, about 2–3
cm from the nostril opening.37 This slender, roughly triangular-shaped slit is much narrower at
its upper part and acts as a dynamic valve to modify the rate and direction of airflow during
respiration. Figure 4 below illustrates the nasal cavity, viewing the nasal valve from the nasal
vestibule.

Figure 438

The mucosal lining of the nasal cavity is not uniform in type or function. The squamous
mucosal lining of the nasal valve region gradually transitions from columnar epithelium to
ciliated respiratory epithelium.39 The region lined with respiratory epithelium usually contains
three nasal turbinates―the superior, the middle and the inferior―which project from the lateral
wall of each half of the nasal cavity, and other complex structures such as the uncinate process,
although there can be substantial person-to-person anatomic variation, including, for example,

36L. Illum . Nasal Drug Delivery—Possibilities, Problems and Solutions. J Control Release. 2003 Feb. 21;
87(1–3):187–198 (attached hereto as Exhibit 22).
37P. Cole . The Four Components of the Nasal Valve. Am. J. Rhinol. 2003 Mar,–Apr.; 17(2):107–110
(attached hereto as Exhibit 23).
38 Reprinted with permission from MediVisual (www.medivisual.com).
39A. Sahin-Yilmaz and R.M. Naclerio. Anatomy and Physiology of the Upper Airway. Proc. Am. Thorac.
Soc. 2011 Mar.; 8(1):31–39 (attached hereto as Exhibit 24).

11
a fourth or “supreme” turbinate. The middle meatus, above the inferior turbinate and
behind/lateral to the middle turbinate, is the area of the nose where the maxillary, frontal and
anterior ethmoid sinuses drain and ventilate. In addition, this region is also where nasal polyps
typically originate, as discussed below.40 The ciliated respiratory mucosa is covered by a
protective mucus blanket designed to trap particles and microorganisms41 which is cleared via
the beating action of cilia towards the nasopharynx.42 The density and functional orientation of
ciliated epithelium cells in different regions of the nasal cavity varies greatly, influencing the
clearance rate and direction and the pathway of mucus clearance in different sub regions.43
Figure 5 below illustrates coronal and sagittal sections through the nasal cavity.

Figure 544

The distribution, density, and functional orientation of ciliated cells and non-ciliated
cells, including goblet cells, in the nasal cavity and sinuses affect the mucus layer and
mucociliary clearance rate, as well as the direction and patterns for particles/drugs deposited to
different regions. Additionally, the chronic inflammatory disease process has a non-uniform
influence on mucociliary activity (by anatomic location and by stage of disease).45 Further, the
distribution of goblet cells producing the mucus layer also has an important impact on the rate

40P.L. Larsen, M. Tos. Origin of Nasal Polyps: An Endoscopic Autopsy Study. Laryngoscope. 2004 Apr.;
114(4):710–719 (attached hereto as Exhibit 25).
41D.F. Proctor. The Mucociliary System. In: Proctor DF, Andersen IB, eds. The Nose: Upper Airway
Physiology and the Atmospheric Environment. Amsterdam: Elsevier Biomedical. 1982. p. 245–278
(attached hereto as Exhibit 26); A.R. Halama, S. Decreton, J.M. Bijloos et al. Density of Epithelial cells in
the Normal Human Nose. A Scanning Electron Microscopic Study. Rhinology. 1990; 28(1):25–32
(attached hereto as Exhibit 27).
42 Sahin-Yilmaz, supra note 39 (Exhibit 24).
43Halama 1990, supra note 41 (Exhibit 27); N. Mygind and R. Dahl. Anatomy, Physiology and Function
of the Nasal Cavities in Health and Disease. Adv. Drug Deliv. Rev. 1998 Jan. 5; 29(1–2):3–12 (attached
hereto as Exhibit 28); N. Jones. The Nose and Paranasal Sinuses Physiology and Anatomy. Adv. Drug
Deliv. Rev. 2001 Sep. 23; 51(1–3): 5–19 (attached hereto as Exhibit 29).
44A.J. Sim , A.I. Levine, S. Govindaraj. Functional Nasal and Sinus Surgery. In: A.Levine, S. Govindaraj,
Jr. S. DeMaria (eds). Anesthesiology and Otolaryngology. Springer, New York, NY.
https://doi.org/10.1007/978-1-4614-4184-7_13; Chapter 13 (2013) (attached hereto as Exhibit 30).
45 Mygind 1998, supra note 43 (Exhibit 28).

12
of dissolution, concentration, and clearance of drugs deposited, and also is greatly influenced by
both anatomic location and the nasal polyp disease process. Available human data on these
properties is very limited and largely includes information published in a 1990 study by
Halama.46 Authors attempting in vitro modeling of drug delivery into the nasal cavity to date
have been forced to make unrealistic simplifying assumptions or have ignored these effects
entirely, as discussed below in section B.III.A.3.c.47

(2) Complexity and dynamics of the nasal cavity

The nose and structures in the nasal cavity, as illustrated in Figure 5 above, are highly
complex and dynamic. They may change in response to physiological, physical, endocrine, and
emotional input. For example, the dimensions of the nasal cavity―in particular, the pliable
segments at and beyond the valve region―dynamically expand and contract with airflow
direction and the associated changes in pressure. The nasal valve, especially its upper narrow
segment, is particularly susceptible to this effect. During inhalation, Bernoulli forces narrow the
valve progressively with increasing inspiratory flow rate and may even cause partial or complete
collapse with vigorous sniffing.48 During exhalation, the valve acts as a “brake” to maintain a
positive expiratory airway pressure that helps keep the pharyngeal and lower airways open and
optimized gas exchange in the lungs.49

The complexity and dynamics of the nasal cavity make effective delivery of medication
and assessment of nasal deposition patterns challenging. The nasal valve, for example, is a
significant barrier to drug particles that are released anterior to the nasal valve.50 The complex
slender anatomy beyond the nasal valve also limits the amount of drug that can reach superior
and posterior areas that may be remote and secluded but critical to efficacy in the treatment of
nasal polyp disease, such as the olfactory epithelium (e.g., olfactory cleft) and middle meatus
where the sinuses ventilate and drain and where nasal polyps typically originate. In addition,
differences in the amount of drug that reaches these areas also can produce a different safety
and adverse event profile due to regional differences in potential exposure to important
structures such as the eye and cranial nerves. Actions such as sniffing at delivery, often
performed by either instruction or spontaneously due to a wet sensation in the nasal vestibule,
may further narrow the nasal valve, additionally limiting posterior deposition. Moreover,
sniffing also draws the drug being administered along a pathway that is mostly along the floor of

46 Halama 1990, supra note 41 (Exhibit 27).

47S. Chari, K. Sridhar, R. Walenga. Computational Analysis of a 3D Mucociliary Clearance Model
Predicting Nasal Drug Uptake. J. Aerosol. Sci. 2021 June; 155:105757 (attached hereto as Exhibit 31).
P. Cole. Nasal and Oral Airflow Resistors. Site, Function, and Assessment. Arch. Otolaryngol. Head
48

Neck Surg. 1992 Aug.; 118(8):790–793 (attached hereto as Exhibit 32).

49P.G. Djupesland, O. Skatvedt, A.K. Brogersen. Dichotomous Physiological Effects of Nocturnal External
Nasal Dilation in Heavy Snorers: The Answer to a Rhinological Controversy? Am. J. Rhinol. Mar.–Apr.
2001; 15(2):95–103 (attached hereto as Exhibit 33); W.M. Hairfield, D.W. Warren, V.A. Hinton et al.
Inspiratory and Expiratory Effects of Nasal Breathing. Cleft Palate J. 1987 July; 24(3):183–189 (attached
hereto as Exhibit 34).
50Cole 1992, supra note 48 (Exhibit 32); R. Fodil, L. Brugel-Ribere, C. Croce et al. Inspiratory Flow in the
Nose: A Model Coupling Flow and Vasoerectile Tissue Distensibility. J. Appl. Physiol. 2005 Jan.;
98(1):288–295 (attached hereto as Exhibit 35).

13
the nose (under the inferior turbinate) directly toward the oral cavity and away from sites more
superiorly located that are targeted with nasal delivery in the treatment of nasal polyp disease.

(3) Variability of the nasal cavity and relevant adjacent

structures

Further, the anatomy of the nasal cavity and mid-facial morphology is highly variable
depending on race, ethnicity, and other factors in ways that interact with the structure and
function of an exhalation delivery system. The XHANCE EDS was designed to provide effective
delivery of fluticasone propionate across a range of anatomies.

