GASTROENTEROLOGY 1997;112:1817-1822
`
`Famotidine for Healing and Maintenancein Nonsteroidal Anti-
`inflammatory Drug—Associated Gastroduodenal Ulceration
`
`NICHOLAS HUDSON,* ALI S. TAHA,* ROBIN |. RUSSELL,* PENELOPE TRYE,® JEREMY COTTRELL,®
`STEPHEN G. MANN, ANTHONY J. SWANELL,' ROGER D. STURROCK,*
`and CHRISTOPHER J. HAWKEY*
`* Division of Gastroenterology, University Hospital, Nottingham, England; ‘Department of Gastroenterology and Rheumatology, Glasgow
`Royal Infirmary, Glasgow, Scotland; Merck Research Laboratories UK, Hoddesdon, Herts, England; and ''Department of Rheumatology,
`University and City Hospitals, Nottingham, England
`
`See editorial on page 2143.
`
`Background & Aims: Nonsteroidal anti-inflammatory
`drugs (NSAIDs) are strongly associated with gastroduo-
`denal ulceration. How to manage patients with NSAID-
`associated ulcers is a commonclinical dilemma. High-
`dose famotidine in the healing and maintenance of
`NSAID-associated gastroduodenal
`ulceration was
`therefore evaluated. Methods: One hundred four pa-
`tients with rheumatoid or osteoarthritis who had gastro-
`duodenal ulceration received famotidine, 40 mg twice
`daily. Sixteen patients stopped and 88 continued their
`NSAID treatment. Ulcer healing was assessed endo-
`scopically at 4 and 12 weeks. Seventy-eight NSAID
`users with healed ulcers were then randomized to re-
`
`ceive 40 mg twicedaily famotidine or placebo and un-
`derwent endoscopy at 4, 12, and 24 weeks. Results:
`Cumulative ulcer healing rates at 12 weeks were 89.0%
`(95% confidence interval [Cl], 82.3%-—95.7%) for pa-
`tients who continued NSAID treatment and 100% (95%
`Cl, 82.9%-100.0%) for those who stopped. The subse-
`quent estimated cumulative gastroduodenal ulcer re-
`lapse over 6 months for NSAID users who took placebo
`was 53.5% (95% Cl, 36.6%-—70.3%). This was reduced
`to 26.0% (12.1%-39.9%) in patients taking famotidine
`(P = 0.011). Conclusions: High-dose famotidine is ef-
`fective ulcer healing therapy in patients who stop or
`continue NSAID treatment and significantly reduced
`the cumulative incidence of gastroduodenal ulcerrecur-
`rence compared with placebo when given as mainte-
`nance therapy.
`
`onsteroidal anci-inflammatory drugs (NSAIDs) are
`
`Nie: associated with gastroduodenal ulceration
`
`and the complications of ulcer hemorrhage and perfora-
`tion.'~* These risks seem to be increased in patients with
`a history of gastroduodenal ulceration.’* Use ofthe pros-
`taglandin analogue misoprostol for prophylaxis against
`development of NSAID-associated ulceration is well es-
`
`tablished,°~'' but there is very little evidence aboutulcer
`healing.'* Treatments that suppress acid are better toler-
`ated, but high doses are needed to prevent acute mucosal
`injury.'*-'> Because managementof patients presenting
`with ulceration represents the most common dilemma
`in the management of such patients, we conducted a
`study to assess the efficacy of high-dose famotidine, an
`H, antagonist, in both the healing of NSAID-associated
`gastroduodenal ulceration and the subsequent prevention
`of ulcer relapse when given as maintenance therapy. This
`enabled us to compare ulcer developmentrates with those
`observed in a study of primary prophylaxis conducted to
`an identical design, at the same time,
`in a cohort of
`patients drawn from the same population.
`
`Patients and Methods
`
`Design
`
`The study consisted of two phases: an open study of
`famotidine, 40 mg twicedaily, in the healing of endoscopically
`proven gastroduodenal ulceration and, in patients with success-
`ful ulcer healing, a prospective randomized double-blind pla-
`cebo-controlled maintenance study of famotidine, 40 mg twice
`daily, as secondary prophylaxis against endoscopically detected
`recurrent gastroduodenal ulceration.
`
`Patients
`
`Adult patients (aged =18 years) with rheumatoid ar-
`thritis or osteoarthritis were recruited from the rheumatology
`and orthopedic clinics, provided they had been receiving an
`NSAID within the range of standard recommended dosage for
`at least
`1 month before endoscopy. Patients were not consid-
`ered for the study if they had been taking antiulcer drugs
`other chan antacids <7 days before study entry or were taking
`steroids at a dosage equivalent =7.5 mg prednisolone daily,
`methotrexate, or antineoplastic drugs. The other main exclu-
`
`Abbreviations used in this paper: Cl, confidence interval; HAQ,
`Health Assessment Questionnaire.
