J Thromb Thrombolysis
`DO[ 10. |00'.7/sl I239-0l3-1035-4
`
`Impact of concomitant low-dose aspirin on the safety
`and tolerability of naproxen and esomeprazole magnesium
`delayed-release tablets in patients requiring chronic nonsteroidal
`anti-inflammatory drug therapy: an analysis from 5 Phase III
`studies
`
`Dominick J. Angiolillo ' Catherine Datto '
`Shane Raines - Neville D. Yeomans
`
`© Springer Science+Business Media New York 2013
`
`Abstract Patients receiving chronic nonsteroidal anti-
`inflammatory drugs (NSAIDS) and concomitant low-dose
`aspirin (LDA) are at increased risk of gastrointestinal (GI)
`toxicity. A fixed—dose combination of enteric-coated (EC)
`naproxen and iininediate-release esomeprazole magnesium
`(NAP/ESO) has been designed to deliver a proton-pump
`inhibitor followed by an NSAID in a single tablet. To
`examine safety data from 5 Phase III studies of NAP/ESO
`in LDA users (5325 mg daily, administered at any time
`during the study), and LDA non-users, data were analyzed
`from 6-month studies assessing NAPIESO versus EC
`naproxen in patients with osteoarthritis,
`rheumatoid
`arthritis, or ankylosing spondylitis (ll = 2), 3-month stud-
`ies assessing NAPIESO vs celecoxib or placebo in patients
`with knee osteoarthritis (H = 2), and a 12-month, open-
`label, safety study of NAPIESO (H = I). In an analysis of
`
`D. J. Angiolillo {E}
`Division of Cardiology, University of Florida College of
`Medicine-Jacksonville, ACC Building 5th Floor.
`655 West 8th Street. Jacksonville, FL 32209. USA
`e-mail: dominick.angiolillo@jax.ufl.edu
`
`C. Datto - S. Raines
`
`Astrazeneca, [800 Concord Pike, Wilmington, DE [9850 USA
`e—mail: cathetine.datto@astrazeneca.com
`
`S. Raines
`e—n1ai|: shane.raines@astrazeneca.com
`
`N. D. Yeornans
`University of Western Sydney, Sydney. NSW, Australia
`e-Inail: n.yeomans@uws.edu.au
`
`N. D. Yeomans
`
`Office for Research. Austin Hospital. Studley Road. Heidelberg,
`Melbourne. VIC 3084. Australia
`
`Published ortlirtc: 25 December 2013
`
`Page1of13
`
`two studies, incidences of endoscopically confirmed gastric
`ulcers (GUS) and duodenal ulcers (DUs) were summarized
`
`by LDA subgroups. In the pooled analysis from all five
`studies. incidences of treatment-emergent adverse events
`(AE5) {including prespecified NSAID-associated upper GI
`AEs and cardiovascular AEs), serious AEs, and AE-related
`
`discontinuations were stratified by LDA subgroups. Over-
`all, 2,317 patients received treatment;
`1,157 patients
`received NAPIESO and, of these, 298 received LDA. The
`cumulative incidence of GUS and DUs in the two studies
`
`with 6-month follow-up was lower for NAPIESO vs EC
`naproxen in both LDA subgroups [GUs: 3.0 vs 27.9 ‘=79,
`respectively. for LDA users, 6.4 vs 22.4 %, respectively,
`for LDA non-users (both P < 0.001); DUs: 1.0 vs 5.8 %
`for LDA users, 0.6 vs 5.3 % for LDA non—users]. The
`incidence of erosive gastritis was lower in NAPi"ESO- vs
`EC naproxen-treated patients for both LDA users [|8.2 vs
`36.5 %,
`respectively (P : 0.004)] and LDA non-users
`[19.8 vs 38.5 "79, respectively [P -5. 0.001)]. Among LDA
`users, incidences of NSAID-associated upper GI AEs were:
`NAPIESO,
`l6.l %; EC naproxen, 31.7 %; celecoxib,
`22.1 %; placebo, 23.2 %. Among LDA non-users.
`inci-
`dences of NSAID-associated upper GI AEs were: NAP!’
`ESO, 20.3 %; EC naproxen, 36.6 %; celecoxib, 18.5 %;
`placebo, 18.9 %. For LDA users,
`incidences of cardio-
`vascular AES were: NAPIESO, 3.0 %; EC naproxen.
`1.0 %; celecoxib, 0 %; placebo, 0 %. For LDA non—users,
`incidences of cardiovascular 13035 were: NAPIESO, 1.0 %;
`
`EC naproxen, 0.6 %; celecoxib, 0.3 %; placebo, 0 %.
`NAPIESO appears to be well-tolerated in patients receiving
`concomitant LDA. For LDA users, AE incidence was less
`
`than that observed for EC naproxen. For most AE cate-
`gories.
`incidences were similar among NAPIESO, cele-
`coxib and placebo groups. The safety of NAP.-‘ESO
`appeared similar regardless of LDA use.
`
`Q Springer
`Patent Owner Ex. 2003
`CFAD v. Pozen
`|PR2015-01718
`
`
`Page 1 of 13
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`IPR2015-01718
`
`

`
`Keywords NSAID - Low—dose aspirin - Naproxen -
`Naproxenfesomeprazole magnesium - Safety profile »
`Tolerability
`
`Introduction
`
`Nonsteroidal anti-inflammatory drugs (NSAIDS) are com-
`monly used for managing the symptoms of many inflam-
`matory conditions, including osteoarthritis (OA), rheumatoid
`arthritis (RA), and other arthritic conditions. However,
`
`chronic NSAID therapy is associated with an increased risk
`of adverse gastrointestinal (GI) and cardiovascular (CV)
`effects. For instance, chronic NSAID users develop endo-
`scopic gastric ulcers
`(GUs) with point prevalences of
`15-30 % [1], serious ulcer complications occur in about
`2-4 % annually [l—4], and an increased incidence of stroke,
`myocardial infarction (MI), and congestive heart failure has
`also been reported with many NSAIDs [5, 6].