First, nasal cavities are complex anatomical structures characterized by high inter-
patient variability.51 Anatomical variants of the nasal cavity, including the septum and middle
turbinate, also play a role in the pathogenesis of inflammatory processes of the nasal cavity and
paranasal sinuses and are more common in chronic rhinosinusitis. This type of variation has
clear potential to interact with the nosepiece, airflow characteristics, and user behaviors with an
exhalation delivery system, in ways that influence drug deposition in regions of interest for the
treatment of nasal polyps. Further, external mid-facial morphology also exhibits an
“exceptionally high level of between and within-group variation” across ethnic populations and
will differently influence fit and function of exhalation delivery systems because they interact
with both the nares and the mouth.52

Multiple studies have evaluated nasal variation in various populations, with the most
commonly studied population being patients presenting to an ear, nose, and throat (“ENT”)
clinic, which is more representative of the population of interest suffering nasal polyp disease.
Rates of anatomic abnormality are higher in the population of interest with chronic
rhinosinusitis. The types and frequency of multiple relevant anatomic variations have been
reported, with the most common being deviated nasal septum (up to 83% prevalence), various
types of concha bullosa (lamellar, bulbous, or extensive types; collectively up to 55%),
pneumatized pterygoid base (up to 73%), Haller cells, and other pneumatized bony structures.53

51 M. Cellina, D. Gibelli, A. Cappella et al. Nasal Cavities and the Nasal Septum: Anatomical Variants and

Assessment of Features with Computed Tomography. Neuroradiol J. 2020 Aug.; 33(4) 340–347 (attached
hereto as Exhibit 36).
52A. A. Evteev, A N. Grosheva. Nasal Vity and Maxillary Sinuses form Variation Among Modern Humans
of Asian Descent. Am. J. Phys. Anthropol. 2019; 169:513–525 (attached hereto as Exhibit 37).
53Cellina 2020, supra note 51, (Exhibit 36); K. Devaraja, S.M. Doreswamy, K. Pujary. Anatomical
Variations of the Nose and Paranasal Sinuses: A Computed Tomographic Study. Indian J. Otolaryngol
Head Neck Surg. 2019 Nov. ; 71(Suppl 3):2231–2240 (attached hereto as Exhibit 38); K.M. Ozcan, A.
Selcuk, O. Ozcan et al. Anatomical Variations of Nasal Turbinates. J. Craniofac Surg. 2008 Nov.;
19(6):1678–1682 (attached hereto as Exhibit 39); H.K.K. Tan, Y.K. Ong. Sphenoid Sinus: An Anatomic
and Endoscopic Study in Asian Cadavers. Clin. Anat. 2007 Oct.; 20(7):745–750 (attached hereto as
Exhibit 40); Y. Lu, J. Pan, S. Qi S et al. Pneumatization of the Sphenoid Sinus in Chinese: The Differences
from Caucasian and its Application in the Extended Transsphenoidal Approach. J. Anat. 2011 Aug.;
219(2):132–142 (attached hereto as Exhibit 41); C.M. Özer, K. Atalar, I.I. Öz et al. Sphenoid Sinus in
Relation to Age, Gender, and Cephalometric Indices. J. Craniofac Surg. 2018 Nov.; 29(8):2319–2326
(attached hereto as Exhibit 42); S. Tomovic, A. Esmaeili, N.J. Chan et al. High-Resolution Computed
Tomography Analysis of the Prevalence of Onodi. Cells. Laryngoscope. 2012 July; 122(7):1470–1473
(attached hereto as Exhibit 43); M.A. Muñoz-Leija, M. Yamamoto-Ramos , F.J. Barrera-Flores et al.

14
These variations can severely distort the nasal anatomy. Other potentially important anatomic
variants, including a 4th turbinate (the supreme turbinate or Santorini’s concha), also occur.

Second, this “static” variation is further compounded by “dynamic” variation noted

above. In addition to changes due to Bernoulli forces, dynamic variation is due to physiologic
variability in the epithelium lining intranasal structures, which is dense with vascular capillaries
and venus sinusoid structures that constantly swell and shrink in response to external
conditions and endogenous stimuli, under the control of the sympathetic nervous system, in a
process called the “nasal cycle” which causes dynamic congestion and decongestion of intranasal
spaces.54 Variable swelling of turbinate tissues due to factors such as inflammation, allergy, or
environmental pollution can result in partial or complete blockage of the channel-like meatuses.
Hormonal changes (pregnancy, premenstrual stage) can also determine prolonged or transitory
turbinate congestion, further suggesting the need for both gender and age-specific modeling―in
addition to modeling of the other sources of variation described―in order to understand
population effects when attempting to simulate in vivo anatomy with in vitro simulations.

Third, race and ethnicity also are recognized as an important source of variation in not
only the internal nasal cavity anatomy, but also in mid-facial morphology and anthropometrics
of regions that directly interact with an exhalation delivery system (e.g., teeth/upper lip distance
from nares), indicating that assessment of performance in a limited racial/ethnic population
introduces significant risk of failure to indicate performance in a broader population even if
intranasal cavity variation were to somehow be fully accounted for. “Systematic nasofacial
analysis” is considered the standard of care by nasal surgeons contemplating rhinoplasty, and
large differences are apparent in frontal, lateral, and basal views of regions that interact directly
with an exhalation delivery system.55 Some examples of this include the shape, size, and
direction of the opening formed by the ala into which the nosepiece is inserted (which can vary
dramatically from antero-posterior oriented to laterally-oriented, with short or long columella
and different columella/lobule ratios) and the nasolabial angle (which can vary from acute to
obtuse and, in conjunction with lip and front teeth, can influence the distance and angle
between lips and nose and change spray angle and direction). Substantiating this recognized
population variation, and as reflected in the design history file, the human factors studies
conducted during design of the specific nosepiece and translating mouthpiece of the XHANCE
EDS also identified a substantial anthropometric range of mid-facial structures that resulted in
different interactions with differently designed devices.

Anatomical Variations of the Ethmoidal Roof: Differences Between Men and Women. Eur. Arch.
Otorhinolaryngol. 2018 July; 275(7):1831–1836 (attached hereto as Exhibit 44); L. Badia, V.J. Lund, W.
Wei et al. Ethnic Variation in Sinonasal Anatomy on CT-Scanning. Rhinology. 2005 Sept.; 43(3):210–214
(attached hereto as Exhibit 45); S.B. Hiremath, A.A. Gautam, K. Sheeja et al. Assessment of Variations in
Sphenoid Sinus Pneumatization in Indian Population: A Multidetector Computed Tomography Study.
Indian J. Radiol. Imaging. 2018 July–Sept.; 28(3):273–279 (attached hereto as Exhibit 46); V. Burulday,
N.B. Muluk, M.H. Akgület al. Presence and Types of Anterior Clinoid Process Pneumatization, Evaluated
by Multidetector Computerized Tomography. Clin. Invest. Med. 2016 June; 39(3):E105–110 (attached
hereto as Exhibit 47).
54 Cellina 2020, supra note 51 ( Exhibit 36).
N.L. Villanueva , P.N. Afrooz, J.A. Carboy. Nasal Analysis: Considerations for Ethnic Variation. Plast.
55

Reconstr. Surg. 2019 June; 143(6):1179e–1188e (attached hereto as Exhibit 48).

15
 The static and dynamic variation in anatomy can interact with the specific pressure and
airflow produced by an exhalation delivery system in ways that are potentially sensitive to even
millimeter variation in initial conditions (for example, angle of delivery resulting from position,
or degree of nasal valve stenting in the most superior aspect versus inferior aspect).

In summary, anatomical characteristics, including anthropometrics, of the nose, mid-

face, nasal cavity, and paranasal sinuses exhibit considerable variation, and many of these
variations depend on age, gender, geography, race, and ethnicity.56 This is further complicated
by a high degree of additional variation due to the indicated disease process (nasal polyps)
which is known to further alter static and dynamic anatomy and physiology due to both the
effects of inflammation (e.g., swelling, impairment of mucociliary motility, epithelial barrier
dysfunction) and the polyp structures themselves. These patient-based sources of variation
interact directly and/or functionally with an exhalation delivery system device, introducing the
risk that non-identical devices may produce materially different drug deposition that is difficult
to predict in vitro. In vitro to in vivo correlation models for exhalation delivery system devices
have not been established and are difficult to develop for the reasons further articulated below;
however, even if they should become possible in the future, a failure to assess a suitable range of
variation in anatomy due to within-population intranasal anatomic variants, and of a suitable
range of racial and ethnic variants in anthropometrics of the midface, would restrict the
generalizability of any conclusions of comparability. Failure to assess a suitable range of
disease-related variation would have a similar consequence. A non-inferiority assessment of
clinical safety and efficacy outcomes is necessary, as discussed below, because for this topically
acting product, the distinct profile of deposition and local absorption is equally or more
important than simply the amount of drug delivered or systemically absorbed and cannot be
predicted in non-generalizable in vitro models.

b) Nasal polyps are variable and arise in the upper parts of the nasal
cavity, typically in the region of the middle meatus or superior
turbinate as well as inside the sinuses, and not on or below the
inferior turbinate.

Nasal polyps are benign lesions that arise from chronically inflamed tissue within the
nasal cavity and are associated with nasal congestion, rhinorrhea, facial pain and pressure, loss
of sense of smell, halitosis, fatigue, impaired sleep, and frequent episodes of acute sinusitis.
These symptoms, among others, contribute to serious morbidity as measured by impairment of
quality of life. The harm to quality of life has been found to be on a par with that associated with
other serious chronic diseases such as chronic obstructive pulmonary disease (“COPD”), chronic
angina, and congestive heart failure (“CHF”).57

Nasal polyps typically arise from the region of the middle meatus but also can be found
higher up on the superior turbinate or on the posterior septum. Nasal polyps also can form on
inflamed mucosa within the paranasal sinuses (typically the ethmoid cells or maxillary sinus).
This polypoid tissue eventually can reach into the nasal cavity via the ostia in the unoperated

56Devaraja 2019, supra note 53 (Exhibit 38); Tan 2007, supra note 53 (Exhibit 40); Lu 2011, supra note
53 (Exhibit 41); Özer 2018, supra note 53 (Exhibit 42); Tomovic 2012, supra note 53 (Exhibit 43);
Muñoz-Leija, supra note 53 (Exhibit 44); Badia 2005, supra note 53 (Exhibit 45); Hiremath 2018, supra
note 53 (Exhibit 46); Burulday 2016, supra note 53 (Exhibit 47); Evteev 2019, supra note 52 (Exhibit 37).
Z.M. Soler, E. Wittenberg, R. J. Schlosser et al. Health State Utility Values in Patients Undergoing
57

Endoscopic Sinus Surgery. Laryngoscope. 2011 Dec.; 121(12):2672–2678 (attached hereto as Exhibit 49).