`© 1997 by the American Gastroenterological Association
`0016-5085/97/53.00
`
`Page 1 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`Page 1 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`1818 HUDSON ET AL.
`
`GASTROENTEROLOGYVol. 112, No. 6
`
`sion criteria were lactation, child-bearing porential, renal fail-
`ure, diabetes, or clinically significant prestudy laboratory ab-
`normalities.
`
`Recruitment Procedures
`
`Recruitment was performed by two gastroenterologists
`who attended the Rheumatology and Orthopedic Clinics and
`approached all potentially suitable arthritic patients regardless
`of dyspeptic symptoms. Patients who accepted the invitation
`to participate underwent screening endoscopy. Ac endoscopy,
`ulcers, erosions, and intramucosal hemorrhages were recorded
`separately for the esophagus, gastric body, gastric antrum,
`duodenal bulb, and second part of the duodenum.
`
`Definitions and Assessments
`
`An ulcer was defined as an excavated mucosal break
`
`>3 mm in diameter, measured using biopsy forceps or a cus-
`tom-made measuring device. Erosions (defined as superficial
`mucosal breaks) and intramucosal hemorrhages (defined as
`hemorrhagic lesions without an overlying mucosal break), to-
`gether with any ulcer in each of the target areas, were used to
`derive Lanza scores.'® Helicobacter pylori status was determined
`by CLO test and gastric antral histology. Before commence-
`ment of the study,
`the two endoscopists (A.S.T. and N.H.)
`attended each other's endoscopic sessions and reviewed still
`and video images as a process of standardization of reporting
`criteria for ulcers and other lesions. Both studies were approved
`by the Nottingham and Glasgow Hospital ethics committees.
`
`Healing Phase
`
`Patients with ulcers were entered into the healing
`study and invited to discontinue their NSAID therapy, Pa-
`tients without ulceration entered a prophylaxis study, reported
`elsewhere.'’ Patients received two 20-mg famotidine tablets
`twice daily.
`In addition, antacid tablets (Maalox; Rhone-
`Poulenc Rorer, Eastbourne, East Sussex, England) were pro-
`vided for relief of dyspepsia as required. Patients underwent
`endoscopyafter 4 weeks oftreatment and, if the ulcer remained
`unhealed, underwent a repeat endoscopy at 12 weeks, Patients
`who were unhealed at 12 weeks were designated as treatment
`failures.
`
`Maintenance Phase
`
`Patients who wished to continue NSAID therapy and
`whose ulcers had healed successfully during the study or during
`the following 4 weeks were invited to enter the maintenance
`study. These patients were randomized to receive either famoti-
`dine, 40 mg twice daily, or placebo in a double-blind fashion
`and underwent further endoscopy at 4, 12, and 24 weeks or
`until ulcer relapse during the study.
`
`Nonendoscopic Assessments
`
`Patients were assessed routinely at baseline and at the
`time of each subsequent endoscopy. In addition to endoscopy,
`the following assessments were performed: NSAID and other
`drug usage, arthritis-related physical disability measured by
`
`the Health Assessment Questionnaire (HAQ), vital signs, com-
`plete physical examination (baseline and end ofstudy), urinaly-
`sis, hematology, and biochemistry. Antacid use and abdominal
`symptoms were recorded daily on specific diary cards. Compli-
`ance with study drugs was assessed by tablet count and adverse
`events by open questioning at each visit.
`
`End Points
`
`The primary end point of the healing study was healing
`of gastroduodenal ulceration at the 4- or 12-week endoscopy.
`The primary end point ofthe maintenance study was the cumu-
`lative incidence of gastroduodenal ulcer relapse as assessed at
`follow-up endoscopies (at 4, 12, and 24 weeks). The main
`secondary end points were the corresponding findings related
`to gastric and duodenalulcers separately, Lanza scoresfor lesser
`degrees of gastroduodenal injury, abdominal pain, and antacid
`consumption. The main safety analyses included assessment
`of adverse events, arthritis, HAQ, physical examination, and
`laboratory results.