`Among the known risk factors for CV toxicity with
`NSAID treatment are older age, hypertension, and estab-
`lished CV disease [7, 8]. Risk factors for NSAID-associated
`
`GI complications include older age, history of ulcers or upper
`GI (UGI) symptoms, and concomitant use of such medica-
`tions as anticoagulants and low-dose aspirin (LDA) [9, 10].
`Twenty percent of NSAID users are estimated to take con-
`comitant LDA, usually as prophylaxis for CV events [1 1].
`A recommended strategy to prevent higher risk patients
`from developing NSAID-associated ulcers is the concom-
`itant administration of a gastroprotective agent, for exam-
`ple, a proton pump inhibitor (PPI) [2,
`l2—l6]. PPIs have
`also been shown to reduce the risk for GUs, duodenal
`
`ulcers (DUs), and their complications associated with the
`continuous use of LDA [17-"—l9].
`However, despite recommendations from guidelines,
`several studies suggest that. although increasing, use of
`concomitant gastroprotective agents with NSAIDS remains
`low [20—24].
`As a potential solution to the under-use of gastropro-
`tective agents, a fixed—dose combination of enten'c—coated
`(EC)
`naproxen
`500 mg
`and
`immediate-release
`(IR)
`esorneprazole magnesium 20 mg (naproxenfesomeprazole
`magnesium; NAPIESO) has been designed to provide
`sequential delivery of, first, a PP], and then an NSAID from
`a single tablet. Phase III trials have demonstrated compa-
`rable efficacy for NAPIESO and celecoxib in the treatment
`of OA of the knee [25], while NAPIESO was associated
`
`D. J. Angiolillo ct al,
`
`adverse events (AEs) [22]. The NAPIESO combination is
`currently licensed in both the United States and Europe for
`the relief of signs and symptoms of OA, RA, and anky-
`losing spondylitis, and to decrease the risk for developing
`NSAID-associated GUs in at-risk patients [28, 29].
`The regulatory studies with the NAPIESO combination
`tablet included a substantial number of patients who were
`also taking LDA, refiecting the frequency with which such
`dual NSAIDILDA therapy occurs in routine clinical prac-
`tice. In order to explore the possible GI and CV effects of
`combining LDA with either the combination tablet or other
`NSAID, prespecified analyses of ulcer incidence in patients
`stratified by [DA use were conducted and AE data from all
`5 Phase III studies were pooled in a post hoc analysis of the
`safety and tolerability of NAPIESO.
`
`Patients and methods
`
`Studies
`
`The study designs of the 5 Phase III studies included in this
`analysis have been reported previously [25—27]. Briefly,
`studies 301 (NCT00527787) and 302 (NCTOI 129011)
`
`were identically designed 6-month, randomized, double-
`blind, parallel-group studies comparing NAPIESO and EC
`naproxen tablets in patients who were at risk of developing
`GUS [26]. The primary endpoint was the cumulative inci-
`dence of patients with endoscopically observed GUs
`(33 mm diameter with depth) at any time throughout the
`6 months of treatment. Studies 307 [NCT00664560) and
`
`309 {NCT0066S43l) were identically designed 3-month,
`randomized, doubie-blind, placebo-controlled, parallel-
`group studies comparing NAPIESO, celecoxib, and pla-
`cebo, whose primary aim was to assess efficacy in pain
`relief of these agents in patients with OA of the knee, using
`the Pain and Function Subscales of the Western Ontario
`
`and McMaster Universities (WOMAC) OA index and the
`
`patient global assessment of OA questionnaire [25]. Study
`304 (NCT00527904) was a 12-month, open-label, multi-
`center study assessing the safety of NAPIESO in patients
`with OA, RA, or other conditions requiring daily NSAIDS
`for at least 12 months and at risk of GI events [27]. For all
`studies, data were collected on treatment-emergent AEs,
`serious AEs (SAES), AEs leading to discontinuation, and
`predefined NSAID-associated UGI AEs. In addition, stud-
`ies 301, 302, 307. and 309 included an assessment of tol-
`
`with a significantly lower incidence of endoscopic GUS
`compared with EC naproxen in patients at risk for devel-
`oping NSAID-associated ulcers [26]. Furthermore,
`long-
`term [12-month) use of NAPIESO was not associated with
`
`any new safety issues. including predefined UGI and CV
`
`erability endpoints, such as heartburn resolution, severity
`of dyspepsia assessment
`(SODA) or modified SODA
`(mSODA), and rescue antacid use, while study 304 col-
`lected data on heartburn and dyspepsia as AES, alongside
`exposure to, and dosage of, acetaminophen [25—27].
`
`Q Springer
`
`Page2of13
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`|PR2015-01718
`
`
`Page 2 of 13
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`IPR2015-01718
`
`

`
`LDA concomitant with NAPIESO: pooled safety data from 5 Phase III studies
`
`Patients
`
`The five studies enrolled patients with OA, RA, ankylosing
`spondylitis, or another condition expected to require chronic
`daily NSAID therapy. Studies 307' and 309 included patients
`with OA of the knee only. Eligible patients were aged 50 years
`or over. In addition, studies 30]. 302, and 304 also permitted
`younger patients (aged 18-49 years) provided they had a his-
`tory of uncomplicated GU or DU within the previous 5 years.