16
patient or through a surgically created antrostomy. Nasal polyps do not form on the more
inferior structures within the nasal cavity, such as the inferior turbinate or floor of the nasal
cavity. Polyps may be mobile (stalked), in which case they can move with respiration, or
comparatively immobile and be in the form of bulk tissue extending from the lining of the
mucosal surface. They almost always present bilaterally and often are quite different in
conformation (shape/size) on each side of the nose. Left untreated, nasal polyps typically will
expand, filling up the spaces within the nasal cavity, inhibiting airflow, and contributing to the
symptoms of congestion and hyposmia commonly reported by patients suffering from this
condition. As polyps continue to expand, they will extend inferiorly and even can reach beyond
the nares if left untreated.58

Although nasal polyps originate from chronically inflamed tissue, they also exacerbate
the inflammation within the nasal cavity. Nasal polyps release large amounts of IL-5, which is a
cytokine known to attract eosinophiles that degranulate within the nasal cavity, exacerbating the
inflammatory burden and further worsening the associated swelling that is typical of this
disease. They also are associated with increased tissue IL-4, IL-13, S. Aureus enterotoxin
(superantigens), and other inflammatory mediators, which may help explain why patients with
nasal polyps are known to suffer from “epithelial barrier dysfunction” (including loss of tight
junction functioning) which has obvious potential to alter drug tissue penetration in specific
segments of the nasal cavity that are so affected. Patients with nasal polyps also tend to have
impaired ciliary clearance. The degree and specific anatomic locations of this impairment can
be quite variable and highly influenced by the severity of inflammation and, therefore, should be
assumed to be inconsistent in different regions of the nasal cavity. Moreover, this impairment is
thought to be reversible with effective treatment, introducing an additional variable in modeling
mucociliary clearance that depends on treatment effects and disease progression; the rate and
degree of improvement varies substantially.59

Patients with nasal polyps often undergo sinus surgery, and a large fraction of patients
who are candidates for nasal medication for nasal polyps have undergone prior sinus surgery
(approximately 30% in one XHANCE pivotal trial and 60% in the other pivotal trial).60
Endoscopic sinus surgery (“ESS”) is not standardized, and multiple variations in the magnitude
of surgical intervention are common. Typically, ESS involves removal of tissue (for example,
nasal structures such as the uncinate process and/or middle turbinates), and the creation of
variably-sized antrostomies between some or all of the paranasal sinuses and nasal cavity,
greatly increasing variability of nasal geometry. This additional “surgically induced” anatomic
variation in polyp patients can influence airflow, pressure, site of polyp recurrence, and other
factors material to prediction of deposition, safety, and efficacy associated with exhalation
delivery system delivery.

58W. W. Stevens, R. P. Schleimer, R.C. Kern. Chronic Rhinosinusitis with Nasal Polyps. J. Allergy Clin.
Immunol. Pract. 2016; 4(4): 565–572 (attached hereto as Exhibit 50).
59K. E. Hulse, W. W. Stevens, B. K. Tan. Pathogenesis of Nasal Polyposis. Clin. Exp. Allergy. 2015 Feb.;
45(2): 328–346 (attached hereto as Exhibit 51).
60 D.A. Leopold, D. Elkayam, J.C. Messina, et al. NAVIGATE II: Randomized, Double-Blind Trial of the
Exhalation Delivery System with Fluticasone for Nasal Polyposis. J Allergy Clin. Immunol. 2019 Jan.;
143(1):126–134.e5 (attached hereto as Exhibit 52); R. Sindwani, J.K. Han, D.F. Soteres et al. NAVIGATE
I: Randomized, Placebo-Controlled, Double-Blind Trial of the Exhalation Delivery System With
Fluticasone for Chronic Rhinosinusitis With Nasal Polyps. Am. J. Rhinol. Allergy. 2019 Jan.; 33(1):69–82
(attached hereto as Exhibit 53).

17
 Because fluticasone propionate is a topically acting medication, deposition on the nasal
polyps themselves is required for them to be reduced, and the magnitude of polyp reduction will
be dependent upon the amount of steroid that can reach the tissue, the polyp tissue surface area
that is coated, and the amount of drug that continues to reach the tissue as it shrinks. The same
is true for mucosal tissues swollen due to inflammation related to the underlying nasal polyp
disease process.

c) The XHANCE EDS is designed to deliver the fluticasone

propionate suspension to the most superior and posterior regions
of the nose.

The XHANCE EDS is specifically designed to deposit drug aerosol on structures and
tissues that are in more superior and more posterior obstructed spaces in the nasal cavity (e.g.,
olfactory region, middle meatus, osteomeatal complex, ethmoid sinus cavity). Through the
XHANCE EDS, the product produces a different distribution of intranasal drug deposition than
a conventional nasal spray and disproportionately places topically acting steroid in the region of
the nasal cavity where nasal polyps typically originate and where paranasal sinuses ventilate and
drain. In addition, the XHANCE EDS has been shown to deposit drug inside sinus cavities that
have been surgically opened to the nasal cavity, which is often the case in the context of patients
with nasal polyp disease61.

The XHANCE EDS achieves this particular deposition pattern through known and
unknown factors sensitively related to critical device components and also to dynamic
interactions with patients’ variable anatomy and physiology. The critical components and
dimensions of the device constituent of XHANCE include not only those for a conventional nasal
spray, such as spray pump and applicator, but also include the device subassembly and its
components that affect the mechanics of the overall performance of the device, including the
nosepiece, mouthpiece, geometry between the nosepiece and mouthpiece, T-valve, grip, and
others.

Specifically, the XHANCE EDS contains a spray pump that produces a metered dose of
medication that is synchronized to be released with the “burst” of exhaled breath when actuated.
The XHANCE EDS also interfaces with both the external nares and nasal valve and the mouth
and serves as conduit for the exhaled breath and the associated pressure to move from the user’s
mouth into the nose in a way that allows for the aerosol droplets to be entrained into the airflow.
Lastly, the asymmetrical nosepiece of the XHANCE EDS seals at the nostril to allow transfer of a
balancing, or proportional, amount of pressure into the nose from the oral cavity (to control the
degree of soft palate elevation), stents the narrow nasal valve (especially superiorly), and
displaces soft tissue without obstructing drug exit from the device, all facilitating deposition of a
larger proportion of medication both beyond the nasal valve and above the inferior turbinate
(i.e., superiorly and posteriorly) in order to reach regions of the nose targeted for treatment of
nasal polyps.

61P. G. Djupesland, J.C. Messina, J.N. Palmer. Deposition of Drugs in the Nose and Sinuses With an
Exhalation Delivery System vs Conventional Nasal Spray or High-Volume Irrigation in Draf II/III Post-
Surgical Anatomy. Rhinology 2020; 58: 2, 175–183 (attached hereto as Exhibit 54).

18
 d) The design of the XHANCE EDS and the interaction of the device
with the patient’s anatomy and physiology are central to the
clinical effect of the drug product.

The design of the XHANCE device and the interaction of the device with the patient’s
anatomy and physiology are central to the product’s drug deposition, and, accordingly,
XHANCE’s clinical effects.

First, the XHANCE EDS creates an airflow that has a key impact on the deposition of
drug within the nose. The XHANCE EDS releases the user’s breath as the bottle is depressed
and simultaneously opens the internal valve inside the device to create an abrupt “burst of air”
synchronized with the release of the spray. The device has important design aspects that
influence the flow of the air into the nose. For example, the XHANCE EDS can prevent
“leakage” of air, and associated pressure, as the breath travels through. This flow efficiency
affects both the volume and velocity of the airflow that enters the nose, and varies depending on
the exhalation pressure and flow. Additionally, the XHANCE EDS synchronizes the release of
the breath with the release of the spray such that the droplets are entrained in airflow at a
particular point in time within the variable airflow curve (flow rate by time curve), contributing
to the product’s unique drug deposition.

