Naproxcinod: A first-in-class cyclooxygenase-inhibiting nitric oxide donor for osteoarthritis pain

Key Points

Abstract

Naproxcinod, a cyclooxygenase-inhibiting nitric oxide donor, is pending FDA approval for the indications of knee and hip osteoarthritis. Treating osteoarthritis pain can be challenging because many agents commonly used for this indication carry potential risk for increased cardiovascular events including increased blood pressure, increased upper gastrointestinal (GI) bleeding, and increased hepatotoxicity. Naproxcinod combines the action of the parent drug, naproxen, with nitric oxide, with the intent of minimizing potential hypertensive and GI adverse effects while maintaining the analgesic effects. Available clinical trial data indicates that naproxcinod has superior efficacy for pain compared with placebo, does not increase blood pressure, but has not been shown to lower the incidence of gastroduodenal ulcer compared with naproxen therapy. The most commonly reported adverse effects include headache, back pain, arthralgia, nausea, dyspepsia, and upper abdominal pain. Unresolved issues include gastroprotective effects of naproxcinod, recommended dosing in patients with chronic kidney disease, and the drug interaction profile of this agent. (Formulary. 2010;45:116-122)

Osteoarthritis is a degenerative disorder affecting an estimated 26 million individuals worldwide, and in the United States, is associated with direct cost in excess of $2,600 annually per patient.¹ Osteoarthritis is characterized by decoupling of normal degenerative and regenerative repair processes, ultimately leading to articular cartilage fragmentation, thinning, and denudation, with associated mobility loss and disability. Although osteoarthritis has been thought to arise from excessive wear and tear, proinflammatory cytokines such as interleukin-1 may also affect the joints.²

The Osteoarthritis Research Society International (OARSI) recently summarized new research on hip and knee osteoarthritis therapies published between January 31, 2006 and January 31, 2009.⁴ Among the available oral therapies, opioids were ranked most effective for pain followed by chondroiton sulfate, glucosamine, selective cyclooxygenase (COX)-2 inhibitors, avocado soybean unsaponifiables, rosehip powder, nonsteroidal anti-inflammatory drugs (NSAIDs), diacerhein, and acetaminophen. Therapies with the greatest body of evidence are opioids, COX-2 inhibitors, NSAIDs, and acetaminophen and these are most commonly prescribed.

Despite its low efficacy, OARSI recommends acetaminophen as a first-line, long-term therapy for knee and hip osteoarthritic pain.⁴ However, the recent research indicates that higher doses of acetaminophen are associated with upper gastrointestinal (GI) side effects, mild renal dysfunction in women, and increased hypertension.⁴ Furthermore, FDA recently recommended that single doses of acetaminophen be limited to 650 mg and that daily doses not exceed 4 g because of hepatotoxicity concerns.⁵

While effective, 25% of patients taking opioids experience side effects leading to medication discontinuation. Adverse effects are more frequent among users of stronger opioid medications, and recent research shows that patients with noncancer pain reported that only strong opioids were more effective than NSAIDs or acetaminophen.⁴

Beginning with the withdrawal of rofecoxib in 2004, NSAIDs, including selective and nonselective COX inhibitors, have been associated with well-established cardiovascular and GI risks and carry related boxed label warnings.^4,6-8 As a result, an international panel of experts has recommended clinical practice guidelines for managing patients taking NSAIDs.⁷ They state that a nonselective NSAID is appropriate for those at average GI risk, and either a nonselective NSAID or a COX-2 inhibitor is appropriate for patients with average cardiovascular risk. Patients with increased GI risk should receive a nonselective NSAID with concomitant proton pump inhibition or misoprostol, and those at high cardiovascular risk should receive naproxen. They recommend no NSAID therapy as appropriate for patients having elevated risk for both GI and cardiovascular events.⁷ Subsequently, results of a 1999 to 2004 retrospective cohort study highlighted the potential advantage of naproxen versus lower doses of diclofenac and higher doses of ibuprofen, celecoxib, or rofecoxib with regard to cardiovascular events in patients with established heart disease.⁸ The mechanism through which NSAIDs, with the possible exception of naproxen, increase cardiovascular risk remains unclear, but the propensity for NSAIDs to elevate blood pressure has been proposed as a possible contributing factor.⁹