`The use of LDA (defined as 5325 mg/day) was allowed at the
`discretion of the treating physicians in all studies. Among the
`key exclusion criteria were uncontrolled or unstable cardiac
`disorder, prior GI disorder or surgery leading to impaired drug
`absorption, allergic reaction, or intolerance to any PPI or any
`NSAJD (including aspirin). In the endoscopic studies (301 and
`302), patients had to be ulcer—free at a baseline endoscopy.
`
`Study treatment
`
`In studies 301 and 302, patients received either oral NAP!
`ESO (EC naproxen 500 mgl'IR esomeprazole 20 mg) twice
`daily or oral EC naproxen 500 mg twice daily.
`In study
`304, patients
`received oral NAPIESO twice daily as
`described for studies 30l and 302. In studies 307 and 309,
`
`patients received oral NAPIESO twice daily, celecoxib
`200 mg twice daily, or placebo.
`Treatment was discontinued if patients withdrew
`informed consent, were judged by the investigator to be at
`significant safety risk, became pregnant. had a creatinine
`clearance of <30 mljmin, or had a confirmed decrease in
`
`hemoglobin level of >20 gfdL. In addition, in studies 30],
`302, and 304,
`treatment was discontinued if patients
`developed an ulcer.
`
`Incidence of ulcers
`
`Studies 301 and 302 assessed GUS and DUs using endos-
`copy. Data from these two studies were Pooled in a pre-
`defined analysis to assess the effect of NAPIESO plus
`concomitant LDA use on the incidence of GUS and DUs.
`
`Safety
`
`AEs and SAEs Occurring from the start of the study drug
`administration to the end of each study were recorded and
`coded using preferred terms from the Medical Dictionary
`for Regulatory Activities {MedDRA) version 10.]. Overall,
`AE and SAE data were pooled across all five studies for all
`patients who received 31 close of study drug). For the
`purpose of comparing safety across all five studies. a data
`cut-off of I20 days was used for studies 301, 302. and 304.
`For consistency across the studies, AEs identified via
`endoscopy were excluded in this analysis.
`
`Page 3of13
`
`CV events were prespecified in study 304 and were
`compiled by the sponsor’s physician and an independent
`cardiologist based on literature and medical expertise. This
`compilation was used for all the other studies. All CV AE
`data were pooled across all
`five studies and presented
`according to LDA users and non-users.
`
`Statistical analyses
`
`Overall. AE and SAE data were stratified by subgroups of
`LDA users and LDA non—users and pooled for post hoc
`analysis. Patients who were taking LDA at any time during
`the study period for a particular study were considered to
`be an LDA user. The incidence of an event refers to the
`
`proportion of patients who reported that event, and not the
`number of occurrences of that event.
`
`A summary of the cumulative observed incidence of
`GUs and DUs at 1, 3, and 6 months was produced based on
`the intent—to—treat (ITT) population in studies 301 and 302
`(i.e., all patients who received :1 dose of study drug and
`had no ulcer as detected by endoscopy at screening);
`however. the ITT and safety populations were identical in
`these two studies. Safety analyses were based on safety
`populations (all patients who received 31 dose of study
`drug)
`in each study. The incidences of endoscopically
`observed GU and DU, and incidences of AEs of erosive
`
`gastritis and erosive duodenitis. were analyzed using
`pooled data from studies 301 and 302 for the prespecified
`subgroups of LDA users and LDA non-users.
`The incidence of prespecified NSAID-associated UGI
`AEs (including dyspepsia, abdominal discomfort, gastritis,
`and vomiting; Table I), and discontinuation rates due to
`any AE or a prespecified NSAID—associated UGI AE, were
`summarized by LDA subgroup in the pooled safety popu-
`lations of all five studies.
`
`In order to accurately compare the safety results across
`the treatment groups in the five studies, which had varying
`study lengths and AE identification methods (e.g., use or
`non-use of endoscopy}, AEs SI3I’[il'lg >120 days after the
`first dose of study medication in studies 301, 302, and 304
`were not
`included in these summaries, nor were AEs
`identified during an endoscopy in studies 301 and 302.
`Statistical summaries were completed using Statistical
`Analysis System (SAS) version 8.
`
`Results
`
`Patients
`
`Overall. 2.31? patients were treated across the five studies.
`and 1,790 patients completed the studies (Fig. 1). Treat-
`ment arms within the individual studies were well-balanced
`
`Q Springer
`Patent Owner Ex. 2003
`CFAD v. Pozen
`|PR2015-01718
`
`
`Page 3 of 13
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`IPR2015-01718
`
`

`
`D. J. Angiolillo et al.
`
`upper
`Table l Prespecified NSAID-associated
`adverse events reported across all five studies
`
`gastrointestinal
`
`Abdominal
`discomfort
`
`Esophageal
`discomfort
`
`G1 hemorrhage
`
`Abdominal pain
`Abdominal
`tenderness
`DU
`
`Esophageal disorder
`Esophageal
`hemorrhage
`Esophageal stenosis
`
`GI mucosal disorder
`Hematemesis
`
`Hemorrhagic
`duodenitis
`
`non-users in the study-level analyses reported elsewhere, as
`the medication they were taking (Aggrenox or BC Powder)
`was not among the original LDA search terms. Two of
`these patients received an LDA dose of 5100 mglday.
`while the third had their LDA dosage classified as “other".