For example, at one extreme, if the breath instead were to be released largely after the
drug aerosol spray, many of the droplets would be deposited before being entrained in the
breath, producing highly altered drug deposition within the nose. Similarly, if the drug aerosol
spray were to be released after the peak of the exhaled air “burst,” the airflow carrying the
droplets would be substantially different purely on the basis of timing. The direction, volume,
and velocity of airflow directed into the nasal cavity will change both the initial deposition of the
spray and the subsequent movement of initially deposited drug, in part by altering the plume
and spray pattern inside the nose and changing droplet size (as airflow causes droplets to impact
one another, thereby altering size). As airflow continues beyond the initial deposition of the
spray (during both the initial breath and during a subsequent breath when using the 2-spray
dose), laminar airflow along mucosal surfaces will continue to move the deposited liquid further
back into the nasal cavity for the duration of the exhalation process, typically lasting 2-3
seconds. This effect of moving already deposited drug particles more superiorly and posteriorly
is further accentuated when a second dose is administered to the same nostril, a dose that is
consistent with currently labeled use of the product. Therefore, the amount of air and the
velocity and pressure of the airflow that is passed through the device will have important effects
on the deposition of fluticasone suspension at the specific sites of action within the nasal cavity
that are relevant for the treatment of nasal polyps.

Airflow and pressure after exiting the device within the nasal cavity also are changed by
the resistance of the air circuit. In a nasal geometry congested by anatomic variation, mucosal
inflammation, and/or nasal polyps, more of the exhaled force will be transferred to pressure and
therefore may change the conformation of the narrow cavity anterior to the obstruction, with a
lower flow rate. The expansion and lower flow rate will enhance the drug deposition on the
obstructing tissues localized anteriorly. With less mucosal swelling and polyp volume, with
variation not only due to intrinsic or disease-related patient differences, but also during
treatment due to the effects of the treatment, there may be less resistance, and the flow rate will
increase and propel droplets progressively deeper into the nasal cavity. This dynamic
interaction between the physiological and pathological status of the sinonasal airway and the
deposition is difficult to model across the full range of expected variation in the target
population, further calling into question the validity of in vitro models that do not account for

19
this variation for assessing equivalence and highlighting the need for non-inferiority clinical
endpoint studies in a suitable target population, as discussed below.

Second, the transfer of the positive pressure associated with the airflow is another
important design feature of the XHANCE EDS and impacts drug deposition within the nasal
passages and target sites. Per the approved Instructions for Use, the user blows into the device
and, while continuing to blow, presses the vial upward.62 The valve within the XHANCE EDS
prevents airflow release through the nosepiece until the valve is opened. This allows pressure to
build within the device before actuation. The amount of “leakage” of air through other parts of
the device (“flow efficiency”) will impact the amount of pressure that will be transferred from
the device into the nasal cavity during use. These pre- and post-actuation flow efficiencies will
affect the force and associated velocity of the airflow through the nosepiece once actuated,
which, as noted above, will influence deposition along with spray pattern, plume geometry, and
droplet size within the nasal cavity.

The transfer of the positive pressure also has a dynamic effect on the nasal cavity itself.
This positive pressure transfer will cause the soft tissues within the nasal passages to expand in
a manner that depends on the magnitude, timing, and direction of the pressure applied. The
positive pressure will affect airflow parameters within the entire nasal cavity circuit, thereby
affecting drug deposition within the nasal passages and target sites.

This is especially true for the area where the nasal valve is located. The transfer of
positive pressure that occurs with the spray release from the nozzle will influence the size and
shape of the opening produced in the nasal valve, which is a key factor in determining the
amount of drug that reaches targeted parts of the nasal passages. Similarly, if the soft palate is
either not sealed, or if the counterbalancing pressures across the soft palate (the oral cavity
pressure from below the vellum balanced against the “mid-circuit” pressure in the nasopharynx
from above the vellum) produce a different positioning of the soft palate during delivery, then
the airflow circuit also will be altered, which will in turn alter airflow, pressures, and deposition.
Notably, this is different from the considerations that apply to conventional nasal sprays, which
interact differently with the patient and nasal cavity. With conventional nasal sprays, not only is
there no transfer of proportional positive pressure to consider, but the opposite: negative
pressures associated with “sniffing” a conventional nasal spray can collapse narrow spaces due
to Bernoulli’s forces. This highlights how dramatic device-related differences can be depending
on device design. Even small differences in the device, such as nosepiece stenting design, soft-
tissue obstruction of device outflow, and others will have potential to influence drug deposition
when deposition is assessed with the anatomic specificity appropriate to nasal polyp disease and
to the wide degree of variation in the target population.

Third, the nosepiece of the XHANCE EDS plays an essential role in achieving the
product’s drug deposition. The nosepiece is designed specifically with a form that achieves a
number of functions that influence drug deposition. The specific XHANCE EDS nosepiece is
designed to slide into the anterior part of the nose to create a seal with the tissue at its base in a
broad range of nasal sizes and shapes and, at the same time, stent open the upper segments of
the narrow nasal valve. For airflow and the associated positive pressure to be proportionately
transferred into the nasal cavity from the device, the pressure drop across the device must be
controlled and, notably, the nosepiece must create a seal with the nostril to prevent air and
pressure from escaping retrograde out of the nostril of administration.

62 XHANCE Instructions for Use (Apr. 2021) (Exhibit 6).

20
 At the same time, it is important that the position and angle of the inserted nosepiece
remain stable. For this to be achieved, the user must be able to insert the mouthpiece between
the lips while still maintaining a seal around the nostril and a consistent position with the
nosepiece (“2-point fixation”). The XHANCE EDS has a flexible mouthpiece that is designed,
and has been demonstrated in human factors studies, to accommodate substantial differences in
the distances between the mouth and nose in different people in order to ensure that a seal
within the nostril and a stable position of the nosepiece can be achieved while still placing the
mouthpiece between the lips. The physical design and architecture of the body of the XHANCE
EDS are engineered to maximize fit in a diverse patient population as it relates to their mid-
facial and nasal anatomic features and dimensions. Elements of the mouthpiece design and
device geometry are important to assure the fit needed to correctly accommodate the anatomic
diversity of a large portion of users.

Variations, such as in the angles between the mouthpiece and the device body or in the
shape or size of the mouthpiece or nosepiece, have the potential to alter the “2-point fixation”
between the mouth and nose and, therefore, the angle of the spray applicator and direction of
delivery of the drug aerosol, and also to alter the interaction of the nosepiece and soft tissues of
the nasal vestibule and nasal valve in ways that change drug deposition.

The nasal valve is the narrowest part of the entire respiratory tract and plays an
important role in the physiology of the nose and interacts closely with the lower airways in many
ways. The nasal valve is located between 2-3 cm into the nostril and is one of the significant
barriers to delivering drug to the more superior and posterior parts of the nasal cavity targeted
for treatment of nasal polyps. The multi-dimensionally asymmetrical nosepiece used on the
XHANCE EDS is elongated and shaped in such a way that it stents the nasal valve at its
narrowest point without creating undue discomfort, or risk of mucosal erosion, at the site of
device interaction because the generally “triangular” form of the nasal valve is mimicked. This
ensures that the air and associated positive pressure exiting the device at the time of actuation
can create the maximum opening of the nasal valve and posterior soft tissues, allowing more of
the spray droplets to get beyond the nasal valve into the target regions of the nasal passages.

Another key consideration is that the nosepiece design was intended to reduce
unintentional obstruction of the spray that could be caused by compression of intranasal soft
tissues during insertion, and to be acceptably comfortable to a broad range of end users.
Ultimately, the nosepiece shape was selected with a specific configuration that best conformed
to the patient’s anatomy and the XHANCE EDS’s intended functions.

Relatively minor variations from the specific design may behave differently in all, or
some, individual users across a diverse population. Results from human factors evaluations
identified potentially important differences in nosepiece design with regard to risk of
obstruction of the exit due to compression primarily of lateral soft tissue, reliability of producing
a seal at the nostril entrance, adequate stenting of the superior aspect of the narrow nasal valve,
and reported comfort, particularly in individuals with smaller noses and/or “shorter” nasofacial
dimensions. As a result of these evaluations, the specific shape and opening at the tip of the
nosepiece were determined to be important to assure a seal at the nostril with varying nostril
shapes and sizes, to achieve the desired in-use performance (e.g., stenting pattern and relief of
soft tissue obstruction), to reduce risk of discomfort or harm when in contact with nasal tissues,
and to accommodate diverse users. The specific configuration of the nosepiece itself, and its
relationship with the mouthpiece, is an important component of the design for any exhalation
delivery system and will have an impact on the factors described above and, therefore, on both
the airflow and positive pressure transferred into the nasal cavity. Different nosepieces also will
influence the number and size of droplets able to pass beyond the nasal valve region, and in

21
what direction and with what force and flow, to reach the specifically targeted subset of regions
that are behind the nasal valve and above the inferior turbinate, and will behave differently
across a spectrum of patients due to patient-level variation.

Additionally, the grip of the XHANCE EDS has a shape that permits a patient to use the
product with either a one-handed or two-handed hold, and also was evaluated with either same-
sided or opposite-sided administration. These grip differences can alter the angle of the device
nozzle in the nostril and greatly influence the direction of spray. Alterations to the grip have the
potential to change the manner in which a patient uses the device in ways that change the
direction of spray or position of the nosepiece in the nasal valve, resulting in potentially major
differences in the amount of drug passing beyond the nasal valve and above the inferior
turbinate.

