in s v s

T H E R A P E U T I C S , IN C .

200 3 17 NGV-a A9 '-,9

October 27, 2017

Dockets Management Branch

Food and Drug Administration
Department of Health and Human Services
5630 Fishers Lane. Room 1061 (HFA-305)
Rockville, Maryland 20852

CITIZEN PETITION

Insys Therapeutics, Inc. (‘insys”) submits this petition under sections 505(j) and
505(q) of the Federal Food, Drug, and Cosmetic Act ("FDC Act5') and the related
regulations, 21 C.F.R. §§ 10.30-31, to request that the Commissioner of Food and
Drugs refrain from approving any Abbreviated New Drug Application (“ANDA'')
referencing its SYNDROS %(dronabinol) Oral Solution, CII approved under New Drug
Application ("NDA”) 205525, if such ANDA relies upon a waiver pursuant to 21
C.F.R. § 320.22. Insys therefore requests that the Food and Drug Administration
(“FDA”) require the submission of in vivo bioequivalence data for approval of any
generic versions of dronabinol oral solution. Insys is aware of at least two received
ANDA referencing SYNDROS and requests that FDA retroactively refuse to accept
this ANDA, as well as any other ANDA that may have been received, or refuse to
approve any ANDA unless it includes adequate in vivo bioequivalence studies.

I. ACTIONS REQUESTED

Insys requests that FDA:

(1) Decline to receive or approve any ANDA application for generic

dronabinol oral solution that relies on SYNDROS as the Reference
Listed Drug (•“RLD”) if the ANDA relies on a waiver in lieu of
establishing in vivo bioequivalence to SYNDROS.
(2) Require that ANDA applicants for generic versions of SYNDROS
include fed and fasted state bioequivalence studies.

Insys is concerned that FDA has permitted and may continue to permit waivers
of bioequivalence testing for products referencing its NDA. SYNDROS has inherent
product characteristics that may affect bioequivalence. To ensure required product

1333 South Spectrum Boulevard, Suite 100, Chandler, A Z 85286 / phone: 480.500.3127 / fax: 602.910.2627 / www.insysrx.com

^ 77 /
safety and efficacy, Insys requests that FDA require generic applicants to conduct and
submit in their ANDAs bioequivalence studies in both the fed and fasted state.

II. STATEMENT OF GROUNDS

A. Factual Background

Dronabinol is a cannabinoid indicated for the treatment of anorexia associated

with weight loss in patients with AIDS and nausea and vomiting associated with cancer
chemotherapy in patients who have failed to respond adequately to conventional
antiemetic treatments. First approved in oral capsule form in 1985 as Marinol, it has
been available as a generic in capsule form since 2008. Insys began marketing a
generic capsule form in 2011, but has subsequently discontinued the product.

Insys received marketing approval for an oral solution version of dronabinol in

NDA 205525 on July 1, 2016; however, at the request of FDA, Insys agreed not to
market SYNDROS until a Drug Enforcement Administration (“DEA”) scheduling
order had been issued. Insys therefore did not market SYNDROS until July 31, 2017 -
several months after the DEA's scheduling order issued on March 23, 2017. Pursuant
to FDC Act § 505(x), the date of approval is considered the date of issuance of the
interim final rule controlling the drug; therefore, the date of SYNDROS approval is
March 23,2017.

In June 2017, Insys received a Paragraph IV certification from Par

Pharmaceuticals (“Par’') notifying Insys of Par’s submission of ANDA 210314
referencing NDA 205525. Insys promptly initiated litigation in August 2017 in the
District Court for the District of Delaware, which triggered a 30-month stay during
which FDA cannot approve Par’s ANDA. Similarly, on August 31, 2017, Insys
received a Paragraph IV certification from Alkem Laboratories ("Alkem") notifying
Insys of Alkem's submission of its ANDA. Insys again promptly initiated litigation
within 45 days. With the filing of such litigation, the existence of Par's and Alkem’s
ANDAs became public knowledge.

Insys began distributing SYNDROS on July 31, 2017: no versions of

SYNDROS were in distribution and available for bioequivalence testing prior to July
31, 2017. Insys therefore believes that Par's ANDA is based on a bioequivalence
waiver under 21 C.F.R. § 320.22.
 Insys notes that it is currently awaiting an FDA exclusivity determination with
respect to SYNDROS and expects to receive three years of exclusivity based on the
submission of new clinical studies essential to approval.

