Newer, safer nonsteroidal
`anti-inflammatory drugs
`Rational NSAID selection for arthritis
`
`William Bensen, MD, FRCPC
`
`Angelo Zizzo, MD, CCFP, FRCP
`
`OBJECTIVE To summarize current evidence that three new additions to nonsteroidal anti-inflammatory
`drugs (NSAIDs) offer comparable efficacy with fewer adverse effects than established NSAIDs.
`QUALITY OF EVIDENCE No large randomized controlled trials (RCTs) have compared all important NSAIDs.
`Several RCTs have shown that H2 antagonists do not protect against NSAID side effects, but some RCTs
`compared the protective effect of misoprostol (Cytotec) used with other NSAIDs; others have compared
`etodolac (Ultradol) or nabumetone (Relafen) with placebo and naproxen (eg, Naprosyn). Postmarketing
`surveys have been used to support claims that the new NSAIDs have few gastric or renal side effects.
`MAIN FINDINGS Using misoprostol in conjunction with traditional NSAIDs reduces gastric and renal
`adverse effects. Misoprostol can be taken at the same time as NSAIDs or in a combination tablet. Two new
`NSAIDS, etodolac and nabumetone, do not inhibit cyclooxygenase 1 prostaglandins, which occur in the
`stomach and kidneys, but more selectively block cyclooxygenase 2 prostaglandins, which cause arthritic
`inflammation. These two NSAIDs have efficacy profiles comparable to older NSAIDs but have markedly
`fewer side effects.
`CONCLUSIONS Safer treatment for arthritis can be achieved by combining misoprostol with traditional
`NSAIDs or by using one of two new agents, nabumetone or etodolac.
`
`OBJECTIF Passer en revue les preuves actuelles demontrant que trois nouveaux anti-inflammatoires non
`sterofdiens (AINS) sont d'efficacite comparable aux AINS traditionnels mais qu'ils comportent moins d'effets
`indesirables.
`QUALITE DES PREUVES A grande echelle, aucun essai randomise et controle (ERC) n'a compare l'ensemble
`des AINS importants. Plusieurs ERC ont demontre que les antagonistes des recepteurs H2 ne protegent pas
`contre les effets indesirables des AINS. Par ailleurs, certains ERC ont compare l'effet protecteur du
`misoprostol (Cytotec) en association avec d'autres AINS; d'autres essais ont compare l'etodolac (Ultradol) et
`la nabumetone (Relafen) au placebo et au naproxen (p. ex. Naprosyn). On a fait appel 'a des sondages post-
`commercialisation pour confirmer les pretentions voulant que les nouveaux AINS comportaient moins
`d'effets indesirables gastriques ou renaux.
`PRINCIPAUX RESULTATS L'utilisation concomitante du misoprostol et des AINS traditionnels reduit les
`effets indesirables gastriques et renaux. Le misoprostol peut se prendre en meme temps que les AINS ou
`combine dans un meme comprime. Deux nouveaux AINS, l'etodolac et la nabumetone, n'inhibent pas les
`prostaglandines cyclo-oxygenase-1 (COX-1) que l'on retrouve dans l'estomac et les reins, mais inhibent de
`fapon plus selective les prostaglandines cyclo-oxygenase-2 (COX-2) responsables de l'inflammation
`arthritique. Ces deux nouveaux AINS montrent des profils d'efficacite comparables 'a ceux des anciens AINS
`mais leur profil d'innocuite comporte beaucoup moins d'effets indesirables.
`CONCLUSION On peut reduire les risques du traitement de l'arthrite en combinant le misoprostol aux AINS
`traditionnels ou en prescrivant l'un des deux nouveaux agents, soit la nabumetone, soit l'etodolac.
`
`This article has been peer reviewed.
`Cet article a fait l'objet d'une evaluation externe.
`Can Fam Physician 1998;44:101-107.
`
`-*- FOR PRESCRIBING INFORMATION SEE PAGE 157
`
`VOL44: JANUARY * JANVIER 1998 + Canadian Family Physician
`
`Le Medecin defamille canadien
`
`101
`
`Page 1 of 5
`
`Patent Owner Ex. 2059
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`CME
`Newer, safer nonsteroidal
`anti-inflammatory drugs
`
`rthritis is the sleeping giant of Canadian
`health care. Arthritic diseases are the
`leading cause of chronic disability,
`workers' compensation injuries, and
`time off work.'