`The median durations of exposure to NAPIESO were
`
`178.5, 85, and 349 [22] days in the safety populations for
`studies 301 and 302 combined, 30? and 309 combined, and
`
`Duodenal
`hemorrhage
`Duodenal scarring
`DU hemorrhage
`
`Duodenitis
`Dyspepsia
`
`Epigastric
`discomfort
`Erosive duodenitis
`Erosive esophagitis
`Erosive gastritis
`
`Esophageal ulcer
`
`Hemorrhagic gastritis
`
`Esophageal varices
`Esophagitis
`
`Gastric hemorrhage
`Gastric mucosal
`lesion
`
`Gastritis
`
`GERD
`Gastrovesophagitis
`GI erosion
`
`1-Iyperchlorhydria
`Nausea
`
`Reflux esophagitis
`Stomach discomfort
`
`Upper abdominal
`pain
`Vomiting
`
`[26] Aliment Pitt.-rmacot' Titer 2010
`Adapted from Goldstein et al.
`with permission from John Wiley and Sons
`
`reflux disease, G.’
`DU duodenal ulcer. GERD gastrwesophageal
`gastrointestinal, NSAID nonsteroidal anti-inflammatory drug
`
`and baseline demographics and characteristics were similar
`for patients within studies [25—27]. Table 2 shows the
`baseline demographics and patient characteristics by LDA
`subgroup. Across the five studies, 4.8 % of patients had a
`previous history of ulcer, while 55.? % of patients had a
`previous history of CV events.
`Overall, 1,157 patients were treated with NAPIESO. Of
`these, 298 were identified as taking concomitant LDA
`(5325 mg/day} during the study (99 patients in studies 301
`and 302 combined. 124 patients in studies 307 and 309
`combined, and 75 patients in study 304). Of the 298
`patients who were identified as taking NAPIESO and
`concomitant LDA, an average daily LDA dose could be
`calculated for 292 patients. An LDA dose of 5 100 mgfday
`was received by 240 (80.5 %) of the NAPIESO patients,
`while 52 (12.4 %) patients received a dose of 101~325 mg}
`day.
`The average daily LDA dose could not be determined
`for 11 of the LDA users (rt = 6 in the NAPIESO group;
`it = 1
`in the placebo group; n = 4 in the EC naproxen
`
`group); for these patients, the LDA dose was classified as
`either “dose not recorded" or “unable to determine".
`
`Of the patients who were determined to be LDA users,
`3 patients (1 in the EC naproxen group of study 301.
`1
`in
`the EC naproxen group of study 302, and I
`in the NAP!
`ESO group of study 304) were originally classified as LDA
`
`study 304, respectively.
`
`Incidence of ulcers
`
`The cumulative incidence of GUS at month 6 by LDA use
`subgroup in studies 301 and 302 has been published pre-
`viously [26]. This publication reported that NAPIESO was
`associated with a significantly lower incidence of GUs than
`EC naproxen, irrespective of concomitant LDA use (3.0 vs
`28.4 %, respectively in LDA users and 6.4 vs 22.2 %,
`respectively in LDA non-users; P 4. 0.001 in favor of
`NAPIESO in both subgroups) [26]. The reclassification of
`two patients‘ LDA status for this analysis did not sub-
`stantially alter these findings: incidence of GUs with NAP!
`E50 vs EC naproxen in LDA users was 3.0 vs 27.9 %,
`respectively, and incidence in LDA non-users was 6.4 vs
`22.4 %, respectively.
`Among LDA users, the cumulative observed incidences
`of GUs at 1, 3, and 6 months in NAPIESO-treated patients
`were low and substantially less than those observed for EC
`naproxen-treated patients (Fig. 2); the cumulative observed
`incidences of DUs among patients receiving NAPKESO and
`concomitant LDA were also low and less than those
`
`observed for EC naproxen-treated patients (Fig. 2). Similar
`trends in the incidence of GUs and DUs were observed in
`
`the LDA non-user group at 1, 3, and 6 months (Fig. 2).
`
`Incidence of erosive gastritis and erosive duodenitis
`
`Overall, erosive gastritis was reported as an AB in fewer NAP!’
`ESO—treated patients
`than EC naproxen—treated patients
`[l9.4 % (83!428} vs 38.0 % (1621426), pooled analysis of data
`from studies 301 and 302; Chi squared P -1 0001]. Among
`LDA users, the incidence of erosive gastritis was significantly
`higher in the EC naproxen group compared with the NAPEESO
`group [36.5 % (381104) vs 18.2 % (18f99); Chi squared
`P = 0.0046]. A similar finding was observed for incidence of
`erosive gastritis among LDA non-users [38.5 % (124t'322) vs
`19.8 57: (65829) for EC naproxen and NAPIESO, respectively;
`Chi squared P < 0.0011. Of the patients who had erosive gas-
`tritis, 4 (0.9 %) patients in the NAPIESO treatment group (all
`LDA non-users) and 39 (9.2 %} patients in the EC naproxen
`group (12 LDA users and 27' LDA non-users) also had a GU.
`The incidence of erosive duodenitis was also lower
`
`among patients treated with NAPIESO than those receiving
`
`Q Springer
`
`Page4of13
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`|PR2015-01718
`
`
`Page 4 of 13
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`IPR2015-01718
`
`

`
`LDA concomitant with NAPIESO: pooled safety data from 5 Phase II] studies
`
`EC naproxen [2.1 % (91428) vs 11.’? % (5011426), pooled
`analysis
`in studies 301 and 302; Fisher's exact P <
`0.0001]. Among LDA users,
`the incidence of erosive
`duodenitis was lower for the NAPIESO group than the EC
`naproxen group [2.0 % (2199) vs 5.8 % (6.1 104)]. However,
`the test for differences was not significant (Fisher's exact
`
`P = 0.28). Among LDA non-users, rates of erosive duo-
`denitis were significantly lower for patients in the NAP!
`ESO group than the EC naproxen group [2.l % (#329) vs
`13.7 % (44,822), respectively; Fisher‘s exact P < 0.0001].