B. Legal Background

Under the FDC Act, an ANDA must refer to an RLD and must contain, among
other things, “information to show that the new drug is bioequivalent to the listed
drug.” 21 U.S.C. § 355(j)(2)(A)(iv). FDA regulations provide that any person
submitting an ANDA to the FDA must include either "evidence demonstrating that the
drug product that is the subject of the abbreviated new drug application is bioequivalent
to the [RLD]” or enough information showing bioequivalence to the RLD to permit
FDA to waive the submission of evidence demonstrating in vivo bioequivalence. 21
C.F.R. § 320.21(b). Two drug products will be considered bioequivalent if there is no
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 of active moiety under similar experimental conditions. Id. § 314.3.

Evidence demonstrating bioequivalence “shall be obtained using one of the

approaches for determining bioavailability set forth" in 21 C.F.R. § 320.24. Id.
§ 320.21(e). In certain circumstances, FDA may waive the requirement for in vivo
bioequivalence testing under 21 C.F.R. § 320.22 when in vivo bioavailability or
bioequivalence may be considered self-evident based on other data in the application.
For an oral solution, this includes products that contain an active drug ingredient in the
same concentration and dosage form as the RLD and contains no inactive ingredient or
other change in the formulation that may significantly affect absorption or availability
of the product. Id. § 320.22(b)(3).

III. DISCUSSION

Insys requests that FDA require Par and any other ANDA applicant to submit
studies demonstrating that the potential generic is bioequivalent to Insys’ SYNDROS.
Because of the complexity of the product, as well as the directions for use with food as
outlined below in subsections A and B, Insys contends that a waiver of bioequivalence
studies under 21 C.F.R. § 320.22 is inappropriate for this drug. FDA itself has made
this determination with respect to capsule versions of the product. Draft Guidance on
Dronabinol (Feb. 2014); Insys therefore requests that FDA exercise its authority under
21 C.F.R. § 320.22(f) to require the submission of evidence of bioequivalence for oral
solution versions of SYNDROS.
 A. The structure and profile of SYNDROS indicate that any
formulation changes require bioequivalence studies.

While FDA has authority to grant waivers if the waiver is ‘'compatible with the
protection of human health." dronabinol oral solution does not meet the limited criteria
that FDA cites as most appropriate for waivers. Waivers of in vivo bioequivalence
studies are typically reserved for limited instances. As noted, for a generic version of
an oral solution, FDA may grant a waiver only if it makes a preliminary assessment
that the product contains no change in formulation that may significantly affect
absorption of the active drug or no change that may significantly affect systemic
availability.

FDA guidance and precedent seems to suggest that a drug with high
permeability is likely to have high bioavailability. See e.g.U.S. Food and Drug Admin.,
Waiver of In Vivo Bioavailability and Bioequivalence Studies for Immediate-Release
Solid Oral Dosage Forms Based on a Biopharmaceutics Classification System: Draft
Guidance for Industry (May 2015). Thus, high permeability, in conjunction with high
solubility and rapid dissolution, can be used as a predictor for absorption. This
assumption, however, does not hold true for dronabinol. The mean oral bioavailability
is 10-20 percent for dronabinol and its active metabolite, which is indicative of a
significant complexity of absorption, distribution, metabolism, and elimination profile
in vivo. AbbVie, Marinolf (dronabinol) Capsules Package Insert. NDA 018651
(revised Aug. 2017).

Because SYNDROS in its oral form presents a complex absorption and

bioavailability profile, changes to the formulation of the product risks bioequivalence
differences. Dronabinol exhibits (intra-subject) variability of 30% or more in two
bioequivalence parameters Cmax and AUC. Advisory Committee for Pharmaceutical
Sciences Meeting Transcript. Why Bioequivalence of Highly Variable Drugs is an
Issue (Apr. 14, 2004); Kamal Midha et al.. The bioequivalence of highly variable
drugs and drug products, 43 Int. J. Clin. Pharmacol. Ther. 485 (2005). Potential
limitations of prior dronabinol formulations include variable bioavailability and
variable absorption, which are important considerations that limit the optimal treatment
of patients. Consequently, intra-subject and inter-subject variabilities are likely to have
important implications for safety and efficacy with dronabinol products, as well as
patient adherence. To explore the possibility of the drug product formulation
contributing to high variability. Davit BM, et.al., investigated relation of dissolution
performance variability to bioequivalence parameter variability by comparing rank
orders of dissolution profile % CV values with rank orders of root-mean-square error
(RSME) values within each inconsistently highly variable drug product line. Barbara
Davit et al.. Highly Variable Drugs: Observations from Bioequivalence data submitted
to the FDA for new generic drug applications, 10 AAPS J. 1481 (March 2008). Results
showed that out of 22 inconsistently highly variable drugs, 11 drugs showed the highest
variability in dissolution profiles correlated with that of the highest RSME. There was
no correspondence between dissolution profiles correlated with that of highest RSME
for other 11 drugs. These results suggest variations in drug formulation could be
contributing to the high variability.