`
`Statistical Methods
`
`The study was designed to be pragmatic rather than
`explanatory. Efficacy data in both the healing and maintenance
`study were therefore subjected to a primary “all patients
`treated” analysis. Product limit (Kaplan—Meier) estimates of
`the cumulative incidence of ulceration, the primary end point,
`were made for each endoscopy visit. Comparisons between
`treatment groups were made using the log rank test co allow
`for those withdrawn for reasons other than ulceration. The
`
`cumulative incidence of gastric ulceration and duodenal ulcer-
`ation were also similarly analyzed separately. Similar per-proto-
`col analyses of efficacy were also performed on assessable pa-
`tients. Assessable patients were defined as those whosatisfied
`the inclusion and exclusion criteria, consumed >80% of both
`the prescribed NSAID and study drugs, did not consume addi-
`tional full-dose salicylates, and had their end-of-study endos-
`copy performed no more than 5 days after the end of study
`treatment. Secondary efficacy end points were also analyzed
`using “‘all patients treated’ as the primary analysis and a per-
`protocol approach as the secondary analysis. The Mantel—
`Haenszel test or Kruskal-Wallis test were used where appro-
`priate. Comparisons were considered significant at P values of
`=0,05.
`
`regression
`Prognostic factors. A Cox proportional
`model was constructed to assess the effects of 26 potential
`prognostic factors on cumulative ulcer incidence. The potential
`prognostic factors were identified before the study. These fac-
`tors were added to or removed from the model in a stepwise
`regressive procedure, based on a threshold P value (<0.1 to
`be added or >0.1 to be removed), with the effect of treacment
`group being included at each stage. The prognostic variables
`included the following; recruiting center, age, sex, smoking
`habit, alcohol use, type of NSAID, duration of prior NSAID
`use, rheumatological diagnosis, duration of arthritis, history
`of peptic ulceration, time to ulcer healing, presence of erosions
`or hemorrhagic lesions at screening endoscopy, ulcer size at
`
`Page 2 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`Page 2 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`June 1997
`
`FAMOTIDINE AND NSAIDs 1819
`
`Table 1. Patient Characteristics in Healing and
`Maintenance Study (‘All Patients Treated”’
`Population)
`
`Maintenance
`
`%healed &
`
`NSAID stopped
`
`NSAID continued
`
`ah
`
`
`
`4
`
`Weeks
`
`8
`
`12
`
`Ulcer healing
`(n = 104)
`
`Famotidine 40 mg
`(n = 39)
`
`Median age (yr, range)
`M/F
`Smokers (55)
`Alcohol consumption (%6)
`Rheumatoid/osteoarthritis
`Previous PUD (%)
`Second-line therapy (9%)
`Mean HAQ score (SD)
`NSAID type
`Naproxen (%)
`Indomethacin (%)
`Diclofenac (%)
`Other(%)
`PUD, peptic ulcer disease.
`
`58 (31-89)
`32/72
`32 (31)
`45 (43)
`82/22
`26 (25)
`25 (24)
`1.4 (0.8)
`
`31 (30)
`419 (18)
`14 (13)
`40 (38)
`
`58 (32-79)
`14/25
`13 (33)
`17 (44)
`34/5
`12 (31)
`12 (31)
`1.5 (0.8)
`
`14 (36)
`3 (8)
`6 (15)
`46 (41)
`
`Placebo
`(n = 39)
`
`55 (35-89)
`12/27
`12 (34)
`18 (46)
`33/6
`11 (28)
`10 (18)
`1.5 (0.8)
`
`14 (28)
`9 (23)
`4 (10)
`415 (39)
`
`endoscopy, abdominal pain at baseline, HAQ score, second-
`line antirheumatoid agent prednisolone, total leukocyte count,
`hemoglobin, and platelet count.
`
`Results
`
`Five hundred seventy patients (Glasgow, n = 299;
`Nottingham, n = 271) were invited to enter the study,
`of whom 389 accepted (Glasgow, n = 235; Nottingham,
`n = 154). Of these, 104 had gastroduodenal ulcers at
`screening endoscopy (gastric ulcers, n = 76; duodenal
`ulcers, n = 42; and both, n = 14) and were entered into
`the healing study; 69 (66%) were recruited from Glasgow
`and 35 (34%) from Nottingham. H. pylori status was
`established (histology and urease test) in 93 patients.
`The patients with no ulcer at screening participated in
`a primary prophylaxis study run concurrently. As shown
`in Table 1, patients were well matched for age, sex,
`smoking status, alcohol usage, underlying arthritis, ulcer
`history, frequency of joint pain, HAQ score, and use of
`individual NSAIDsor disease-modifying drugs. Eighty-
`two patients had rheumatoid arthritis and 22 had osteo-
`arthritis. Sixteen patients agreed to stop their NSAID
`therapy during the course of the healing study, and 88
`continued NSAID therapy.