In light of the poor efficacy, side effects, and increased cardiovascular and GI risks associated with the commonly prescribed therapies for hip and knee osteoarthritis pain, researchers have focused on developing safer therapeutic alternatives. Naproxcinod (pronounced na proks' sin od), 1 of 5 COX-inhibiting nitric oxide donators (CINOD), combines naproxen with nitric oxide, in an effort to minimize potential hypertensive and adverse GI effects while maintaining analgesic effects.¹⁰ The FDA Arthritis and Drug Safety and Risk Management Advisory Committees will meet to consider the new drug application for naproxcinod on May 12, 2010.¹¹

CHEMISTRY AND PHARMACOLOGY

Naproxcinod (4-(nitrooxy)butyl-(2S)-2-(6-methoxy-2-naphthyl) propanoate) belongs to a new class of CINODs, with a molecular weight of 347.4 g/mol. This agent consists of the nonselective NSAID naproxen, linked through a butyl group to a nitric oxide donating moiety.¹² Nitric oxide may contribute to maintaining gastric mucosa integrity and circulation by inhibiting gastric acid secretion and enhancing gastric mucus alkaline secretion. Furthermore, nitric oxide may counteract NSAID-induced gastric damage secondary to increased leukocyte adherence, impairment of mucosal repair, and decreased gastric blood flow.^13,14 Nitric oxide has been shown to activate guanylyl cyclase, which allows for the conversion of GTP into cyclic GMP, the second messenger primarily responsible for nitric oxide signaling.¹⁴ In addition to its gastroprotective effects, nitric oxide signaling has been shown to exert potentially beneficial cardiovascular effects, because vascular nitric oxide generation controls blood pressure. The inhibitory effects of nitric oxide on vascular smooth muscle cell proliferation, prevention of low-density lipoprotein oxidation, and macrophage activation may slow atherogenesis.¹⁴

Arachidonic acid, the precursor of prostaglandins, is formed by the metabolism of phospholipids in the cell membranes by phospholipase A₂. COX-1, -2, and -3 convert arachidonic acid to prostaglandin (PGG₂) and peroxidases convert PGG₂ to PGH₂. Various stimuli activate tissue-specific isomerases to convert PGH₂ to PGD_2a, PGE₂, prostacyclin (PGI₂), PGF₂ and thromboxane (TXA₂), which bind respectively to DP, EP, IP, FP, and TP receptors.¹⁵ Because NSAIDs inhibit the production of PGE₂ and PGI₂, they have been used extensively as analgesic, antipyretic, and anti-inflammatory agents. However, decreased gastric prostaglandin synthesis induces mucosal damage through leukocyte adherence to the vascular endothelium and through reductions in gastric mucus bicarbonate secretion, along with decreased mucosal blood flow. Moreover, since PGE₂ and PGI₂ play key roles in the regulation of electrolyte excretion and blood circulation in the kidneys, adverse effects such as sodium retention, hyperkalemia, and decreased glomerular filtration rate may emerge in patients with prostaglandin-dependent renal blood flow.^15,16 The detrimental effects of COX inhibition in the kidney could theoretically lead to increases in blood pressure, given the major role that PGI₂ has in regulating blood pressure and renal filtration.¹⁴

PHARMACOKINETICS

Following oral administration, naproxcinod undergoes hydrolysis in the intestinal tract to form naproxen, nitric oxide, and nitrates.^12,17 The fractions (f_a) of intact naproxcinod and naproxen absorbed from human small intestine data are estimated using the following assumptions: (1) small intestine radius (r) of 1.75 cm, (2) transit time of 3 hours (tres), and (3) use of the equation f_a=1 – e^{-(5.6x(GI permeability)x(tres)/r)}. ¹² Approximately 23% to 24% of the naproxcinod capsule (self-emulsifying drug delivery system) was predicted to be absorbed, with 94% to 97% uptake of naproxen from the absorbed capsule (71% to 73% as naproxen formed within the GI tract). The investigators also predicted that a single dose of naproxcinod 750 mg would yield a maximum nitrate plasma concentration (C_max) of 60 μM, a 3- to 4-fold lower level than the concentration observed following a nitrate-rich meal (1,000 mg).