`Only one patient across both studies experienced both
`erosive duodenitis and a DU (an EC naproxen-treated
`patient in the LDA non—user group).
`
`Safety
`
`Adverse events
`
`the incidence of
`Among LDA users across all 5 studies,
`reported AEs was similar across all treatment groups; 56.0 %
`(167.1293) of NAPIESO-treated patients reported AEs com-
`pared with 58.7 % (6 1! I 04) of EC naproxemueated patients,
`
`53.8 % (561104) of celecoxib-treated patients, and 5'11 %
`(32156) of placebo-treated patients. Among LDA non-users,
`the corresponding incidences were also similar across treat-
`ment groups: 54.9 % (472.1859) for NAP/ESO; 59.6 % (192.-"
`322) for EC naproxen; 48.4 % (1861384) for celecoxib; and
`49.5 % (941190) for placebo (Table 3). GI disorders were the
`
`most commonly reported AEs in patients treated with NAP!
`ESO; the most common GI AE was dyspepsia (Table 3).
`Among LDA users.
`the incidences of prespecified
`NSALD-associated UGI AEs were lowest for NAP.-"ESO
`
`[l6.l % (48:"298)], highest for EC naproxen [31.7 % (33!
`104)], and were 22.1 % (231104) for celecoxib, and 23.2 %
`(13356) for placebo. The difference between NAPIESO and
`EC naproxen was statistically significant (Chi squared test,
`P = 0.001). The most common prespecified NSAID—asso—
`ciated UGI AEs were dyspepsia, nausea, and upper
`abdominal pain (Table 4). Among LDA non-users, pre-
`specified NSAID-associated UGI AEs were observed in
`20.3 % (174I859) of NAP)‘ESO—treated patients. 36.6 %
`(118.1322) of EC naproxen-treated patients (the highest
`incidence amongst the treatments considered), 18.5 % (71.1
`384) of celecoxib-treated patients, and 18.9 % (361190) of
`
`2317 patients
`
`I
`
`
`
`
`Study 309
`
`515 randomized
`
`Prernalure disoontinuatiens
`
`Nanreso: n = 41 me. n = I6: wimanaw ounsaril.
`n= 1i':tosthoFU.n =3'.oIhar. n = 5]
`Cdaentxib: n = 59 (AE. n = 22: withdraw consent.
`:1 = 25: hat in Iallaw-up. n : 3: other. n = 9:
`Placebo: n = 26 ME. n = 5: wlthalew consent.
`n = t5;!ns1InFU.n = homer. n =4]
`
`
`
`
`
`
`
`
`
`
`
`Premature
`
`cliscontinuations
`n = 98 ME. n = 45: withdraw
`
`
`consent. n = 21: Intel to
`FU. rt = azolnar. n = 22:
`
`
`
`
`
`C-:un-iplclncl
`n = 1 -‘J13
`
`
`
`
`
`Cnrr:J::[olr_'|:l'
`12r'nrJ:n.—. 135
`
`Premalure
`
`
`
`
`
`Premature
`discontinuations
`disoontinuations
`Premature discontinuations
`N.IP}'E%: II = 31! (RE. II = 14',
`NAPIEED: n = 61 (RE, fl = 20.‘
`wulhmew consent. in = 13: reel to
`WI|.hd|'9W consent.n=2\-|:Ios1 to
`N.nFJ'E5o: n = 40 (AE. n = 19: withdnaw nunsanl.
`Ft}. n =5:EHJ,n=1:o1net.n=5)
`FL|.n=8:DU, n =2:cLher. n = 1:]
`n=9:|osl!oFU.n=U:o1l'iar.n=I2)
`EC nllfllax: n = 6? LIE, n = 24;
`EC I'M‘-“Olin: 53ls\E.n=30:
`Calaouxlb: n = 39 |AE. n = 16: WIHUIEW oonsanl.
`
`wimcIreweonsnnt.n=8:|oe1rn
`wilhcltew oonsenl. n = 25: Iosl to
`n:13:Ios1toIollaw-up.n=2:o1ne¢.n=a:
`iollaw-uo. II = 2: UU. n = 10;
`rolowvup. n = 7; DU. I1 = 3:
`Pl-aneho-. n = 191:5. n = 7; withdrew consent.
`other. :1 = B?
`
`ulhar. II = 5]
`n=7;IoatoFu_n=n:o1nor.n=5;
`
`
`Completed study’
`n=18l3
`n:-.153
`
`
`
`Completed study
`n=203
`n = 16-3
`
`
`
`
`n=9El
`
`
`
`_n=151
`
`n=15,3
`
`Fig. 1 Patient disposition in the five studies. AE adverse event, DU
`duodenal ulcer, EC enteric-coated. FU follow-up. IT?‘ intent-to-treat.
`mo month.
`rm'TT modified intent—t0—treat, NAP/E80 naproxen!
`esomeprazole magnesium, naprox.
`naproxen, pop. population,
`
`S safety. *Patients completed 6 months of study treatment or
`discontinued due to gastric ulcer. 1Patients completed 2348 days on
`study treatment
`
`Page 5of13
`
`Q Springer
`Patent Owner Ex. 2003
`CFAD v. Pozen
`|PR2015-01718
`
`
`Page 5 of 13
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`IPR2015-01718
`
`

`
`Table 2 Patient demographics and baseline characteristics across ali five studies in the subgroups of (a) LDA users and (b) LDA non-users (ITT
`populations)
`Chameteristic
`
`Studies 3017302
`
`Studies 3077309
`
`NAPIESO
`71 = 99
`
`EC nuproxen
`11 = 104
`
`NAPIESO
`:1 = 124
`
`Celecoxib
`71 = 104
`
`D. J. Angiolillo et al.