`
`Ulcer Healing Study
`
`Figure 1. Proportion of patients with ulcers healed with famotidine,
`40 mg twice daily, at 4 and 12 weeks in those who continued or
`discontinued NSAID therapy.
`
`NSAID therapy. Differences in healing rates were not
`significant at either 4 or 12 weeks (Figure 1). Ulcers
`failed to heal in 9 patients after 12 weeks of therapy, all of
`whom continued NSAID therapy. For the corresponding
`results per protocol, the cumulative healing rates were
`67.6% (95% CI, 56.7%—-78.5%) vs. 76.9% (95% CI,
`54.0% —99,8%) at 4 weeks and 91.2% (95% CI, 84.4% —-
`97.9%) vs. 100% (95% CI, 79.4%—-100.0%) at 12
`weeks in NSAID users and nonusers, respectively.
`Of the prognostic variables, only ulcer size was sig-
`nificantly and inversely correlated with ulcer healing.
`Nonetheless, the 12-week cumulative healing rate for 49
`ulcers that were >5 mm in diameter was 81.6% (95%
`CI, 70.7-92.5). Healing occurred in 85.7% (95% CI,
`76.6% —94.8%) of patients who were H. pylori negative
`and 93.0% (95% CI, 86.4% —99.6%) of those who were
`H. pylori positive. When healing rates for gastric and
`duodenal ulcer were analyzed separately, the cumulative
`healing rates for gastric ulcers in 63 patients who contin-
`ued NSAID therapy were 63.5% (95% CI, 51.6%—-
`75.4%) at 4 weeks and 86.7% (95% CI, 78.2% —95.2%)
`at 12 weeks. For duodenal ulcers, healing rates were
`75.7% (95% CI, 61.9% —89.5%) and 97.0% (95% CI,
`91.1% —100.0%), respectively. Similar results were ob-
`tained from the per-protocol analysis.
`
`Maintenance Study
`
`Seventy-eight patients who continued to use
`NSAIDsagreed to enter the maintenance study, although
`1 patient did not undergo repeat endoscopy. Patient char-
`acteristics are shown in Table 1. On “all patients treated”
`
`At 4 weeks, 65.9% (95% confidence interval [CI],
`56.0% —75.8%) of patients who continued and 81.3%
`(95% CI, 54.4% -96.0%) who discontinued NSAID
`therapy had successfully healed ulcers on “all patients
`treated” analysis. Two patients unhealed at 4 weeks were
`not assessed further (see below). At 12 weeks, the cumu-
`lative healing rate was 89.0% (95% CI, 82.3% —95.7%)
`in patients who remained on NSAID therapy and 100%
`(95% CI, 82.9% -100.0%) in those who discontinued
`
`Weeks
`4
`12
`24
`
`
`+} _
`te
`—f — ——
`
`Famotidine
`omgba
`Placebo
`
`me 100
`= 2
`a2
`5 5 50
`ee 0
`
`Figure 2. Life table of cumulative incidence of ulcer recurrence rates
`at 4, 12, and 24 weeksin patients taking maintenance famotidine,
`40 mg twice daily, compared with placebo,
`
`Page 3 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`Page 3 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`1820 HUDSON ET AL.
`
`GASTROENTEROLOGY Vol. 112, No. 6
`
`Table 2. Prognostic Variables for Ulcer Relapse During
`Maintenance Phase
`
`Cumulative relative risk
`
`
` (95% Cl) P value
`
`Famotidine 40 mg twice daily
`Duodenalulcer in healing study
`Length of time to healing
`HAQ score
`Use of ketoprofen
`Shorter duration of arthritis
`Model with H. pylori included
`Famotidine 40 mg twice daily
`Shorter duration of arthritis
`Baseline H. pylori status
`
`0.093 (0.025-0.34)
`4.919 (1.33-18.16)
`4.944 (1.25-19.53)
`3.361 (1.41-8.01)
`8.067 (0.91-71.39)
`0.935 (0.87-1.01)
`
`0.290 (0.108-0.774)
`0.953 (0.893-1.017)
`2.08 (0.696-6.218)
`
`0.0003
`0.017
`0.023
`0.006
`0.061
`0.071
`
`0.013
`0.149
`0.190
`
`analysis, the estimated cumulative incidence of gastrodu-
`odenal ulceration over the following 24 weeks for the 39
`patients taking placebo was 53.5% (95% CI, 36.6% —
`70.3%) compared with 26.0% (95% CI, 12.1% —39.9%)
`(P = 0.011) in patients taking famotidine, 40 mg twice
`daily (Figure 2), The crude ulcer incidence at 4 weeks
`was 34.2% (95% CI, 19.1% —49.3%) in the placebo
`group vs. 12.8% (95% CI, 2.3% —23.3%) in the famoti-
`dine group, and at 12 weeks was 49.9% (95% CI,
`33.3% —66.5%) for placebo vs. 20.5% (95% CI, 7.8%-
`33.2%) for famotidine. Similar results were obtained
`following the per-protocol analysis.