Data from 16 studies conducted with healthy volunteers and patients were analyzed to determine the C_max, time to maximum concentration (T_max), and half-life of naproxcinod. Additional pharmacokinetic parameters determined for naproxen included area under the plasma concentration-time curve (AUC), clearance, and volume of distribution. Oral consumption of naproxcinod 375-mg and 750-mg capsules by fasting subjects yielded a C_max of <4 to 64 nM and <4 to 57 nM, respectively. Concomitant intake of food with naproxcinod resulted in a 2- to 4-fold increase in C_max as seen by levels of 6 to 111 nM in subjects treated with naproxcinod 375 mg and 8 to 192 nM in subjects who received naproxcinod 750 mg. The median T_max increased from 1.5 hours to 2 hours in fasting subjects compared with nonfasting subjects. The mean C_max of naproxcinod did not display a dose-proportional increase and the C_max observed with single doses were: 375-mg dose, <4 nM; 750-mg dose, 17 nM; 1,500-mg dose, 42 nM; and 2,250-mg dose, 61 nM. The estimated terminal half-life of naproxcinod 350 mg to 750 mg approached 3 to 4 hours.

Naproxcinod 375-mg administration resulted in delayed and less-extensive uptake of naproxen compared with naproxen 250-mg administration, as shown by 27% and 7% differences in C_max and AUC, respectively, between the 2 drugs.¹² The median T_max of naproxen occurred at 3 hours after naproxcinod administration; the terminal half-life (15 to 22 hours) did not differ between subjects who took naproxcinod or naproxen. Concomitant administration of naproxcinod with food did not affect the C_max or T_max of naproxen. Among median-weight subjects taking naproxcinod 125 mg twice daily up to 750 mg twice daily, the clearance and volume of distribution were estimated to be 467 L/h and 7,660 L, respectively. Clearance increased at a rate of 0.6% per kg and the volume of distribution increased at a rate of 1.5% per kg.¹² Naproxen is metabolized through the cytochrome P450-dependent pathway (40% of metabolism) and glucuronidation to form the conjugated metabolites M₂ (ether glucuronide of 6-O-desmethyl naproxen), M₄ (sulfate conjugate of 6-O-desmethyl naproxen), M₅ (acyl glucuronide of 6-O-desmethyl naproxen), and M₈ (acyl glucuronide of naproxen).¹² The metabolites recovered in urine collected up to 48 hours following dosing accounted for 95% of the dose.

In the same pharmacokinetic analysis, the median T_max of nitrate occurred at 4 hours, following consumption of [15N]-[3H]-naproxcinod 750 mg by nitrate-restricted subjects.¹² The steady state nitrate levels of subjects on normal diets increased from a pre-dose level of 51 μM pre-dose to 90 μM after administration of naproxcinod 750 mg twice daily. Patients who did not have a nitrate-restricted diet yielded 1,630±362 μmol of [¹⁵ N] in urine over a 24-hour period. The renal clearance of [¹⁵ N]-nitrate approached 30 to 40 mL/min.

BLOOD PRESSURE RESULTS

Naproxcinod did not increase blood pressure in a phase 3, randomized, double-blind, 13-week study of 916 patients with knee osteoarthritis, including 207 patients with hypertension who were being treated with renin-angiotensin system blockers or diuretics.¹⁸ Patients with uncontrolled hypertension were excluded from the trial. Patients were randomly assigned to receive naproxcinod 750 mg or 375 mg, naproxen 500 mg, or placebo twice daily. Evaluation of blood pressure with standardized measurements was a predefined end point.