The XHANCE EDS also offers potential therapeutic effects, including those related to
carbon dioxide, nitric oxide, change in pH, or positive pressure, the direct effects of which are
not well characterized, that are independent and/or complementary to the therapeutic effect of
the drug being delivered.63 As noted in the XHANCE labeling, for example, the carbon dioxide
present in the exhaled breath delivered into the nose through the XHANCE EDS may influence
inflammatory mediator activity and neuropeptide activity.64 These effects would have to be
replicated by a differently designed device in order for a proposed generic product to be
“expected to have the same clinical effect and safety profile” as XHANCE, and, therefore, to
achieve therapeutic equivalence to XHANCE.65

In summary, the design of the XHANCE EDS, including its critical components, acts in a
particular way to achieve drug deposition that is the result of both the device design and by
interaction with the nose and nasal cavity, and mouth and oral cavity, and with the static and
dynamic anatomy and physiology of the patient. Deposition of the topically acting drug is
central to the safety and efficacy of XHANCE for the treatment of nasal polyps. Even if a
proposed generic liquid drug formulation is identical, differences in the delivery device can
produce materially different activity, especially through different drug deposition, with the clear
potential to alter safety or efficacy in ways that would cause the proposed product not to be
therapeutically equivalent to XHANCE.

2. The in vitro methods that FDA recommends for conventional nasal

spray products are necessary but not sufficient to identify safety and
efficacy differences resulting from differences from the XHANCE EDS.

FDA’s Nasal Spray Draft Guidance recommends the following seven in vitro methods to
evaluate bioequivalence for locally acting drugs delivered by nasal aerosol or nasal spray: (1)
single actuation content through container life; (2) droplet size distribution by laser diffraction;
(3) drug in small particles/droplets, or particle/droplet size distribution by cascade impactor;
(4) drug particle size distribution by microscopy; (5) spray pattern; (6) plume geometry; and (7)
priming and repriming.66 These methods are necessary, but insufficient without additional data,
to establish that a proposed ANDA product generates an equivalent drug deposition within the

63 See note 7, supra.

64 XHANCE Prescribing Information § 12.1 (Apr. 2021) (Exhibit 6).
65 21 C.F.R. § 314.3(b) (defining “therapeutic equivalents”).
66 Nasal Spray Draft Guidance, supra note 23, at 9–10 (Exhibit 17).

22
nose to that produced by the XHANCE EDS, including through the interaction of the device with
the patient’s anatomy and physiology.

The XHANCE EDS introduces multiple new variables into the intranasal delivery process
that influence drug deposition in the nasal cavity in ways that are difficult to predict. Some of
these variables relate to the device itself, and other variables relate to the device’s interaction
with the patient, patient’s anatomy (static, dynamic, and disease-related), and patient’s normal
physiology. The in vitro assessments of droplet size, plume geometry, spray pattern, and other
properties that FDA generally recommends for conventional nasal spray devices are insufficient
to fully evaluate a proposed generic version of the XHANCE EDS. As discussed, the XHANCE
EDS interacts directly with the nasal anatomy (e.g., sealing the nostril, stenting the nasal valve,
providing an initial burst of air at actuation synchronized with spray emission, and generating a
subsequent continuous high airflow lasting up to 2-3 seconds which dynamically expands the
nasal cavity as well as―along with subsequent actuations―moving already-deposited drug even
higher and deeper). These interactions will influence droplet size, spray pattern, and plume
geometry in the nose, and will modify the nasal valve and other nasal geometry, in addition to
re-routing airflow and modifying the “air resistance circuit” as air that reaches the nasopharynx
reverses course due to the sealed soft palate. Physiologic behavior and anatomy have the
potential to be different based on race, gender, disease state, and other factors. Therefore, in
vitro testing of droplet size, spray pattern and plume geometry between XHANCE and a test
product remain necessary in the assessment of equivalence; however, an evaluation of these
factors alone is insufficient to establish equivalent drug deposition within the nose in a diverse
target population.

Similarly, the in vitro morphologically-directed Raman spectroscopy (“MDRS”) method

that supported FDA’s approval of generic versions of NASONEX® (mometasone furoate nasal
spray) is insufficient to assure equivalent drug deposition within the nose between XHANCE
and a proposed generic product.67 MDRS measures particle morphological characteristics (size
and shape) using its microscopic component, and performs chemical identification by analyzing
Raman spectra.68 The MDRS method was used to characterize the particle size distribution
(“PSD”) of active pharmaceutical ingredient (“API”) in the suspension drug product by
distinguishing the API from excipients in the liquid drug suspension. FDA’s general draft
product-specific guidance for proposed generic products referencing FLONASE includes this
MDRS method as a potential alternative to an in vivo clinical endpoint study to support
bioequivalence. However, MDRS is insufficient to assess the bioequivalence and therapeutic
equivalence of a proposed generic version of the XHANCE EDS. MDRS does not assess
intranasal drug deposition at all, and only assesses the distribution of particles suspended in the
liquid formulation in the vial prior to administration. Therefore, it does not address important
potential differences between differently designed exhalation delivery system devices, either
dimensionally or in functional interaction with the patient. Most notably, it does not address
whether there may be differences in the deposition of the aerosol droplets introduced into the
nasal cavity by differently designed devices. It also does not address differences in potential
direct device therapeutic effects independent of the drug delivered.

Lastly, we recognize that FDA has issued draft product-specific guidance documents for
certain orally inhaled drug products that provide for in vitro studies in conjunction with other

67 Liu 2019, supra note 27 (Exhibit 18).

68 Id. at 16.

23
types of data as an alternative approach to conducting comparative clinical endpoint studies.69
However, methods used for the evaluation of orally inhaled drug products cannot be presumed
to be appropriate for locally acting nasal sprays. Orally inhaled drugs are introduced into a
much larger organ system, the lungs, with a dissimilar degree of normal or disease-related
variation relative to the nasal cavity. Additionally, there is a greater sensitivity to small
distances in the (significantly smaller) nasal cavity, where a fraction of a centimeter can be
material, and because of major differences in anatomic specificity and other disease effects
between nasal polyp disease and more diffuse diseases such as COPD or asthma.

3. There is no alternative to a non-inferiority study to ensure that a

proposed generic product is as safe and effective as XHANCE.

In recent years, alternative methods for modeling nasal cavity deposition have been
proposed, including imaging using radiolabeled drug and gamma scintigraphy cameras in
humans, in vitro models using 3D-printed silicone casts, and computational fluid dynamics
(“CFD”). However, no standard has been established for any of these methods, none has
established an in vitro-to-in vivo correlation, and none accounts for the relevant population
variations that are known to occur. Therefore, none of these methods is valid for establishing
equivalence in drug deposition at the sites of action for the target population. These methods
cannot support a conclusion that a proposed generic product would be expected to have the
same safety and efficacy as XHANCE. Below we briefly address some of the limitations
associated with these methods.

a) In vivo imaging

Many studies have sought to quantify deposition by using radiolabeled solutions coupled
with scintigraphy measurement of deposition.70 The high degree of technical difficulty of these
studies has resulted in small patient sample sizes, limiting generalizability to diverse
populations. Further, there are limits on the dose of radioactivity that safely can be
administered to the nose in these experiments, particularly because of concerns over ocular
exposure. As a consequence, in order to obtain sufficient sensitivity of measurements at the
radioactivity doses that are used, even with comparatively coarse image resolution, gamma
emissions must be recorded over a relatively prolonged time period of 1-2 minutes (a “long
exposure image”). The coarse image resolution limits precise characterization of deposition in
the small but relevant anatomic target regions that are relevant in nasal polyp disease. Despite
these limitations, in vivo imaging can be a useful tool for illustrating dramatic differences in
deposition patterns such as those produced by a conventional nasal spray and exhalation
delivery systems, see Figure 1A, supra. However, the sensitivity with which drug delivery to
specific locations within the nose is delineated and quantified and methodological factors
addressing spatial scattering of the radioactivity and differences in tissue attenuation in
different regions of the nasal airway, preclude the use of such imaging for establishing

69See, e.g., FDA, Draft Guidance on Tiotropium Bromide Spray, Metered; Inhalation (Nov. 2020),
https://www.accessdata.fda.gov/drugsatfda_docs/psg/PSG_021936.pdf (attached hereto as Exhibit 55).
70A. Skretting, P.G. Djupesland. A New Method for Scintigraphic Quantification of Deposition and
Clearance in Anatomical Regions of the Human Nose. Nucl. Med. Commun. 2009 Aug.; 30(8):629–638
(attached hereto as Exhibit 56).