`More than 3 million Canadians are affected by
`osteoarthritis (OA), and 300 000 Canadians have
`rheumatoid arthritis (RA). As our population ages
`(the number of Canadians older than 65 is expected
`to double during the next 40 years), the prevalence of
`arthritic diseases, especially OA, is likely to increase
`dramatically.2
`Both RA and OA, although caused by distinct dis-
`ease processes, have the common pathway of inflam-
`mation characterized by pain, swelling, and stiffness.
`Nonsteroidal anti-inflammatory drugs (NSAIDs),
`which limit inflammation, form the backbone of cur-
`rent arthritis therapy.
`Most patients with OA or RA require NSAIDs to
`control inflammation and relieve symptoms. Despite
`the recent preoccupation with NSAID safety and the
`recommended use of acetaminophen, most Canadian
`rheumatologists believe the NSAIDs remain baseline
`treatment for both OA and RA.3 Only a few patients
`with "noninflammatory OA" get sufficient pain relief
`with acetaminophen alone.4
`All NSAIDs are not equal. The more than
`20 NSAIDs and 40 dosing options currently available
`in Canada cause confusion over which drugs are
`superior for managing arthritis. Despite the current
`lack of specific, comparative, double-blind trials, we
`have clear choices between NSAIDs based on their
`safety, efficacy, cost, and convenience.
`
`Therapeutic choice
`Despite studies comparing safety, efficacy, and tolera-
`bility of NSAIDs, there has never been (and likely
`never will be) a head-to-head comparative study of
`the relative merits of the most widely used NSAIDs.
`Most physicians rely on their own clinical experience
`and on lessons learned from multiple therapeutic tri-
`als between one or two NSAIDs.
`To manage arthritis it is important to use the
`"best" NSAIDs currently available. New evidence-
`based guidelines for caring for arthritic patients also
`
`Dr Bensen is an Associate Clinical Professor at
`McMaster University and a rheumatologist at StJoseph's
`Hospital in Hamilton, Ont. Dr Zizzo, a Fellow ofthe
`College, is an Assistant Professor in Clinical Medicine at
`McMaster University and a staffphysician at StJoseph's
`Hospital.
`
`stress this point.3 Choosing the best NSAIDs would
`profoundly affect patients' control of arthritis inflam-
`mation.
`
`Gastrointestinal complications
`Nonsteroidal anti-inflammatory drugs exert their anti-
`inflammatory effects by inhibiting prostaglandins
`(PGE2, PGI2). Because prostaglandins protect the gas-
`tric mucosa, NSAID inhibition predisposes patients to
`gastric (and to a lesser extent duodenal) ulcers,
`GI hemorrhage, and perforation. Nonsteroidal anti-
`inflammatory drugs with the most potent inhibitory
`properties (a dose-related effect) have been shown to
`produce the most GI injury.5
`Studies suggest that 15% to 20% of arthritic
`patients treated with NSAIDs develop gastric or duo-
`denal ulcers.6 Of these patients, an estimated 1% to 3%
`develop life-threatening complications, including
`upper GI bleeding and perforation.7'8 An estimated
`1900 deaths each year in Canada are caused by
`NSAID-related complications-more deaths than are
`attributed to either violence or AIDS.9
`Unfortunately, GI complications caused by
`NSAIDs usually develop without warning or symp-
`toms. Gastrointestinal safety, therefore, must be
`ensured by identifying patients at higher risk for
`NSAID-induced gastropathy. Currently, risk factors
`include:
`* being older than 60 and especially older than 70,
`* a history of peptic ulcer disease or GI bleeding,
`* a history of cardiovascular disease, and
`* using other drugs (steroids) or having severe con-
`comitant diseases.
`These risk factors are helpful in identifying the
`patients most likely to develop GI bleeding or
`perforation when they use NSAIDs.'1"3
`
`Renal effects
`Prostaglandins are key to maintaining renal function
`and to regulating renal blood flow and perfusion.