`
`Study 304
`NAPIESO
`71 = 75
`
`Total
`NAPIESO
`71 = 298
`
`L19/1 u.s'e1'.r
`
`Age group. y (n. 93)
`<50
`
`<60
`360
`Sex, female. n (*2)
`
`Diagnosis. :1 (%)"
`OA
`RA
`
`A8
`Other
`
`History of ulcer, 71 (%)
`GU
`DU
`Both GU and DU
`
`Peptic ulcer
`G1 ulcer
`
`None
`
`CV history. 71 ('56)
`Yes
`No
`
`2 ( 1 .9)
`42 (40.4)
`62 (59.5)
`63 (60.6)
`
`87 (33.7)
`4 (3.8)
`1 (1.0)
`24 (23.1)
`
`8 (7.7)
`
`1 (1.0)
`0
`
`3 (3.0)
`
`37 (37.4)
`62 (62.6)
`48 (48.5)
`
`83 (83.8)
`6 (6.1)
`
`1 (1.0)
`20 (202)
`
`9 (9.1)
`2 (2.0)
`0
`
`—
`—
`
`0
`
`32 (25.8)
`92 (74.2)
`70 (56.5)
`
`0
`
`33 (31.7)
`71 (68.3)
`56 (53.8)
`
`20 (35.7)
`36 (64.3)
`
`31 (55.4)
`
`124(I00)
`
`104(I00)
`
`56(100)
`
`I (0.8)
`
`1 (0.8)
`0
`
`0
`0
`
`2 (1.9)
`
`0
`0
`
`3 (2.9)
`0
`
`CJDDCDCJ
`
`1 (1.3)
`29 (38.7)
`46 (61.3)
`41 (54.7)
`
`59 (78.7)
`7 (9.3)
`
`I (1.3)
`l4(l8.7)
`
`6 (8.0)
`0
`0
`—
`—
`
`4 (1.3)
`98 (32.9)
`200 (67.1)
`159 (53.4)
`
`266 (89.3)
`I3 (4.4)
`
`2 (0.7)
`34(11.4)
`
`16 (5.4)
`3 (1.0)
`0
`0
`0
`
`88 (88.9)
`
`95 (91.3)
`
`122 (98.4)
`
`99 (95.2)
`
`56 (I00)
`
`69 (92.0)
`
`279 (93.6)
`
`71 (71.7)
`28 (28.3)
`
`73 (70.2)
`3] (29.8)
`
`92 (74.2)
`32 (25.8)
`
`69 (66.3)
`35 (33.7)
`
`40 (71.4)
`16 (28.6)
`
`57 (76.0)
`18 (24.0)
`
`220 (73.8)
`78 (26.2)
`
`0
`
`Oral steroid use at baseline, 11 (%)h
`Yes
`0
`
`No
`Characteristic
`
`99 (I00)
`Studies 3017302
`
`0
`
`104 (100)
`
`0
`
`I24(100)
`Studies 3077309
`
`104(100)
`
`55(100)
`
`NAPIESO
`71 = 329
`
`EC naproxen
`n = 322
`
`NAPIESO
`n = 366
`
`Celecoxib
`71 = 384
`
`Placebo
`71 = I 90
`
`LDA n0r1—L1se'rs
`
`Age group, y (71, %)
`-<50
`<60
`360
`Sex. female. 71 (%)
`
`11 (3.3)
`179 (54.4)
`150 (45.6)
`234 (71.1)
`
`7 (2.2)
`175 (54.3)
`147 (45.7)
`223 (70.3)
`
`0
`174 (47.5)
`192 (52.5)
`250 (68.3)
`
`I (0.3)
`192 (50.0)
`192 (50.0)
`245 (63.8)
`
`0
`
`91 (47.9)
`99 (52.1)
`128 (67.4)
`
`Diagnosis or reason for NSAID use. 71 (%)"
`0A
`262 (79.6)
`265 (32.5)
`RA
`27 (8.2)
`13 (4.0)
`A5
`2 (0.6)
`1 (0.3)
`
`Other
`
`78 (23.7)
`
`71 (22.0)
`
`366 (I00)
`
`384 (100)
`
`190 (I00)
`
`23 (7.1)
`4 (1.2)
`1 (0.3)
`
`15 (4.6)
`5 (1.5)
`0
`
`—
`—
`
`3 (0.8)
`0
`0
`
`0
`0
`
`3 (0.8)
`I (0.3)
`0
`
`3 (0.8)
`0
`
`309 (93.9)
`
`294 (91.3)
`
`363 (99.2)
`
`377 (98.2)
`
`3 (1.6)
`2 (1.1)
`0
`
`1 (0.5)
`
`1 (0.5)
`133 (96.3)
`
`History of ulcer, 71 (%)
`GU
`DU
`Both GU and DU
`
`Peptic ulcer
`GI ulcer
`
`None
`
`Q Springer
`
`Page 6 of 13
`
`2 (2.7)
`73 (97.3)
`
`Study 304
`
`NAPKESO
`71 = 164
`
`6 (3.7)
`90 (54.9)
`74 (45.1)
`I27 (77.4)
`
`I30 (79.3)
`14 (8.5)
`I (0.6)
`
`38 (23.2)
`
`1] (6.7)
`3 (1.8)
`0
`—
`
`—
`
`2 (0.7)
`296 (99.3)
`
`Total
`
`NAPIESO
`77 = 859
`
`17 (2.0)
`443 (51.6)
`416 (48.4)
`611 (71.1)
`
`758 (88.2)
`41 (4.8)
`3 (0.3)
`
`I16 (13.5)
`
`29 (3.4)
`8 (0.9)
`0
`0
`
`0
`
`I50 (91.5)
`
`822 (95.7)
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`|PR2015-01718
`
`
`Page 6 of 13
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`IPR2015-01718
`
`

`
`LDA concomitant with NAPIESO: pooled safety data from 5 Phase Ill studies
`
`Table 2 continued
`
`Characteristic
`
`Studies 30 U302
`
`Studies 3071309
`
`NAPIESO
`n = 329
`
`EC naproxen
`n = 322
`
`HIAPIESO
`n = 366
`
`Celecoxib
`H = 334
`
`Placebo
`:1 = 190
`
`Study 304
`
`NAP/ESO
`:1 = [64
`
`Total
`
`NAPIESO
`H = 859
`
`CV history. ii (96)
`174 (52.9)
`Yes
`155 (47.1)
`N0
`Ural steroid use at baseline. :1 (%)h
`Yes
`0
`
`156 (48.4)
`166 (51.6)
`
`I73 (47.3)
`I93 (52.7?)