24
equivalent drug deposition produced by different exhalation delivery system devices at the site
of nasal polyps where differences can be expected to affect each product's safety and efficacy.71

b) Use of nasal airway models

Similar to in vivo imaging, nasal casts can detect the dramatic difference in deposition
patterns generated by a conventional nasal spray and the XHANCE EDS, see Figure 1B, supra;
however, they are subject to significant limitations which preclude the use of such models for
establishing equivalent drug deposition produced by different exhalation delivery system
devices at the site of nasal polyps. These limitations include, among others, challenges in
defining air tissue boundaries when producing the airway model and the inaccurate static
and/or dynamic anatomy at the nasal valve and in other areas, unrepresentative tissue pliability,
and inaccurate simulation of the soft palate, and the impact of these items on nasal
aerodynamics.72 Additionally, there is a lack of in vitro to in vivo correlation using nasal casts as
it relates to the efficacy and safety of a topically acting medication in regions of the nose affected
by nasal polyps, and no clear pathway to simulate the exceptionally broad range of patient and
disease population variation that occurs in this context.73 This correlation is further limited by
the large in vivo intra-patient variability in anatomy in particular as regards the location, shape
and dimensions of the nasal valve area and the lack of pathological models. Given the various
targets pursued through nasal drug deposition (including the middle meatus, olfactory region,
sinuses, and nasal polyps, among others) and the complexity and dynamics of the nasal
anatomy, the correlation between nasal drug deposition and some biological effects may be less
straightforward than for other targets of inhaled drug delivery, such as lung deposition.74

c) Computational fluid dynamics

An alternative approach to examining the deposition of drug from nasal sprays is “in
silico” using 3D computer modeling. While many studies have “validated” the observed in silico
results against in vitro tests (e.g., nasal casts; see limitations discussed above), this has typically
involved simply confirming the amount of drug that is deposited on either side of a simplified
assumption for the nasal vertical valve location.75 A key challenge for diseases such as sinusitis

Skretting 2009, supra note 70 (Exhibit 56); A.S. Shah, R. L. Berger, J. McDermott et al. Regional
71

Deposition of Mometasone Furoate Nasal Spray Suspension in Humans. Allergy Asthma Proc. 2015 Jan.–
Feb.; 36(1):48–57 (attached hereto as Exhibit 57).
72Fodil 2005, supra note 50 (Exhibit 35); K. Zhao, J. Jiang. What is Normal Nasal Airflow? A
Computational Study of 22 Healthy Adults. Int’l Forum Allergy Rhinol. 2014 June; 4(6):435–446
(attached hereto as Exhibit 58).
S. Le Guellec, S. Ehrmann, L. Vecellio. In Vitro - In Vivo Correlation of Intranasal Drug Deposition.
73

Adv. Drug Deliver Review. 2021 Mar.; 170:340–352 (attached hereto as Exhibit 59).
74R.L. Walenga, P.W. Longest. Current Inhalers Deliver Very Small Doses to the Lower Tracheobronchial
Airways: Assessment of Healthy and Constricted Lungs. J. Pharm. Sci. 2016. Jan.; 105(1):147–159
(attached hereto as Exhibit 60).
75Chari, supra note 47 (Exhibit 31); S. Hosseini, T.A. Schuman, R. Walenga et al. Use of Anatomically-
Accurate 3-Dimensional Nasal Airway Models of Adult Human Subjects in a Novel Methodology to
Identify and Evaluate the Internal Nasal Valve. Comput. Biol. Med. 2020 Aug.; 123:103896 (attached
hereto as Exhibit 61); Shah 2015, supra note 71 (Exhibit 57).

25
and nasal polyps is that the targeted regions are far more specific and are in deeper, more
posterior regions of the nose (e.g., the middle meatus, olfactory cleft, or ethmoid sinus cavity),
and the degree of relevant variation (patient- or delivery-related) is much greater.76 Simply
demonstrating that the amount of drug reaching beyond the nasal valve (vertical dividing line
often incorrectly called the nasal valve) is the same between a test product and reference
product is inadequate for establishing equivalence of drug deposition, both in the context of a
condition such as nasal polyps and in the context of an exhalation delivery system.77

The impact of mucociliary clearance on the location where fluticasone penetrates

through the mucus layer into a target site of action is another important factor in attempting to
model efficacy and safety and also is a challenge for computational fluid dynamics (“CFD”).78 To
accurately use in silico modeling of intranasal drug deposition, the distribution of ciliated/non-
ciliated cells and of goblet cells in the nasal cavity and sinuses must be understood, as must the
functional variation in different internasal anatomic sites, including the mucus layer, relative
production rate of mucus (affecting drug dissolution/penetration), density of ciliated cells,
mucociliary clearance rates, and direction of clearance.79 Due to the different density and
distribution of ciliated cells and goblet cells, mucociliary clearance is highly variable in different
regions of the nasal mucosa. Total clearance patterns for the entire nose become misleading,
even if separated into anterior and posterior segments.80

Even if data on cell density and distribution (squamous epithelium, ciliated cells, goblet
cells, olfactory cell/filaments) are taken into account, it is essential to be aware that the available
data are based on biopsies from cadavers of 4 individuals aged 58 to 78 who died from

76Larsen 2004, supra note 40 (Exhibit 25); R. Richard, Orlandi et al. International Consensus Statement
on Allergy and Rhinology: Rhinosinusitis 2021. Int’l Forum Allergy Rhinol. 2021; 11:213–739 (attached
hereto as Exhibit 62); W.J. Fokkens, V.J. Lund, C. Hopkins et al. European Position Paper on
Rhinosinusitis and Nasal Polyps 2020. Rhinology. 2020 Feb. 20; 58(Suppl S29):1–464 (attached hereto
as Exhibit 63); Djupesland 2013 supra note 3, (Exhibit 2).
77P.G. Djupesland, A. Skretting, M. Windern et al. Breath Actuated Device Improves Delivery to Target
Sites Beyond the Nasal Valve. Laryngoscope. 2006 Mar.; 116(3):466–472 (attached hereto as Exhibit 64);
P.G. Djupesland , A. Skretting, Nasal Deposition and Clearance in Man: Comparison of a Bidirectional
Powder Device and a Traditional Liquid Spray Pump. J Aerosol. Med. Pulm. Drug Deliv. 2012 Oct.;
25(5):280–289 (attached hereto as Exhibit 65).
78 Shah 2015, supra note 71 (Exhibit 57); P.G. Djupesland, R. Mahmoud. Letter to the Editor: Incorrect

Conclusions Regarding Deposition of Conventional Mometasone Furoate (MF) Nasal Spray. Allergy
Asthma Proc. 2015 Sept.–Oct. 36(5): e104–e105 (attached hereto as Exhibit 66); Djupesland 2006, supra
note 77 (Exhibit 64); Djupesland 2012, supra note 77 (Exhibit 65).
79S. Gizurarson. The Effect of Cilia and the Mucociliary Clearance on Successful Drug Delivery. Biol.
Pharm. Bull. 2015; 38(4):497–506 (attached hereto as Exhibit 67); Halama 1990, supra note 41 (Exhibit
27); C. Rusznak, J.L. Devalia, S. Lozewicz et al. The Assessment of Nasal Mucociliary Clearance and the
Effect of Drugs. Respir. Med. 1994 Feb.; 88(2):89–101 (attached hereto as Exhibit 68); F. Féron, C. Perry,
J.J. McGrath et al. New Techniques for Biopsy and Culture of Human Olfactory Epithelial Neurons. Arch.
Otolaryngol. Head Neck Surg. 1998 Aug.;124(8):861–866 (attached hereto as Exhibit 69).
80Shah 2015, supra note 71 (Exhibit 57 ); Djupesland 2015, supra note 78 (Exhibit 66); Djupesland 2006,
supra note 77 (Exhibit 64); Djupesland 2012, supra note 77 (Exhibit 65).

26
cardiopulmonary incidents and who lacked sinonasal pathology.81 These limitations are not
recognized and addressed in the recent CFD studies arguing that CFD analysis may serve as a
model to predict nasal drug deposition and uptake.82 Furthermore, the impact of acute and
chronic inflammatory sinonasal disorders causing mucosal congestion and modifying the rate of
secretion production and mucociliary clearance is not taken into account in the simplified model
applied.83 Chronic inflammation causes remodeling of the entire sinonasal mucosa with entirely
changed cell distribution.84 In addition, polyp formation, surgical procedures, and medical
intervention will significantly alter the anatomy, physiology, and aerodynamics of the sinonasal
geometry. Assumptions related to these factors intended to simplify and enable in silico
modeling must be accurate and well-substantiated, which so far has not been possible despite
published efforts.85 Accurate modeling of the nasal valve (e.g., not a simplified assumption of a
vertical plane) and other intranasal features, all subject to a high degree of static and dynamic
disparity across patients and over the course of time, also is necessary. It is crucial to
understand that there is great variation, relevant to both efficacy and safety in the treatment of
nasal polyps, among regions that have historically been characterized simply as “behind the
nasal valve.” Recent studies accurately have described the complexity and variability of the
nasal valve and the significant inter- and intra-subject variability in deposition in nasal replicas
from healthy individuals.86 Nonetheless, more recent publications by some of the same authors
do not address these important findings regarding the dynamic features of the nasal valve
anatomy, airflow, and the impact of inflammation, polyp formation, and surgery,87 in favor of a
simplified model of mucociliary clearance and drug uptake.88 Furthermore, with the
introduction of novel devices like the XHANCE EDS―a device where delivery and drug