`They are also involved in sodium and water excretion
`and in glomerular filtration. Inhibition of renal
`prostaglandins (PGE2, PGI2) thus could lead to hemo-
`dynamic renal failure and retention of salt and water,
`especially among elderly patients.'4
`
`Patient fear
`Many patients fear NSAIDs because they have heard
`of or experienced NSAID toxicity. Fearful patients
`are unlikely to comply with NSAID regimens, leading
`to suboptimal efficacy, an important barrier to
`overcome in NSAID use.
`
`102 Canadian Family Physician
`
`Le Medecin defamille canadien * VOL44: JANUARY * JANVIER 1998
`
`FOR PRESCRIBING INFORMATION SEE PAGE 168 --D-
`
`Page 2 of 5
`
`Patent Owner Ex. 2059
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`Issues and controversies
`Cytoprotective strategies. Concomitant administra-
`tion of misoprostol (a prostaglandin El analog) with
`NSAIDs reverses the inhibition of protective gastro-
`prostaglandins and decreases the incidence of
`ulceration and serious complications, including
`GI hemorrhage, by 40% in patients with RA. Other
`GI agents, such as cimetidine (eg, Tagamet) or raniti-
`dine (eg, Zantac), often relieve symptoms but have
`little effect on ulceration.'5-'7 The combination of
`diclofenac and misoprostol, marketed under the trade
`name Arthrotec, has been shown to prevent
`NSAID-induced GI damage to a similar degree as
`diclofenac (eg, Voltaren) and misoprostol separately.7'18
`Thus, when patients' risk factors make them more
`likely to develop GI complications, concomitant use of
`misoprostol and an NSAID, or the combination formu-
`lation of misoprostol and diclofenac, are both appro-
`priate gastroprotective strategies. Unfortunately,
`dosing with misoprostol and an NSAID does not pro-
`tect all patients completely. Studies indicate that up to
`6% of patients receiving misoprostol alone or in combi-
`nation with diclofenac develop GI ulcers. Some
`patients do not tolerate misoprostol well. Side effects,
`including diarrhea, abdominal pain, cramping, and
`flatulence, occur with misoprostol, especially during
`the first week of therapy.",9 Although many patients
`can overcome these adverse reactions, in some they
`are intolerable and misoprostol alone or combined
`with diclofenac must be withdrawn. Clinicians and
`patients require other, potentially safer, NSAIDs.
`
`Cyclooxygenase selectivity. Until 1993 we believed
`that all NSAIDs caused potential GI and renal
`toxicity.2022 However, two distinct cyclooxygenase
`(COX) enzymes have now been identified: COX-1,
`which facilitates production of gastric and renal
`prostaglandins (the housekeeping enzyme), and
`COX-2, which is responsible for production of
`prostaglandins in inflammatory tissues. In Canada
`two new agents, etodolac (Ultradol) and nabumetone
`(Relafen), have become available as alternative
`NSAIDs during the past 3 years. These two
`drugs selectively inhibit COX-2 more than COX-1
`enzymes.23'24
`In vitro human cellular assays used to evaluate
`COX-1 versus COX-2 selectivity have shown that
`etodolac produced a seven- to 10-fold selectivity for
`COX-2 inhibition relative to COX-1.23 Other NSAIDs
`have demonstrated either COX-1 selectivity or nearly
`equal inhibition of COX-1 and COX-2.23 Laine and
`colleagues25 have demonstrated that gastromucosal
`
`CME
`Newer, safer nonsteroidal
`anti-inflammatory drugs
`
`prostaglandin production decreases significantly with
`naproxen (eg, Naprosyn) but is unchanged after
`treatment with placebo or etodolac.
`The preferential selectivity of the newer NSAIDs
`for COX-2 rather than COX-1 would be expected to
`lead to fewer adverse effects on the gastric mucosa
`and kidneys.26 Research supports this hypothesis.