There were no differences in systolic blood pressure among patients taking either dose of naproxcinod compared with placebo and both doses showed a significant reduction in blood pressure compared with naproxen (P<.04). The naproxcinod 750-mg dose was associated with a greater significant blood pressure decrease versus naproxen (P<.02).¹⁸ The percentage of patients experiencing a systolic blood pressure increase of 10 mm Hg or more during the trial was greater for those receiving naproxen than naproxcinod and was 22% and 14%, respectively (P=.04 and P=.055, respectively). Among the 207 (23%) patients with hypertension being treated with renin-angiotensin system blockers or diuretics during the trial, the mean change in systolic blood pressure from baseline was 6.5 mmHg less in the naproxcinod 750-mg group than in the naproxen group.¹⁸

EFFICACY AND SAFETY RESULTS

Phase 3 efficacy and safety trials include an extension of the 916-patient blood pressure assessment trial to 52 weeks, a 53-week comparison of naproxcinod 375 mg and 750 mg to placebo and naproxen in 1,020 patients with knee osteoarthritis, and a 13-week comparison of naproxcinod 750 mg with placebo and naproxen in 800 patients with hip arthritis.^19–21 The results have been submitted to FDA but have not been published.

A 6-week study of 645 patients with knee osteoarthritis was conducted to define the efficacy, safety, and tolerability of naproxcinod at twice daily doses of 125 mg, 375 mg, and 750 mg.² Comparators included rofecoxib 25 mg once daily, naproxen 500 mg twice daily, and placebo. Change in the Western Ontario and McMaster Universities Score (WOMAC) pain subscale after an average 4- to 6-week treatment was the primary outcome measure, comparing naproxcinod with placebo for superiority and with rofecoxib for noninferiority. Compared with placebo, all but the naproxcinod 125-mg twice daily group showed significant mean differences on individual WOMAC parameters. In addition, the 375-mg and 750-mg doses of naproxcinod were noninferior to rofecoxib with regard to the primary end point.

In a 6-week study of 522 patients with hip and knee osteoarthritis, the efficacy, safety, and tolerability of naproxcinod 750 mg once or twice daily, and naproxcinod 1,125 mg twice daily were assessed.²³ The primary end point was mean difference in WOMAC pain score, and the comparators were placebo and rofecoxib 25 mg once daily. All patients treated with naproxcinod and rofecoxib had superior reductions in WOMAC pain subscale scores at 4 to 6 weeks, compared with placebo-treated patients (P≤.02). However, patients receiving once-daily naproxcinod displayed significantly lower mean changes in WOMAC pain subscale scores at 4 to 6 weeks compared with naproxcinod 750 mg twice daily (P=.0259), naproxcinod 1,125 mg twice daily (P=.0370), and rofecoxib 25 mg once daily (P=.0016). Naproxcinod 1,125 mg twice daily did not demonstrate enhanced efficacy compared with the 750-mg twice daily dose.

The GI tolerance of naproxcinod 750 mg twice daily compared with naproxen 500 mg twice daily and placebo was evaluated in a 6-week study including 970 patients with hip and knee osteoarthritis.²⁴ The primary end point was incidence of endoscopic gastroduodenal ulcer (minimum diameter of 3 mm). Naproxcinod-treated patients trended toward a lower incidence of gastroduodenal ulcer compared with naproxen-treated patients (9.7% vs 13.7%, P=.07), and none of the placebo-treated patients displayed endoscopic gastroduodenal ulcer. Naproxcinod showed similar efficacy as naproxen in lowering WOMAC pain scores (estimated pairwise difference 0.54; 95% CI, -1.87 to 2.95), stiffness (estimated pairwise difference 0.23; 95% CI, -2.45 to 2.91), and function (estimated pairwise difference 0.17; 95% CI, -2.11 to 2.45).