81 Halama 1990, supra note 41 (Exhibit 27).

82Chari, supra note 47 (Exhibit 31); Y. Shang, K. Inthavong, J. Tu. Development of a Computational Fluid
Dynamics Model for Mucociliary Clearance in the Nasal Cavity. J. Biomech. 2019 Mar. 6; 85:74–83
(attached hereto as Exhibit 70).
83V.K. Pandya, R.S.Tiwar. Nasal Mucociliary Clearance in Health and Disease. Indian J. Otolaryngol
Head Neck Surg. 2006 Oct.; 58(4):332–334 (attached hereto as Exhibit 71); Gizurarson 2015, supra note
79 (Exhibit 67); S.M. Birdi, S. Singh, A. Singh. Mucociliary Clearance in Chronic Sinusitis. Indian J.
Otolaryngol Head Neck Surg. 1998 Jan.; 50(1):15–19 (attached hereto as Exhibit 72).
84 Fokkens 2020, supra note 76 (Exhibit 63); Hulse 2015, supra note 59 (Exhibit 51).
85Chari 2021, supra note 47 (Exhibit 31); P. W. Longest, A. Rygg, and M. Hindle. Bioequivalence Testing:
Can Systemic Pharmacokinetic Profiles from Corticosteroid Nasal Sprays Be Used to Elucidate Local Drug
Deposition within the Nose? Respiratory Drug Delivery 2016 (attached hereto as Exhibit 73); Shang 2019,
supra note 82 (Exhibit 70).
86Hosseini, 2020, supra note 75 (Exhibit 61); M.D. Manniello, S. Hosseini, A. Alfaifi et al. In Vitro
Evaluation of Regional Nasal Drug Delivery Using Multiple Anatomical Nasal Replicas of Adult Human
Subjects and Two Nasal Sprays. Int’l J. Pharm. 2021 Jan. 25; 593:120103 (attached hereto as Exhibit 74).
87 K. Zhao, P.W. Scherer, S.A. Hajiloo et al. Effect of Anatomy on Human Nasal Air Flow and Odorant

Transport Patterns: Implications for Olfaction. Chem. Senses. 2004 June; 29(5):365–379 (attached
hereto as Exhibit 75); M. R. Wofford, J.S. Kimbell, D.O. Frank-Ito. A Computational Study of Functional
Endoscopic Sinus Surgery and Maxillary Sinus Drug Delivery. Rhinology. 2015 Mar. ;53(1):41–48
(attached hereto as Exhibit 76).
88 Chari, supra note 47 (Exhibit 31).

27
deposition occur through an active and dynamic interaction between the patient, the device
design, and flow and pressure properties that constantly change with the state of sinonasal
pathology―such simplified CFD models lack utility.

One of the functions of the nasal cavity is olfaction, the loss of which is a safety concern
for intranasal treatments and also a cardinal symptom of nasal polyp disease. Olfactory sense is
highly location-dependent across small anatomic regions, reinforcing the critical importance of
precision and sensitivity of deposition and clearance determination in assessing clinical
equivalence for in vitro models. Olfactory nerve filaments extend through the olfactory cleft
and, variably and with less density, into anterior and posterior parts of the middle turbinate.89
Trigeminal nerve fibers, which interact with olfactory fibers in a complex way and contribute to
the “common chemical sense,” also modulate olfactory receptor activity and are involved in
reflexes to minimize exposure to potentially noxious substances by altering nasal patency and
airflow and by changing properties of the mucus blanket.90 Loss of trigeminal function, in
addition to olfactory nerve function, may underlie the loss of sense of smell.91 These
considerations further highlight the importance of an accurate (i.e., precise and specific)
understanding of drug deposition and movement in the nasal cavity after delivery from different
devices, something which cannot currently be achieved in vitro or in silico, particularly in
consideration of the degree of population heterogeneity which must be considered.

An in silico model for an exhalation delivery system would present particular

complexities. There are a number of factors related to the device that could influence the
velocity and volume of air delivered, including, for example, the efficiency of the device in
preventing leakage of air before reaching the nasal cavity and whether or not a sealing nosepiece
is employed. A key challenge for CFD modeling is that there is no definitive way of knowing how
these factors impact deposition in the nose itself. Additionally, a model would have to take into
consideration the pressure transfer that occurs with the XHANCE EDS and the efficiency of the
device in the transfer of pressure, as well as how a change to the shape or form of the nosepiece
(and its relationship to the mouthpiece) in a test product would affect placement, the geometry
and dimensions of the nasal tissues adjacent to the tip of the nostril, “trap conditions” for
movement of drug along mucosal surfaces under conditions of continuing airflow (variable
length and strength of breath and variable number of actuations), changes in droplet size
distribution due to two-way interactions with the intranasal airflow after exiting the device, and
the intranasal circuit resistance to airflow, among other factors.

4. A non-inferiority clinical endpoint study is necessary to demonstrate

bioequivalence and therapeutic equivalence to XHANCE.

No existing in vitro method or any of the emerging technologies described above can
adequately assess the drug deposition that will result from a proposed generic product’s device
and the interaction of that device with the patient’s anatomy and physiology. As discussed,

89Féron 1998, supra note 79 (Exhibit 69); D. Leopold, T. Hummel, J.E. Schwobet al. Anterior
Distribution of Human Olfactory Epithelium. Laryngoscope. 2000 Mar. ;110(3 Pt 1):417–421 (attached
hereto as Exhibit 77).
90 T. Hummel, A. Livermore. Intranasal Chemosensory Function of the Trigeminal Nerve and Aspects of
its Relation to Olfaction. Int’l Arch. Occup. Environ. Health. 2002 June; 75(5): 305–313 (attached hereto
as Exhibit 78).
A. Husner, J. Frasnelli, A. Welge-Lüssen et al. Loss of Trigeminal Sensitivity Reduces Olfactory
91

Function. Laryngoscope. 2006 Aug.; 116:1520–1522 (attached hereto as Exhibit 79).

28
those approaches lack sufficient precision, accuracy, reproducibility, and validity across a
sufficient range of patient-related and disease-related variation to establish equivalence of nasal
deposition at the small and anatomically specific sites that are most relevant in this disease.
Given the absence of any suitable alternatives, a non-inferiority clinical endpoint study is
necessary to demonstrate bioequivalence and support therapeutic equivalence to XHANCE.

The site of deposition is expected to be central to the safety and efficacy of a proposed
generic product. Because fluticasone propionate is a locally acting medication, deposition on
the nasal polyps themselves is required for them to be reduced and the magnitude of polyp
reduction will be dependent upon the amount of steroid that can reach the tissue, the polyp
tissue surface area that is coated, and the amount of drug that continues to reach the tissue as it
shrinks. Importantly, the same is true for mucosal tissues swollen due to inflammation related
to the underlying nasal polyp disease process.

The ability to reach these target regions is essential even when treating patients who
have extensive polyp tissue that extends inferiorly or anteriorly to regions accessed by any nasal
delivery system. This is because the mechanism of XHANCE EDS delivery not only deposits
drug on large polyps, but introduces the potential for the deposition profile to change as the
patient anatomy changes, enabling drug to reach increasingly posteriorly (up to and including
the middle meatus, the typical site of origin) as polyps shrink during treatment. This is
consistent with the observation in pivotal trials of patients taking XHANCE in whom all polyp
tissue inside the middle meatus was resolved during treatment. Products which are different in
ways that have the potential to alter deposition cannot be assumed to produce the same clinical
outcomes in all patients with the disease throughout the course of therapy. Both the physical
presence of the polyps and the inflamed mucosa in the upper regions of the nasal cavity
contribute to symptoms experienced by a patient. It therefore is inadequate, especially when
comparing exhalation delivery system devices, to simply rely on crude segmentation, such as the
proportion of drug that reaches beyond a single region of interest in the nasal passages (for
example, an arbitrary segmentation between “anterior” and “posterior” at a vertical nasal valve)
to determine equivalent deposition. Exactly where the nasal fluticasone suspension is deposited
has obvious potential to influence the efficacy of the product. There also is potential for regional
differences to influence safety, as, for example, might be considered with regard to the amount
of steroid deposited in regions near the eye (contributing to existing concerns over risk for
glaucoma or cataract), deposition at the nasolacrimal duct, or deposition in the olfactory cleft or
sphenopalatine ganglion, which may introduce different risks to nerve function. Importantly,
accurate assessments are not available to establish which specific aspects of device design
contribute in what way to the particular pattern of deposition produced by the XHANCE EDS.

As noted, the XHANCE EDS also offers potential therapeutic effects, the direct effects of
which are not well characterized, that are independent of and/or complementary to the
therapeutic effect of the drug being delivered.92 These effects would have to be replicated by an
alternatively designed device in order for a proposed generic product to be “expected to have the
same clinical effect and safety profile” as XHANCE, and, therefore, to achieve therapeutic
equivalence to XHANCE.93

Based on these considerations, the only valid approach to ensure bioequivalence and
therapeutic equivalence between a proposed generic product and XHANCE is through a

92 See note 7, supra.

93 21 C.F.R. § 314.3(b) (defining “therapeutic equivalents”).

29
comparative clinical endpoint study that demonstrates non-inferiority of safety and efficacy in
patients with nasal polyps.

There are two particularly important considerations for the design of such a comparative
clinical endpoint study. First, any such study should be required to demonstrate non-inferiority
on the same co-primary endpoints—improvement in nasal congestion and improvement in nasal
polyp grade—as evaluated in XHANCE’s pivotal studies. There is a risk of non-equivalence
being missed (i.e., Type 1 error) if reduction in polyp grade is not observed starting from the
same baseline degree of disease, since topical drug will differently reach polyps extending to
different parts of the nasal cavity. Additionally, given the fact that most nasal polyps originate
in the middle meatus, the non-inferiority of a test product on polyp grade should be
demonstrated not only for mean change in this measure from a comparable baseline, but also
for the proportion of subjects who achieve a polyp score of 0=no polyps, demonstrating
comparable activity in the middle meatus. In pivotal studies with XHANCE, it was shown that
between 14-20% of patients treated with the high dose of XHANCE had polyps eliminated
(grade=0; not observed in the middle meatus or elsewhere) in at least one nostril after 16 weeks
and up to 28% after 24 weeks.94 This likely reflects the particular intranasal deposition
associated with XHANCE that enables continued deposition of topical steroid on the polyp
tissue as it shrinks back to its site of origin. Although this activity through different stages of
treatment is not necessary for all drugs that treat the disease, equivalent intranasal deposition is
necessary to produce a therapeutically equivalent effect; therefore, any proposed generic version
of XHANCE should be able to produce an equal rate of polyp elimination to ensure
bioequivalence and therapeutic equivalence.