`Endoscopic evaluation of normal healthy volunteers
`showed that the area of gastric ulcers after 4 weeks
`of therapy was greater with naproxen (58.3 mm2)
`than with either etodolac (13.9 mm2) or placebo
`(29.0mm2).25 Both etodolac and nabumetone cause
`less gastric ulceration than older NSAIDs.25,27,28
`Large-scale postmarketing studies involving
`almost 5000 patients with RA, OA, or ankylosing
`spondylitis identified only three adverse reactions
`related to etodolac, for an adverse reaction rate
`of 0.1%.29,30 A larger survey of more than
`50000 patients identified only 21 severe adverse
`reactions with etodolac use, for an adverse reaction
`rate of 0.0005%.30 A final evaluation of the data from
`both double-blind studies as well as open-label
`trials with etodolac found no evidence of hemor-
`rhage or perforation and a gastroduodenal ulcer
`incidence of 0.3%.2930
`Large-scale postmarketing surveys have similarly
`associated nabumetone with a very low risk of seri-
`ous adverse events. In a US-based study of close to
`2000 patients, ulcers were detected in 13 patients,
`but there were no reports of either GI bleeding or
`perforation.29 In a second study of 8865 patients
`receiving nabumetone, only two patients experi-
`enced GI bleeding; in a third trial of more than
`10000 patients, there were only 11 serious events,
`seven of them GI bleeding.3'
`A large meta-analysis has also indicated that rates
`of NSAID-associated GI complications are lower for
`newer, low-risk NSAIDs.32 Thomas Schnitzer,
`Director of Clinical Research at Northwestern
`University, reported at the international conference
`on immunopharmacology in March 1997 a study
`comparing etodolac and ibuprofen (eg, Advil) for
`GI toxicity in RA patients. Serious GI events
`occurred in 0.43% of patients with low-dose etodolac,
`0.67 patients with high-dose etodolac, and 4.7% of
`patients with high-dose ibuprofen. Both etodolac
`groups showed more clinical improvement than the
`ibuprofen group.33
`Both in vitro and postmarketing studies show that
`two newer NSAIDs, nabumetone, and etodolac,
`which are selective for COX-2, are clearly safer than
`traditional NSAIDs. Thus the doclofenac-misoprostol
`
`VOL44: JANUARY * JANVIER 1998 + Canadian Family Physician . Le Medecin defamille canadien
`
`105
`
`Page 3 of 5
`
`Patent Owner Ex. 2059
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`CME
`Newer, safer nonsteroidal
`anti-inflamato y drugs
`
`Table 1. Approximate costs of nonsteroidal anti-inflammatory drugs
`
`DRUG NAME
`Tiaprofenic acid (eg, Surgam SR)
`
`DOSE
`300mg
`
`NUMBER OF PILLS
`60
`
`PRICE PER MONTH
`($9.95 DISPENSING FEE INCLUDED)
`$53.81
`
`................................................................................................................................................................................................................................
`
`Tiaprofenic acid (generic)
`
`300mg
`
`60
`
`$36.60
`
`................................................................................................................................................................................................................................
`
`Piroxicam (eg, Feldene)
`
`20 mg
`
`30
`
`$34.56
`
`................................................................................................................................................................................................................................
`
`50mg
`
`90
`
`$49.87
`
`Diclofenac (eg, Voltaren)
`Diclofenac and misoprostol (Arthrotec)
`
`................................................................................................................................................................................................................................
`
`50 mg and 200 ,ug
`
`90
`
`$79.61
`
`................................................................................................................................................................................................................................
`
`Etodolac (Ultradol)
`
`300 mg
`
`60
`
`$65.15
`
`................................................................................................................................................................................................................................
`
`$79.61
`
`$39.84
`
`............................................................................................................................................................................................................................
`
`Nabumetone (Relafen)
`Misoprostol (Cytotec)
`
`500mg
`
`90
`
`200,ug
`
`60
`
`combination, nabumetone, or etodolac represent
`the safest alternatives for patients who cannot toler-
`ate misoprostol or who require higher NSAID doses
`than usual.
`
`Efficacy. All forms of arthritis are not the same.
`Clinical experience suggests patients respond differ-
`ently to different NSAIDs. After a period of initial suc-
`cess, patients frequently report that an NSAID loses
`its efficacy and an alternative drug must be found.
`Having alternative NSAIDs is, therefore, critical to
`appropriate management of most arthritic diseases.
`Newer, safer NSAID alternatives include the com-
`bination of diclofenac and misoprostol, nabumetone,
`and etodolac. Because these drugs have equivalent
`efficacy to the older drugs and a notable advantage in
`terms of gastric and renal toxicity, they have become
`preferred for most patients with OA and RA.
`Although etodolac is not indicated specifically for
`ankylosing spondylitis, our expenence suggests that
`it is particularly useful for the morning stiffness and
`pain of this condition.