`Famotidine also significantly reduced the incidence of
`gastric ulceration at 24 weeks when analyzed separately,
`from 41.4% (95% CI, 24.0% —58.7%) in the placebo
`group to 19.1% (95% CI, 6.3% —31.9%) in the famoti-
`dine group (P = 0.026). For duodenal ulcers, the esti-
`mated rate was 16.7% (95% CI, 2.8%-—30.6%) com-
`pared with
`7.9% (95% CI, 0.0%-16.5%) with
`famotidine (P = 0.31). Similar results were obtained
`following the per-protocol analysis.
`
`Risk Factors
`
`The Cox proportional regression analysis (Table
`2) confirmed the efficacy of famotidine. Famotidine, 40
`mg twice daily, was associated with an estimated condi-
`tional risk ratio of 0.093 (95% CI, 0.025—0.337) com-
`pared with placebo (P < 0.0003). Other significant ad-
`verse prognostic factors were as follows: not having a
`duodenal ulcer in the healing study, long time to ulcer
`healing, longer duration of arthritis,
`low HAQ score,
`and use of ketoprofen.
`included H.
`A similar analysis was performed that
`pylori status as a prognostic variable. In this analysis,
`baseline H. pylori infection was associated with a 2.08
`(range, 0.70—6.22) cumulative relative risk of relapse,
`although this did not reach statistical significance (P =
`0.19; Table 2). Estimated cumulative relapse rates in
`
`patients taking placebo were 63.6% (range, 43.5%—
`83.7%) in patients who were H. pylori positive (n = 25)
`and 23.8% (range, 0%—52.8%) in those who were H.
`pylori negative (n = 10). For famotidine, the figures were
`23.1% (range, 3.1% —43.0%) for H. pylori—positive pa-
`tients (n = 18) and 21.4% (range, 4.7% -—50.8%) for H.
`pylori—negative patients (n = 14).
`
`Secondary Efficacy Analyses
`
`Gastric mucosal injury expressed as a Lanza score
`was significantly lower in the famotidine group compared
`with placebo at the 4-week endoscopy in both the healing
`and maintenance study (P = 0.042). Datafor later endo-
`scopies could not be analyzed directly because they were
`confounded by selective dropoutof ulcer patients. Differ-
`ences between the treatment groups for abdominal pain
`scores, mean daily antacid consumption, and arthritic
`pain scores were not significant, and there was no correla-
`tion between abdominal pain and the presence or absence
`ofulceration.
`
`Safety and Withdrawals
`
`In the ulcer healing study, 1 patient died of bron-
`chopneumonia and multisystem disorder and 1 of pancre-
`atic cancerafter completing the study. One patient with-
`drew because of nausea and vomiting. In the maintenance
`study, 4 patients in the famotidine group and 7 in the
`placebo group were withdrawn for reasons other than
`ulcer relapse. One patient
`in each group was lost
`to
`follow-up evaluation, and 3 patients in the placebo group
`were unwilling to continue. There were three withdraw-
`als in both the famotidine and placebo groups because
`of adverse events; none of them were drug related.
`Famotidine was well
`tolerated as both healing and
`maintenance therapy. In the healing study, 10.6% of
`patients appeared to have drug-related adverse events,
`and 1 patient withdrew.
`In the maintenance study,
`28.2% of patients taking famotidine had adverse events,
`of which 12.8% were believed to be drug related, and
`35.9% of those in the placebo group had adverse events,
`of which 12.8% were also believed to be drug related.
`
`This study shows that famotidine, 40 mg twice
`daily, heals ulcers in arthritic patients who continue
`NSAIDtherapy and reduces subseqent relapse when con-
`tinued after successful healing. There was no significant
`retardation of healing in patients who continued NSAID
`therapy compared with those who stopped. Although
`relatively few patients stopped NSAID therapy and we
`cannot completely exclude the possibility that retarda-
`tion in those who continued was missed, this is unlikely
`
`Page 4 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`Page 4 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`June 1997
`
`FAMOTIDINE AND NSAIDs 1821
`
`to be a major effect because the healing rate in this group
`was high and similar to that previously reported for ome-
`prazole, 40 mg daily.'* Famotidine waseffective in heal-
`ing both gastric and duodenal ulcers and in those who
`were either H. pylori positive or negative and seemed to
`be effective for larger as well as smaller ulcers. Subgroup
`analysis of the relapse data must be more guarded because
`of the possibility of confounding due to the differential
`relapse rate and because numbers were small. It seems
`clear that famotidine was able to prevent gastric ulcer
`relapse. In our study, there were too few duodenal ulcers
`to know whether the relapse rate was truly reduced.