As dronabinol is a highly variable drug, small changes in drug formulation in

relation to excipients can result in changes in bioequivalence parameters. Accordingly,
an ANDA sponsor should be required to conduct traditional bioequivalence studies
comparing its generic formulation and SYNDROS to demonstrate that excipient
changes do not affect bioequivalence. Not all of the excipients in approved drug
formulations are “inactive;" some are known to alter drug metabolism and/or
transporter activities. Excipients may alter systemic exposure of the drug itself or other
potentially co-administered therapeutic agents. Inactive ingredients may alter drug
absorption and may change the properties of pH, viscosity, drug permeation, residence
time, as well as drugs onset of action. Changes in excipients might lead to dose
accumulations or slower elimination rates of the drug leading to safety and efficacy
issues. Literature suggests that Ethanol, Polyethylene glycol 400 (PEG 400), and
Propylene Glycol may all have these types of effects on dronabinol. See Jonas H.
Fagerberg, et al.. Concomitant intake of alcohol may increase the absorption of poorly
soluble drugs, 67 Eur. J. of Pharma. Sci. 12 (Jan. 25, 2015); Stefanie Bendels, et al.,
PAMPA - Excipient Classification Gradient Map, 23 Pharm Res 2525 (2006).

Sprague and Craigmill have reported that two significant vehicle effects on the
absorption of THC (dronabinol) were observed. It was shown that treatment of mice
twice daily for 11 days with 10% propylene glycol (PG) + 1% Tween-80 in isotonic
saline impaired the elimination of a subsequently administered dose of either ethanol or
THC, because the primary route of metabolism of propylene glycol involves the
enzymes similar to those suggested to metabolize ethanol and THC. It was also found
that (2) either 6 or 11-day treatment with the propylene glycol + Tween-80 mixture
diminished the observed behavioral effect resulting from THC administration. Gary
Sprague et al.. Ethanol and delta-9-tetrahydrocannabinol: Mechanism for cross­
tolerance in mice, 5 Pharmacology Biochemistry & Behavior 409 (1976).
 Par's product contains ethanol and propylene glycol amounts that may lead to
variable exposure of dronabinol in patients. Therefore, in accordance with 21 C.F.R.
§ 320.22(b)(3), FDA should require bioequivalence testing to show that exposure and
absorption remains the same amongst products and that changes to excipients do not
lead to dose accumulation or any additional safety and efficacy risks.

Finally, it was observed from the Insys in vitro study that SYNDROS forms a
turbid nanoemulsion in both the fed state and fasted state in simulated gastric/intestinal
fluids. Measurement of the particle size of this nanoemulsion showed the particle size
range between 700 nm to 950 nm. It is well known that any change in the particle size
affects the rate and extent of drug absorption. Additionally, any changes to the
formulation, for example the excipients, may affeet the formation of nanosuspension,
which in turn, affects the absorption profile of dronabinol. Therefore, changes to a
formulation that affects particle size may affect bioequivalence. It is thus necessary for
any generic formulations to include traditional bioequivalence studies.

B. FDA policy requires fed and fasted state BE studies for a product
like SYNDROS.

The FDA guidance “Food-Effect Bioavailability and Fed Bioequivalence

Studies” explains that administration of a drug product with food may change the
bioavailability of a product by affecting either the drug substance or the drug product.
FDA Guidance (Dec. 2002). Food effects are most likely to result from a more
complex combination of factors that influence the in vivo dissolution of the drug
product and/or the absorption of the drug substance. In these cases, the relative
direction and magnitude of food effects on formulation bioavailability and the effects
on the demonstration of bioequivalence may be difficult to predict without conducting
a fed and fasted bioequivalence study. For this reason, FDA guidelines state:

In addition to a BE study under fasting conditions, we recommend a BE

study under fed conditions for all orally administered immediate-release
drug products, with the following exceptions:

When both test product and RED are rapidly dissolving, have
similar dissolution profiles, and contain a drug substance with
high solubility and high permeability (BCS Class I), or
11. When the DOSAGE AND ADMINISTRATION section of the
RED label states that the product should be taken only on an
empty stomach, or
 iii. When the RLD label does not make any statements about the
effect of food on absorption or administration.