Second, the subjects included in the study need to be representative of the population in
which XHANCE has been demonstrated to be effective in order to accurately assess whether the
test product has equivalent effects on shrinking nasal polyps. The need for representativeness
applies to both polyps and patients. Regarding polyps, in the XHANCE pivotal studies, subjects
with grade 1-3 polyps were enrolled. Larger polyps extend further down into the nasal cavity,
and therefore are more easily reached by products that produce limited deposition in more
difficult-to-access regions. Smaller polyps (grades 1 and 2) do not reach much further than the
middle turbinate, and the amount of drug deposition in more superior/posterior regions will
have a substantial impact on the magnitude of polyp reduction in the population of patients with
that grade of disease. Therefore, the inclusion criteria for the clinical study should allow for a
sufficient number of subjects with both large and small polyps. Regarding patients, given the
important patient-based population variations that interact with the design and functioning of
an exhalation delivery system device in domains including not only the extent of disease, but
also behavior, intranasal anatomy, and mid-facial morphology (sometimes but not always linked
to race/ethnicity/age/gender), even minor changes in the design or handling of the delivery
device require validation of equivalence of clinical performance in a suitably diverse population.

5. A pharmacokinetic study is a necessary element of an ANDA product’s

demonstration of bioequivalence and therapeutic equivalence to
XHANCE.

Additionally, a demonstration of bioequivalence and therapeutic equivalence to

XHANCE should include a pharmacokinetic study to assess systemic absorption, consistent with

94 Leopold 2019, supra note 60 (Exhibit 52); Sindwani 2019, supra note 60 (Exhibit 53).

30
FDA’s recommendations in the Nasal Spray Draft Guidance.95 Requiring an ANDA applicant to
conduct a pharmacokinetic study also is consistent with FDA’s precedents. For example, the
applicant for the generic mometasone furoate nasal spray conducted a pharmacokinetic study
and obtained measurable plasma concentrations at a therapeutic dose.96 The general draft
product-specific guidance for fluticasone propionate nasal spray also recommends a single-dose,
two-way crossover pharmacokinetic study.97 Accordingly, FDA should require a
pharmacokinetic study to ensure equivalent systemic absorption between XHANCE and a
proposed generic drug product.

6. An ANDA formulation product should be Q1/Q2 the same as XHANCE.

FDA also should require an ANDA product formulation to be qualitatively (Q1) and
quantitatively (Q2) the same as XHANCE. This reflects a recommendation in FDA’s Nasal Spray
Draft Guidance and also aligns with FDA’s recommendations for generic products referencing
FLONASE.98

7. A generic device should have equivalent design and performance

characteristics to the XHANCE EDS and should not include differences in
user interface that would preclude approval in an ANDA.

Finally, as part of the required showing of bioequivalence and therapeutic equivalence,

FDA should require that a generic device have equivalent design and performance
characteristics to those of the XHANCE EDS. The elements of the XHANCE EDS that are
believed to play the greatest role in XHANCE’s safety and efficacy are described above in
sections B.II.A and B.III.A.1.c-d. Additionally, FDA should require an ANDA applicant to
analyze any differences in user interface and show that such differences do not prevent the
proposed generic combination product from being substituted for XHANCE without the
intervention of a health care provider and/or without additional training prior to use. These
requirements are essential given the central role played by the device constituent in the safety
and efficacy of XHANCE.

FDA included similar recommendations in the general draft product-specific guidance

for fluticasone propionate nasal spray and in the Nasal Spray Draft Guidance. First, in the
general draft PSG for products referencing FLONASE, FDA recommended that prospective
applicants consider the following characteristics of the reference (“R”) product in designing the
test product: external operating principles and external critical design attributes of the R
product, size and shape of the R product and number of doses in the R product99

FDA further recommended that prospective applicants refer to the draft guidance
entitled “Comparative Analyses and Related Comparative Use Human Factors Studies for a

95 Nasal Spray Draft Guidance, supra note 23, at 7, 24–26 (Exhibit 17).
96 Liu 2019, supra note 27 (Exhibit 18).
97 FLONASE Draft Guidance, supra note 28, at 4–5 (Exhibit 19).
98Nasal Spray Draft Guidance, supra note 23, at 6–7 (Exhibit 17); FLONASE Draft Guidance, supra note
28, at 8 (Exhibit 19).
99 FLONASE Draft Guidance, supra note 28, at 9 (Exhibit 19).

31
Drug-Device Combination Product Submitted in an ANDA.”100 This draft guidance recommends
that an ANDA applicant “analyze the overall user interface of a proposed generic combination
product to identify differences in design when compared to the RLD,” including through
comparative use human factors studies if the differences “may not be minor.”101 In assessing
whether such design differences are acceptable, FDA “intends to consider whether the generic
combination product can be substituted for the RLD without the intervention of a health care
provider and/or without additional training prior to use of the generic combination product.”102

Similarly, in FDA’s Nasal Spray Draft Guidance, FDA recommended that the container
and closure system of the test product formulation be comparable to that of the reference listed
drug, and that an applicant include a side-by-side comparison of the components of the
container and closure system, listing brand and model, dimensions of critical components, and
engineering drawings, if possible.103

Although these analyses are an essential part of an ANDA applicant’s showing of

bioequivalence and therapeutic equivalence, they are insufficient, without the additional data
described in this petition, to establish the equivalence of a proposed product. This is because it
is not possible to accurately establish, particularly in light of patient variability, how each design
feature of the XHANCE EDS, individually or collectively, contributes to the drug deposition
achieved by XHANCE, including through the interaction between the XHANCE EDS and the
nose and nasal cavity and mouth and oral cavity, and the patient’s physiology, and the product’s
resulting clinical effect.

B. FDA should issue a draft product-specific guidance for ANDAs

referencing XHANCE that is consistent with these requests.

In 2020, FDA added XHANCE to the Agency’s list of planned new product-specific
guidance documents for complex generic drug products.104 To provide clarity to ANDA
applicants and other stakeholders on FDA’s current thinking on the appropriate methods for
showing bioequivalence and therapeutic equivalence to XHANCE, OptiNose requests that FDA
issue a draft product-specific guidance for ANDAs referencing XHANCE that aligns with these
requests.

100Id.; FDA, Draft Guidance for Industry, Comparative Analyses and Related Comparative Use Human
Factors Studies for a Drug-Device Combination Product Submitted in an ANDA (Jan. 2017)
(“Comparative Analyses Draft Guidance”) (attached hereto as Exhibit 80).
101 Comparative Analyses Draft Guidance, supra note 100, at 5, 8 (Exhibit 80).
102 Id. at 4.
103 Nasal Spray Draft Guidance, supra note 23, at 7 (Exhibit 17).
104See FDA, Upcoming Product-Specific Guidances for Complex Generic Drug Product Development
(current as of Mar. 2,
2020),https://web.archive.org/web/20200516235430/https://www.fda.gov/drugs/guidances-
drugs/upcoming-product-specific-guidances-complex-generic-drug-product-development (attached
hereto as Exhibit 81).

32
 C. FDA should require the same type of non-inferiority clinical endpoint
study for a section 505(b)(2) product to receive an “A” rating.

Generally, a section 505(b)(2) product may be approved with a range of differences from
the listed drug relied upon. For example, a section 505(b)(2) product is unlikely to be Q1/Q2
the same as XHANCE and also can be expected to differ from XHANCE in the device design.
The same type of non-inferiority in vivo clinical endpoint study described above should be
required to show the bioequivalence and therapeutic equivalence of a section 505(b)(2) product
if the applicant seeks a determination that the product is therapeutically equivalent to XHANCE
(i.e., an “A” rating in the Orange Book).

IV. Conclusion

For the foregoing reasons, OptiNose requests that FDA not approve an ANDA
referencing XHANCE unless the applicant demonstrates bioequivalence and therapeutic
equivalence through the type of non-inferiority clinical endpoint study described above in
addition to in vitro studies, pharmacokinetic study data, and an adequate showing of device
equivalence. Additionally, OptiNose requests that FDA issues a draft product-specific guidance
document that is consistent with this approach. Finally, OptiNose asks that FDA require a
section 505(b)(2) applicant to demonstrate bioequivalence and therapeutic equivalence through
the same type of non-inferiority clinical endpoint study in order to obtain an “A” rating in the
Orange Book.

C. Environmental Impact
The actions requested herein are subject to categorical exclusion under 21 C.F.R.
§§ 25.30 and 25.31.

D. Economic Impact
Pursuant to 21 C.F.R. § 10.30(b), an economic impact statement will be submitted only
upon the request of the Commissioner.

33