`However, there was a trend in this direction, and previous
`studies have shown that duodenal ulcers are easier to
`
`prevent than gastric ulcers using standard doses of H
`antagonists under primary prophylaxis conditions.”'? Re-
`lapse rates in patients taking famotidine were similar
`whether patients were H. py/ori positive or negative, but
`small numbers limit the confidence of this conclusion.
`
`Whether H. pylori eradication would affect the clinical
`course of NSAID users is unknown and is currently the
`subject of investigation. However, epidemiological data
`suggest that the risk of NSAID-associated ulcer compli-
`cations may not be substantially influenced by H. pylori
`status.””
`
`We used a high dose of famotidine because previous
`work has suggested that healing rates with standard doses
`of H, antagonists are reduced if NSAID therapy is con-
`tinued*!* but
`that this effect
`is abolished if a more
`profound effect with procon pumpinhibitors is used.'®
`We have also found that only the higher dose of famoti-
`dine, when used as primary prophylaxis, was capable of
`reducing significantly the incidence of gastric ulcer.'’
`Our data are compatible with earlier evidence that more
`profound acid suppression can overcomethe retardation
`of healing by NSAID therapy.'® These data are also sup-
`ported by short-term studies in volunteers that show that
`proton pumpinhibitors or high doses of famotidine are
`moreeffective than standard doses of H2 receptor antago-
`13-1
`*-!9 and more
`nists in preventing acute gastric erosions
`recent evidence, reported in abstract form, concerning
`large trials of the proton pump inhibitor omeprazole.”
`Previous studies have examined the use of both rani-
`tidine and misoprostol in the prevention of NSAID-asso-
`ciated gastroduodenal ulceration.~”'’ Misoprostol ap-
`pears to confer protection against both gastric and
`duodenal ulcers, whereas standard dosesof ranitidine pre-
`vented duodenal ulcers but were relatively ineffective
`in preventing gastric ulcers. However, in most of these
`studies, patients with ulceration at baseline endoscopy
`were either excluded from entry or a mixed group was
`studied and H. py/ori status was notestablished, Epidemi-
`
`ological evidence suggests that patients with a history of
`peptic ulcer disease are at greater risk of subsequent
`ulceration and ulcer complications.”” This phenomenon
`was also observed in previous studies in that the rate of
`ulceration observed endoscopically was greater in patients
`with an ulcer history’ or in those in whom ulcers were
`healed before entry.** Our study directly establishes the
`importance of ulceration detected during NSAID useas
`a risk factor for further ulceration, because the placebo
`relapse rate was 32.3%, significantly higher than the
`25.8% we found in patients who did not have ulcers at
`baseline endoscopy and who were studied under primary
`prophylaxis conditions.'? We are confident that this dif-
`ference is likely to be genuine for a variety of reasons.
`Patients in the two studies were drawn from the same
`
`populations and studied concurrently to an identical pro-
`tocol. All examinations were conducted by two endoscop-
`ists, each of whom agreed on criteria for ulcer diagnosis,
`thereby avoiding the large interobserver variation in the
`assessment of NSAID-associated gastroduodenal lesions
`that is likely to characterize other large studies conducted
`in multiple centers.” The high subsequent relapse rate
`associated with detection of an ulcer in NSAID users has
`important implications for management and reinforces
`the notion that this is a group in whom maintenance
`treatment should be considered if the NSAID therapy
`cannot be stopped. Our data suggest that treatment with
`famotidine, 40 mg twice daily, would be appropriate to
`consider for these patients.
`
`References
`
`1. Griffin MR, Piper JM, Daugherty JR, Snowden M, Ray WA. Non-
`steroidal anti-inflammatory drug use and increasedrisk for peptic
`ulcer disease in elderly persons. Ann Intern Med 1991;114:
`257-263.
`2. Guess HA, West R, Strand LM, Helston D, Lydick EG, Bergman
`U, Wolski K. Fatal upper gastrointestinal haemorrhage or perfora-
`tion among users and nonusers of anti-inflammatory drugs in
`Saskatchewan, Canada, 1983. J Clin Epidemiol 1988;41:35-
`45.
`3. Hawkey CJ. Non-steroidal anti-inflammatory drugs and peptic ul
`cers. Facts and figures multiply, but do they add up? Br Med J
`1990; 300:278-284.