Id. at 3-4.

Based on these criteria, FDA should ensure the ANDA applicants demonstrate
bioequivalence under both fasted and fed conditions. The SYNDROS label makes
explicit statements about the effect of food on absorption or administration. The
approved labeling states: “Administer the first dose on an empty stomach at least 30
minutes prior to eating,’' signifying the effect of food on SYNDROS (dronabinol) oral
solution.

Further, SYNDROS is categorically not a BCS Class I drug. In FDA draft

guidance “Waiver of In Vivo Bioavailability and Bioequivalence Studies for
Immediate-Release Solid Oral Dosage Forms Based on a Biopharmaceutics
Classification System Guidance for Industry,” FDA explains that BCS Class I drugs are
solid oral dosage forms that are highly soluble and rapidly dissolvable. Dronabinol is
neither a solid oral dosage form nor is it highly soluble, and FDA has clearly indicated
that waivers are inappropriate for the oral capsule version of this product. U.S. Food
and Drug Admin: Draft Guidance on Dronabinol (Feb. 2014). Therefore, waivers
based on BCS are not applicable here.

Food also appears to decrease the bioavailability of SYNDROS compared to the

fasted state. As shown in Study INS004-15-059, the extent of oral absorption in the fed
state may be considerably less than in the fasted state. There, the administration of
4.25 mg oral solution after a high fat meal resulted in delayed rate and extent of
absorption relative to fasted state; food resulted in 2.5-fold increase in total body
exposure and approximately 4- to 7-hour delay in T max. Additionally, changes in the
formulation composition, qualitatively or quantitatively, may have a significant impact
on the bioavailability of the drug in the presence or absence of food. This suggests that
any generic product of SYNDROS that varies in formulation composition should not
quality for a waiver.

In fact, dronabinol has a variable pharmacokinetic profile. Insys studies

demonstrated an appreciable food effect on dronabinol oral solution. As a part of the
development program and as recommended by FDA, Insys conducted a bioavailability
study to evaluate the effect of food. During its investigation of the pharmacokinetics of
SYNDROS, Insys conducted two studies to examine comparative bioavailability of
SYNDROS in the fasted and fed states. The first study was a single dose, randomized
crossover with 4.25mg of SYNDROS under fasting conditions. In this study, in the
fasted state, dronabinol oral solution had a C max of 1.81 ng/mL, the T max and extent of
absorption AUCo-t is 3.44, and AUCo-^^ of 3.75 (h.ng/mL). The second study was a
randomized single-dose three period crossover study in which 54 subjects were given
4.25 mg dronabinol oral solution under fed (standardized high fat meal) conditions.

Absorption of dronabinol from dronabinol oral solution after a high-fat meal

resulted in 2.6-fold increase in mean total exposure AUCo-t, increase in 2.7-fold for
AUCo-x compared to fasted state, approximately a 6.7-hour delay in median TmaX, and a
22% decrease in Cmax compared to the fasted state. In addition, under the fed
conditions, the mean Cmax values of 1l-OH-delta-9-THC was 1.23 ng/mL for
dronabinol oral solution 4.25 mg indicating a 43% lower Cmax with dronabinol oral
solution 4.25 mg. Mean 1l-OH-deIta-9-THC total exposure (AUCO-t and AUCO-inO
was about 15% lower with dronabinol oral solution 4.25 mg under fed conditions.

FDA itself has recognized the need for fed and fasting studies in dronabinol. In
the Draft Guidance on Dronabinol, FDA explicitly requires this testing for dronabinol
oral capsules. Draft Guidance on Dronabinol (Feb. 2014). Nothing about the structure
or properties of the oral solution indieates that the same considerations need not apply
in the oral solution.