`
`Cost. Often patients must consider cost when filling
`prescriptions, which affects compliance. Newer
`NSAIDs are similar in cost to older NSAIDs com-
`bined with misoprostol (Table 1).
`
`Key point
`Newer NSAIDs with COX-2 selectivity, such as
`nabumetone and etodolac, offer clear advan-
`tages over older medications by reducing
`gastrointestinal toxicity.
`
`Conclusion
`Although 20 NSAIDs are available in Canada,
`they are not equal. Newer NSAIDs with COX-2
`selectivity, such as nabumetone and etodolac,
`offer clear advantages over older medications in
`terms of GI toxicity. Another choice is to add
`cytoprotection as misoprostol to a more toxic
`NSAID or to take the NSAID in combination with
`diclofenac as Arthrotec. These three NSAIDs
`remain the NSAIDs of choice for managing
`arthritic diseases because the evidence currently
`available indicates they offer the best balance of
`4
`efficacy and safety.
`
`Correspondence to: Dr William Bensen, 25 Charlton
`Ave E, Suite 203, Hamilton, ON L8N1Y2
`
`References
`1. Arthritis Society of Canada. First quarter booklet Toronto:
`The Arthritis Society of Canada; 1995.
`2. Statistics Canada. Population projectionsfor Canada,
`provinces and territories (1993-2016). Ottawa: Statistics
`Canada; 1991 Cat No. 91 520.
`3. Tannenbaum H, Davis P, Russell AS, Atkinson MH,
`Maksymowych W, Huang SHK, et al. An evidence-based
`approach to prescribing NSAIDs in musculoskelatel
`disease: a Canadian consensus. Can Med Assoc J 1996;
`155(1):77-88.
`4. Brooks PM, Day RO. Nonsteroidal antiinflammatory
`drugs-differences and similarities. NEnglJMed 1991;
`324(24):1719-24.
`5. Lanza FL. A review of gastric ulcer and gastroduodenal injury
`in nonnal volunteers receiving aspirin and other non-steroidal
`anti-inflammatory drugs. ScandJ Gastroenterol Suppl 1989;
`163(Suppl):24-31.
`
`106 Canadian Family Physician. Le Mdecin defamille canadien * VOL 44: JANUARY * JANVIER 1998
`
`Page 4 of 5
`
`Patent Owner Ex. 2059
`Mylan v. Pozen
`IPR2017-01995
`
`

`

`CME
`Newer, saer nonsteroidal
`anti-inflammatory dmgs
`
`22. Levy RA, Smith DL Clinical differences among nonsteroidal
`antiinflammatory drugs: implications for therapeutic substitu-
`tion in ambulatory patients. Drug Intell Clin Pharm 1989;
`23:76-85.
`23. Glaser KM. Cyclooxygenase selectivity and NSAIDs.
`Cyclooxygenase-2 selectivity of etodolac (Lodine).
`Inflammopharmacol 1995;3:1-12.
`24. Dandona P, Jeremy JY. Nonsteroidal anti-inflammatory drug
`therapy and gastric side effects. Does nabumetone provide a
`solution? Drugs 1990;40(Suppl 5):16-24.
`25. Laine L, Sloane R, Ferretti M, Cominelli F. A randomized,
`double-blind comparison of placebo, etodolac and naproxen
`on gastrointestinal injury and prostaglandin production.
`Gastrointest Endosc 1995;42:42&33.
`26. Masferrer JL, Zweifel BS, Manning PT, Hauser SD,
`Leahy KM, Smith WG, et al. Selective inhibition of inducible
`cyclooxygenase 2 in vivo is antiinflammatory and nonulcero-
`genic. Proc Natl Acad Sci USA 1994;91:3228-32.
`27. Bianchi Porro G, Caruso I, Montrone MPF, Ardizzone S.
`A double-blind gastroscopic evaluation of the effects of
`etodolac and naproxen on the gastrointestinal mucosa of
`rheumatic patients.JIntern Med 1991;229:5-8.
`28. Roth SH. Endoscopy-controlled study of the safety of
`nabumetone compared with naproxen in arthritis therapy.
`AmJMed 1987;83:25-30.
`29. Lanza FL. Gastrointestinal toxicity of newer NSAIDs.
`AmJ Gastroenterol 1993;88:131&23.