`4. Langman MJS, Weil J, Wainwright P, Lawson DH, Rawlins MD,
`Logan RFA, Murphy M, Vessey MP, Colin—Jones DG. Risk of
`bleeding peptic ulcer associated with individual non steroidal
`anti-inflammatory drugs. Lancet 1994; 343:1075-1078.
`5. Ehsanullah RSB, Page MC,Tildesley G, Wood JR. Prevention of
`gastroduodenal damage induced by non-steroidal anti-inflamma-
`tory drugs: controlled trial of ranitidine. Br Med J 1988;297:
`1017-1020.
`6. Graham DY, White RH, Moreland LW, Schubert T, Katz R, Jaszew-
`ski R. Duodenal and gastric ulcer prevention with misoprostolin
`arthritic patients taking NSAIDs: Misoprostol Study Group. Ann
`Intern Med 1993; 119:257-262.
`7. Fries JF, Williams CA, Bloch DA, Michel BA. Nonsteroidal anti-
`inflammatory drug—associated gastropathy: incidence and risk
`factor models, Am J Med 1991; 91:213-222,
`
`Page 5 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`Page 5 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`1822 HUDSON ET AL.
`
`GASTROENTEROLOGY Vol. 112, No. 6
`
`10.
`
`11.
`
`12.
`
`13.
`
`14.
`
`15.
`
`. Laporte J-R, Carme X, Vidal X, Moreno V, Juan J. Upper gastroin-
`testinal bleeding in relation to previous use of analgesics and
`non-steroidal anti-inflammatory drugs. Lancet 1991; 337:85-89.
`. Graham DY, Agrawal NM, Roth SH. Prevention of NSAID-induced
`gastric ulcer with misoprostol: multicentre, double blind, placebo
`controlled trial. Lancet 1988; 2:1277-1280.
`Agrawal NW, Roth S, Graham DY, White RH, Germain B, Brown
`JA, Stromatt SC. Misoprostol compared with sucralfate in the
`prevention of non steroidal anti-inflammatory drug induced gas-
`tric ulcer: a randomised controlled trial. Ann Intern Med 1991;
`115:195-200.
`Silverstein FE, Graham DY, Senior JR, Davies HW, Struthers BJ,
`Bittman RM, et al. Misoprostol reduces serious gastrointestinal
`complications in patients with rheumatoid arthritis receiving non
`steroidal anti-inflammatory drugs: a randomised, double blind
`placebo controlled trial. Ann Intern Med 1995;123:241-249.
`Roth S, Agrawal N, Mahowald M, Montoya H, Robbins D, Miller
`S, Nutting E, Woods E, Crager M, Nissen C, Swabb E. Misoprostol
`heals gastroduodenalinjury in patients with rheumatoid arthritis
`receiving aspirin. Arch Intern Med 1989;149:775-779.
`Daneshmend TK, Pritchard PH, Bhaskar NK, Millns P, Hawkey
`CJ. Use of microbleeding and an ultrathin endoscope to assess
`gastric mucosal protection by famotidine. Gastroenterology
`1989; 97:944-949.
`Daneshmend TK, Stein AG, Bhaskar NK, Hawkey CJ. Abolition
`by omeprazole of aspirin induced gastric mucosal
`injury in hu-
`mans. Gut 1990;31:514-517.
`Cole AT, Brundell S, Hudson N, Hawthorne AB, Mahida YR,
`Hawkey CJ. Ranitidine: differential effects on gastric bleeding
`and gastric mucosal damage inducedby aspirin. Aliment Pharma-
`col Ther 1992;6:707-715.
`Lanza FL. A double blind study of the prophylactic effects of
`misoprostol on lesions of the gastric and duodenal mucosain-
`duced by oral administration of tolmetin in healthy subjects. Dig
`Dis Sci 1986; 31(Suppl):131-136.
`Taha AS, Hudson N, Trye P, Cottrell J, Mann S, Swannell AJ, Trye
`PN, Cottrell J, Mann SG, Simon TJ, Sturrock RD, Russell RI.
`Prevention of NSAID related gastric and duodenal ulcers by fa-
`motidine: a placebo controlled double blind study. N Engl J Med
`1996; 334:1435-1439.
`Walan A, Bader J-P, Glassen M, Lammern CB, Piper DW. Effect
`
`16.
`
`17.
`
`18.
`
`of omeprazole and ranitidine on ulcer healing and relapse rates
`in patients with benign gastric ulcer. N Engl J Med 1989;320:
`69-75.
`Robinson MG, Griffin JW, Bowers J, Kogan FJ, Kogut DG, Lanza
`FL, Warner CW.Effect of ranitidine on gastroduodenal mucosal
`damage induced by non steroidal anti-inflammatory drugs. Dig
`Dis Sci 1989; 34:424-428.