`
`209 (54.4)
`175 (45.6)
`
`92 (48.4)
`93 (51.6)
`
`85 (51.8)
`T9 (43.2)
`
`432 (50.3)
`427 (49.7)
`
`0
`
`0
`
`0
`
`0
`
`0
`
`0
`
`No
`
`329 (100)
`
`322 (I00)
`
`366 (100)
`
`384(l0'0)
`
`190 (I00)
`
`l64(l00)
`
`859(l00)
`
`AS ankylosing spondylitis, CV cardiovascular, DU duodenal ulcer. EC enteric-coated. GI gastrointestinal, GU gastric ulcer, l'T1"intent-to-treat.
`LDA lowvdose aspirin, NAP/ESO naproxenfesomeprazole magnesium, 0A osteoarthritis, RA rheumatoid arthritis
`“ Patients may have had more than one reason for use of NSAIDS
`
`" Any medication taken in "corticosteroids for systematic use. plain" drug class
`
`I N;.p;Ego 01:99}
`1: EC Naproxenljn = 104}
`
`([3)
`
`DU
`
`25 -
`
`tau
`
`mi
`'
`
`Fig. 2 Pooled cumulative
`observed incidence of gastric
`ulcers and duodenal ulcers at
`Month 0, 3, and 6 in :1 LDA
`users and b LDA non—user-s
`(intent-to-treat population,
`pooled data from studies 301
`
`and 302). EC enteric-coated,
`LDA low—dose aspirin, NAP!
`ESO naproxenfesomeprazole
`magnesium
`
`(a)
`
`an
`
`25
`
`20
`
`gr
`j;
`E
`g
`E 15
`
`GU
`
`23.1
`
`M
`
`10
`
`5
`
`0
`
`DD
`
`1
`
`00
`3
`Monlh
`
`2%
`
`I E
`
`O
`
`I NAPIESO (rt = 329;
`:| E0 Naproxen (n = 322:
`
`DU
`
`1|
`
`0.5
`
`5‘,
`
`53
`
`0.5
`
`3
`Month
`
`0.5
`
`5
`
`5
`
`i
`E
`.
`
`I
`
`I
`
`171
`
`10.2
`
`64
`
`H
`
`2|
`
`Month
`
` 5 _
`
`
`
`Cumulativeincidence(is)
`
`G
`
`5
`
`0
`
`patients receiving placebo (Table 4). Again, the difference
`between NAPIESO and EC naproxen was statistically
`significant (Chi squared test, P < 0.001).
`
`patients receiving EC naproxen, celecoxib, and placebo,
`respectively; CV disorders that occurred in two or more
`patients were palpitations and cardiornegaly (Table 5).
`
`Among LDA users, CV AEs occurred in 3.0 % (93
`298) of NAPIESO-treated patients compared with 1.0 %
`(U104), 0 % and 0 % of patients receiving EC naproxen,
`celecoxib. and placebo, respectively: cardiovascular dis-
`orders that occurred in two or more patients were cor-
`onary artery disease and palpitations
`(Table 5). One
`NAPIESO-treated patient who was
`in the LDA user
`subgroup experienced atrial fibrillation, which was clas-
`sified as mild. For LDA non-users, CV AEs were
`
`reported for 1.0 % (91859) of NAPi’ESO—treated patients
`compared with 0.6 % (2i"322), 0.3 ‘X: (H384). and 0 % of
`
`the LDA non—user
`in
`One NAPi’ESO—treated patient
`group had an SAE of peri-operative MI. This serious CV
`event, which occurred 6 days after the patient was hos-
`pitalized for unstable angina, subsequently resolved, and
`was assessed by the investigator as being unrelated to
`NAPIESO use.
`
`incidences of SAES among LDA users
`The overall
`across the five pooled studies were 2.7 % (8i'298)
`in
`NAPIESO-treated
`patients.
`2.9% (3il0-4)
`in
`EC
`naproxen—treated patients, 4.3 % (5i'l04)
`in celecoxib—
`treated patients, and 0.0 % in placebo-treated patients
`
`Page7of13
`
`Q Springer
`Patent Owner Ex. 2003
`CFAD v. Pozen
`|PR2015-01718
`
`
`Page 7 of 13
`
`Patent Owner Ex. 2003
`CFAD v. Pozen
`IPR2015-01718
`
`

`
`D. J. Angiolillo et al.