`30. Semi U. Global safety of etodolac: reports from worldwide
`postmarketing surveillance studies. Rheumatol Int 1990;
`10(Suppl):23-7.
`31. Berhard G. Worldwide safety experience with nabumetone
`JRheumatol Suppl 1992;19(Suppl):4&56.
`32. Ferrez M, Maetzel A, Bombardier C. Meta-analysis compar-
`ing the effects of low risk nonsteroidal anti-inflammatory
`drugs (NSAIDs) on the gastric and duodenal mucosa
`[abstract]. Arthritis Rheum 1995;38(9 Suppl):S261.
`33. Schnitzer T. The International Conference on
`Inflammapharmacology. San Francisco; 1997 March 17-19.
`
`* * 0
`
`6. Larkai EN, Smith JL, Lidsky MD, Graham DY.
`Gastroduodenal mucosa and dyspeptic symptoms in arthritic
`patients during chronic nonsteroidal anti-inflammatory drug
`use. Am J Gastroenterol 1987;82:1153-8.
`7. 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 nonsteroidal anti-inflammatory drugs.
`Ann Intern Med 1995;123:241-9.
`8. Kaplan B, Swain RA. NSAIDs. Are there any differences?
`Arch Fam Med 1993;2:1167-74.
`9. ArthroScope. Toronto: The Arthritis Society; 1995.
`10. Gabriel SE, Jaakkimainen L, Bombardier C. Risk for serious
`gastrointestinal complications related to use of nonsteroidal
`anti-inflammatory drugs. A meta-analysis. Ann Intern Med
`1991;115:787-96.
`11. Fries JF. NSAID gastropathy: the second most deadly
`rheumatic disease? Epidemiology and risk appraisal.
`J Rheumatol Suppl 1991;18(Suppl 28) :6-10.
`12. Griffin MR, PiperJM, Daugherty JR, Snowden M, Ray WA.
`Nonsteroidal anti-inflammatory drug use and increased risk
`for peptic ulcer disease in elderly persons. Ann Intern Med
`1991;114:257-63.
`13. Silverstein FE, Gilbert DA, Tedesco FJ, Buenger NK, PersingJ.
`The national ASGE survey on upper gastrointestinal bleeding:
`II. Clinical prognostic factors. GastrointestEndosc 1981;27:80-93.
`14. Murray MD, Brater DC. Renal toxicity of the nonsteroidal
`anti-inflammatory drugs. Am Rev Pharmacol Toxicol 1993;
`33:435-65.
`15. Ehsanullah RS, Page MC, Tildealey G, Wood JR. Prevention of
`gastroduodenal damage induced by nonsteroidal anti-tiflamma-
`tory drugs: controlled trial of ranitidine. BMJ 1988;297:1017-21.
`16. Agrawal NM, Roth S, Graham DY, White RH, Germain B,
`Brown JA, et al. Misoprostol compared with sucralfate in the
`prevention of nonsteroidal anti-inflammatory drug-induced
`gastric ulcer. A randomized, controlled trial. Ann Intern Med
`1991;115:195-200.
`17. Oddsson E, Gudjonnsson H, Thjodleifsson B.
`Protective effect of omeprazole or ranitidine against naprox-
`en induced damage to human gastroduodenal mucosa.
`Abstracts ofthe World Congresses ofGastroenterology. Sydney:
`Medical Group (UK), Ltd, Abingdon; 1990. p. 22.
`18. McKenna F. Review of Arthrotec clinical data.
`EurJRheumatol Inflamm 1994;14(Suppl 3):12-6.
`19. Geis GS. Overall safety of Arthrotec. ScandiRheumatol
`Suppl 1992;96(Suppl):33-6.
`20. Caruso I, Bianchi Porro G. Gastroscopic evaluation of
`anti-inflammatory agents. BMJ 1980;280:75-8.
`21. Garcia-Rodriguez LA, Walker AM, Perez-Gutthann S.
`Nonsteroidal antiinflammatory drugs and gastrointestinal
`hospitalizations in Saskatchewan: a cohort study.
`Epidemiology 1992;3:33742.
`
`VOL44: JANUARY * JANVIER 1998, Canadian Family Physician
`
`Le Medecin defamille canadien
`
`107
`
`Page 5 of 5
`
`Patent Owner Ex. 2059
`Mylan v. Pozen
`IPR2017-01995
`
`