`Cullen DJE, Hawkey GM, Humphries H, Cave R, Sheperd V, Logan
`RFA, Hawkey CJ. Role of non steroidal anti-inflammatory drugs
`and Helicobacter pylori in bleeding peptic ulcer (abstr). Gastroen-
`terology 1994; 106:A66.
`Lancaster-Smith MJ, Jaderberg MR, Jackson DA. Ranitidine in
`the treatment of non-steroidal anti-inflammatory drug associated
`gastric and duodenal ulcers. Gut 1991;32:252-256.
`Bank S, Greenberg RE, Zucker S. Ranitidine and non steroidal
`anti-inflammatory drug (NSAID) associated gastric and duodenal
`ulceration. Gut 1991;32:963-964.
`Hawkey CJ, Swannell AJ, Eriksson S, Walan A, Lofberg |, Taure
`E, Wicklund |, Yeomans ND. Benefit of omeprazole over misopros-
`tol in healing NSAID associated ulcers (abstr). Gastroenterology
`1996;110:A131.
`Elliot SD, Yeomans ND, Buchanan RRC, Duckham ML, Boyden
`KN, Newnham RG, Smallwood RA.Long term effects of misopros-
`tol on gastropathy induced by non steroidal anti-inflammatory
`drugs (NSAIDs) (abstr). Gastroenterology 1990; 98:A40.
`Hudson N, Everitt S, Hawkey CJ. Inter observer variation in the
`endoscopic classification of NSAID associated gastric lesions.
`Gut 1994; 35:1030-1032.
`
`19.
`
`20.
`
`21.
`
`22.
`
`23.
`
`24.
`
`25.
`
`Received June 19, 1996. Accepted February 4, 1997.
`Address requests for reprints to: Christopher J. Hawkey, M.D.,
`F.R.C.P., Division of Gastroenterology, University Hospital, Notting-
`ham NG7 2UH, England. Fax: (44) 0115-9422232.
`This study was developed in collaboration with Merck Research
`Laboratories UK following an approach from the academic investiga-
`tors and was funded by Merck Research Laboratories UK.
`The authors thank Dr. G. Birnie, Dr. D. H. Bossingham, and Dr.
`J. K. Lloyd-Jones for their help in recruiting patients to the study, and
`David Thompson of Applied Statistics for performing the statistical
`analysis.
`
`Page6 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`Page 6 of 12
`
`Patent Owner Ex. 2049
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`GASTROENTEROLOGY 1997;112:1817-1822
`
`Famotidine for Healing and Maintenancein Nonsteroidal Anti-
`inflammatory Drug—Associated Gastroduodenal Ulceration
`
`NICHOLAS HUDSON,* ALI S. TAHA,* ROBIN |. RUSSELL,* PENELOPE TRYE,® JEREMY COTTRELL,®
`STEPHEN G. MANN, ANTHONY J. SWANELL,' ROGER D. STURROCK,*
`and CHRISTOPHER J. HAWKEY*
`* Division of Gastroenterology, University Hospital, Nottingham, England; ‘Department of Gastroenterology and Rheumatology, Glasgow
`Royal Infirmary, Glasgow, Scotland; Merck Research Laboratories UK, Hoddesdon, Herts, England; and ''Department of Rheumatology,
`University and City Hospitals, Nottingham, England
`
`See editorial on page 2143.
`
`Background & Aims: Nonsteroidal anti-inflammatory
`drugs (NSAIDs) are strongly associated with gastroduo-
`denal ulceration. How to manage patients with NSAID-
`associated ulcers is a commonclinical dilemma. High-
`dose famotidine in the healing and maintenance of
`NSAID-associated gastroduodenal
`ulceration was
`therefore evaluated. Methods: One hundred four pa-
`tients with rheumatoid or osteoarthritis who had gastro-
`duodenal ulceration received famotidine, 40 mg twice
`daily. Sixteen patients stopped and 88 continued their
`NSAID treatment. Ulcer healing was assessed endo-
`scopically at 4 and 12 weeks. Seventy-eight NSAID
`users with healed ulcers were then randomized to re-
`
`ceive 40 mg twicedaily famotidine or placebo and un-
`derwent endoscopy at 4, 12, and 24 weeks. Results:
`Cumulative ulcer healing rates at 12 weeks were 89.0%
`(95% confidence interval [Cl], 82.3%-—95.7%) for pa-
`tients who continued NSAID treatment and 100% (95%
`Cl, 82.9%-100.0%) for those who stopped.