C. Metabolite Bioequivalence should also be measured.

Under FDA's Guidance. Bioavailability and Bioequivalence Studies for Orally

Administered Drug Products — General Considerations, metabolite bioequivalence, in
addition to parent drug bioequivalence, should be measured when the metabolite
eontributes meaningfully to safety and/or efficacy of a product. FDA Guidance, 18
(Mar. 2003). Here, dronabinol has been shown to undergo extensive first-pass hepatic
metabolism by the presystemic gut wall yielding the major metabolite 11-OH-A9-THC,
which contributes meaningfully to safety and efficaey of the active ingredient. For
example, after an intravenous administration of 11-OH-A9-THC (1 mg) to the human
subjects, there were marked pharmacologic and psychologic effects. All subjects
responded to the effect of this drug w ithin 2-3 min. There was an increase in heart rate
ranging from 20 to 60 beats/min. Louis Lemberger et al.. Comparative Pharmacology
of A9 Tetrahydrocannabinol and its Metabolite, 11-OH-A9- Tetrahydrocannabinol, 52
J. of Clin. Investigation 2411 (Oet. 1973). Additionally, changes to the excipients in a
dronabinol product may have an effect on the hepatic metabolism of dronabinol.
Therefore, a generic formulation of the product, including formulations w ith changes in
excipients, should be required to demonstrate metabolite bioequivalence, as well as
parent drug bioequivalence. In its Draft Guidance on Dronabinol, FDA requires
metabolite bioequivalence testing for oral capsules; again, there is no reason this same
requirement should not apply to oral solutions. See Draft Guidance on Dronabinol
(Feb. 2014).
IV. CONCLUSION

SYNDROS offers medically significant benefits for the management of anorexia

associated with weight loss in patients with AIDS and nausea and vomiting associated
with cancer chemotherapy in patients who have failed to respond adequately to
conventional antiemetic treatments. As demonstrated above, FDA should not waive
bioequivalence testing for generic versions of the product. In the absence of rigorous
requirements for approval, a proposed generic dronabinol oral solution may fail to
demonstrate bioequivalence to SYNDROS putting cancer and AIDS patients at risk of
receiving an unsafe or ineffective product.

To ensure clinical equivalency to SYNDROS, Insys requests that any ANDA

applicant using SYNDROS as the RLD be required to establish in vivo bioequivalence
under both fasted and fed states. Changes in formulation composition can markedly
impact the bioequivalence of a drug product, making it imperative that generic
applicants of equivalent formulations of SYNDROS be required to conduct the
bioequivalence trials proposed above.

For these reasons, Insys respectfully requests that FDA refuse to approve Par’s
ANDA and refuse to receive or approve any other ANDA for generic versions of
SYNDROS based on waivers of bioequivalence.

V. ENVIRONMENTAL IMPACT

This petition is categorically exempt from the requirement for an environmental

assessment or an environmental impact statement pursuant to 21 C.F.R. §§ 25.30-31.

VI. ECONOMIC IMPACT

Information on the economic impact of the petition will be provided upon

request.
 Vll. CERTIFICATION

I certify that, to my best knowledge and belief; (a) this petition includes all
information and views upon which the petition relies; (b) this petition includes
representative data and/or information known to the petitioner which are unfavorable to
the petition; and (c) 1 have taken reasonable steps to ensure that any representative data
and/or information which are unfavorable to the petition were disclosed to me. I further
certify that the information upon which I have based the action requested herein first
became known to the party on w'hose behalf this petition is submitted on or about the
following date: June 22, 2017. If I received or expeet to receive payments, including
cash and other forms of consideration, to file this information or its contents, I received
or expect to receive those payments from the following persons or organizations: Insys
Therapeutics Inc. I verify under penalty of perjury that the foregoing is true and correct
as of the date of the submission of this petition.

Sincerely,

Stephen Sherman
Sr. Vice President, Regulatory Affairs
Insys Therapeutics. Inc.

10
 20 .
rr

ri
ORIGIN IDCHDA (480) 500-3147 SHIP DATE: 07NOV17
r-
LISA THELEN ACTWGT: 0.50 LB o
rj
INSYS THERAPELITICS CAD: 110607081/INET3920
1333 S SPECTRUM BLVD STE 100

CHANDLER AZ 85286 BILL SENDER

UNITED STATES US

TO LISAM ELEN
FDA/ DEPT. OF HEALTH & HUMAN SERVIC
5630 FISHERS LN, ROOM 1061(HFS-305)
DOCKETS MANAGEMENT BRANCH
ROCKVILLE MD 20852
( 00 0 ) 000-0000

0 (0 (1)

W ED-08 NOV 10:30A

PRIORITY OVERNIGHT
7706 9229 4686
c 0)
20852 3 O

M D -U S lAD o c >*

= C w
Q. O

nj o

3 0)