`
`Table 3 Summary of treatment-related adverse effects occurring in 335 % of patients in any treatment group according to LDA use (safety
`population. pooled data from five studies)
`LDA users
`
`LDA non-users
`
`Adverse event, 11 (%) of patients
`
`Dyspepsia
`Diarrhea
`Nausea
`
`Upper abdominal pain
`Headache
`
`Cough
`URTI
`
`Lower abdominal pain
`
`NAPFESO EC naproxen
`(1: = 298)
`(11 = 104)
`
`Celocoxib
`(ti = 104)
`
`Placebo
`(11 = 56)
`
`NA PFESO
`(:1 = 859)
`
`EC naproxcn
`(fl =
`
`Celecoxib
`(11 = 334)
`
`Placebo
`(11 = 190)
`
`23 (7.7)
`17 (5.7)
`15 (5.0)
`[4 (4.7)
`[4 (4.7)
`10 (3.4)
`9 (3.0)
`
`4 (1.3)
`
`19 (18.3)
`4 (3.8)
`3 (2.9)
`9 (3.7)
`2 (1.9)
`7 (6.7)
`5 (4.3)
`
`2 (1.9)
`
`8 (7.7)
`4 (3.8)
`7 (6.7)
`5 (4.8)
`4 (3.8)
`0 (0.0)
`2 (1.9)
`
`0 (0.0)
`
`5 (8.9)
`2 (3.6)
`4 (7.1)
`3 (5.4)
`4 (7.1)
`2 (3.6)
`3 (5.4)
`
`3 (5.4)
`
`97 (11.3)
`41 (4.8)
`32 (3.7)
`31 (3.6)
`15 (1.7)
`10 (1.2)
`25 (2.9)
`
`17 (2.0)
`
`85 (26.4)
`15 (4.7)
`15 (4.7)
`22 (6.8)
`3 (0.9)
`3 (0.9)
`8 (2.5)
`
`8 (2.5)
`
`441-(11.5)
`[0 (2.6)
`8 (2.1)
`[6 (4.2)
`[4 (3.6)
`3 (0.8)
`4 (1.0)
`
`2S(|3.2)
`7 (3.7)
`5 (2.6)
`5 (2.6)
`9 (4.7)
`5 (2.6)
`2 (1.1)
`
`5 (1.3)
`
`0 (0.0)
`
`Excludes events identified during endoscopy for studies 301 and 302. Excludes events that occurred after day 120 for studies 301, 302, and 304
`EC enteric~coated, LDA low—dose aspirin. NAP/ESO naproxenfesoineprazole magnesium, URT1 upper respiratory tract infection
`
`Table 4 Summary of prespecified NSAID-associated UGI AEs (occurring in 31 patient in the NAPIESO treatment group) according to LDA
`use (safety popuiation. pooled data from five studies)
`LDA users
`
`LDA non-users
`
`Adverse event, H (9%))
`
`Any
`Dyspepsia
`Nausea
`
`Upper abdominal pain
`Abdominal pain
`Abdominal discomfort
`GERD
`Stomach discomfort
`
`Vomiting
`Reflux esophagitis
`Abdominal tenderness
`Duodenal ulcer
`
`Epigastric discomfort
`Gastritis
`
`Hyperchlorhydria
`Duodenal ulcer hemorrhage
`
`Erosi ve gastritis
`Esophagitis
`G1 hemorrhage
`Hematemesis
`
`NAPIESO
`(:1 = 293)
`
`48 (16.1)
`23 (7.7)
`
`15 (5.0)
`14 (4.7)
`5 (1.7)
`3 (1.0)
`2 (0.7)
`2 (0.7)
`2 (0.7)
`1 (0.3)
`
`0 (0.0)
`0 (0.0)
`0 (0.0)
`0 (0.0)
`0 (00)
`NR
`NR
`NR
`NR
`NR
`
`EC naproxen
`(11 = 104)
`
`Celecoxih
`(1) = 104)
`
`Placebo
`(H = 56)
`
`NAPIESO
`(:1 = 359)
`
`EC naproxen
`(H = 322)
`
`Celecoxib
`(rt = 384)
`
`Placebo
`(ii = 190)
`
`33 (31.7)
`19 (18.3)
`3 (2.9)
`9 (8.7)
`0 (0.0)
`0 (0.0)
`4 (3.8)
`0 (0.0)
`0 (0.0)
`0 (0.0)
`
`0 (0.0)
`0 (0.0)
`1 (1.0)
`1 (1.0)
`0 (0.0)
`NR
`NR
`NR
`NR
`NR
`
`23 (22.1)
`8 (7.7)
`7 (6.7)
`5 (4.8)
`0 (0.0)
`1 (1.0)
`2 (1.9)
`1 (1.0)
`1 (1.0)
`0 (0.0)
`
`0 (0.0)
`1 (1.0)
`0 (0.0)
`1 (1.0)
`1 (10)
`NR
`NR
`NR
`NR
`NR
`
`13 (23.2)
`5 (8.9)
`4 (7.1)
`3 (5.4)
`2 (3.6)
`0 (0.0)
`1 (1.8)
`1 (1.8)
`1 (1.8)
`0 (0.0)
`
`2 (3.6)
`0 (0.0)
`0 (0.0)
`1 (1.8)
`
`0 (0.0)
`NR
`NR
`NR
`NR
`NR
`
`174 (20.3)
`97 (11.3)
`32 (3.7)
`31 (3.6)
`13 (1.5)
`6 (0.7)
`6 (0.7)
`5 (0.6)
`11 (1.3)
`1 (0.1)
`
`118 (36.6)
`85 (26.4)
`15 (4.7)
`22 (6.8)
`5 (1.6)
`6 (1.9)
`5 (1.6)
`4 (1.2)
`5 (1.6)
`0 (0.0)
`
`1 (0.1 )
`NR
`
`1 (0.1)
`1 (0.1)
`2 (0.2)
`0 (0.0)
`1 (0.1)