ADVERSE EFFECTS

In the phase 3 blood pressure trial that has been reported, investigators observed no significant changes from baseline in serum creatinine, potassium, sodium, or blood urea nitrogen in patients being treated with naproxcinod, naproxen, or placebo.¹⁸ Symptoms possibly related to low systolic blood pressure such as dizziness, near syncope, and lightheadedness occurred in 7.3% of patients taking naproxcinod 750 mg BID, 2.1% of those taking the 375-mg BID dose, in 1.3% of the naproxen-treated patients, and in 3.2% of placebo patients. One patient in the naproxcinod 750-mg group had a myocardial infarction during the trial and one placebo group patient developed congestive heart failure. Both patients had cardiovascular disease at study entry.¹⁸

In one phase 2 study, naproxcinod-treated patients showed a 30% lower incidence of endoscopic gastroduodenal ulcers compared with naproxen-treated patients, but this difference did not achieve statistical significance (P=.07).²⁴

Moreover, two other phase 2 studies showed higher rates of predefined NSAID-related GI adverse effects in naproxcinod-treated patients compared with rofecoxib-treated patients and/or naproxen-treated patients.^22,23 In the first, pre-defined NSAID-related GI adverse effects included dyspepsic signs and symptoms, flatulence, bloating/distention, nausea, vomiting, abdominal pain, and diarrhea.²² Twenty percent to 23% of naproxcinod-treated patients experienced GI adverse effects compared with 17% of rofecoxib-treated patients and 15% of naproxen-treated patients. Similarly, in the second phase 2 trial, 45% of patients receiving naproxcinod 750 mg twice daily and 48% of patients receiving naproxcinod 1,125 mg twice daily experienced GI effects (dyspepsia, nausea, upper abdominal pain, gastroesophageal reflux, diarrhea, loose stools, constipation), compared with 35% of rofecoxib-treated patients.²³ Other commonly reported adverse effects observed in naproxcinod recipients from phase 2 clinical trials include headache, back pain, and arthralgia.^22–24

Drug Interactions

Drug interaction information for naproxcinod has not been published. Absorption of the parent compound, naproxen, may be delayed with concomitant use of cholestyramine, antacids (magnesium oxide or aluminum hydroxide), or sucralfate.²⁵ Inhibition of renal prostaglandin synthesis by naproxen may increase mean minimum lithium concentration as much as 15% and reduce lithium clearance by 20%, and may reduce the natriuretic effects of furosemide and thiazides.²⁵ Naproxen has been shown to decrease tubular secretion of methotrexate in animal models, thereby increasing the risk of methotrexate toxicity.²⁵ Use of aspirin, warfarin, or selective serotonin reuptake inhibitors with naproxen can increase the potential for bleeding in susceptible patients.²⁵

DOSING AND ADMINISTRATION

The NDA filed with FDA is for a 375-mg orally administered dose.¹¹ In clinical trials naproxcinod 375 mg was administered twice daily.^19–21 Because naproxcinod releases a nitric oxide moiety, patients with a history of orthostatic hypotension should theoretically avoid using this agent.^22–24

FORMULARY CONSIDERATIONS

Naproxcinod, 1 of 5 CINODs currently undergoing investigation, represents a novel agent that combines the action of the parent drug naproxen with nitric oxide. The rationale of combining a nitric oxide moiety with naproxen was to minimize potential adverse effects on blood pressure, renal function, and to minimize the risk of GI injury, while maintaining the analgesic effects of naproxen.

However, one phase 2 trial showed a nonsignificant reduction in the incidence of endoscopic gastroduodenal ulcers when comparing naproxcinod and naproxen.²⁴ A phase 1 study in 60 healthy volunteers indicated that the nitric oxide moiety of naproxcinod did not confer renal-sparing effects, as shown by a similar magnitude of reduction in glomerular filtration rate (GFR) in subjects taking this agent and other NSAIDs, along with significant reductions in sodium excretion and potassium excretion.¹⁶ During normal sodium intake, recipients of non-naproxen treatment regimens showed maximal absolute changes from baseline GFRs of -16.7 mL/min (placebo) up to -31.3 mL/min (rofecoxib). None of the recipients of naproxcinod or rofecoxib showed statistically significant differences in maximal absolute changes from baseline GFR compared with placebo, whereas patients treated with naproxen 500 mg twice daily did show a significantly larger maximal reduction in GFR compared with placebo (-36.5 mL/min, P<.05).¹⁶ Similarly, during low sodium intake, recipients of non-naproxen treatment regimens showed maximal absolute changes from baseline GFRs of -13.9 mL/min (placebo) up to -33.7 mL/min (naproxcinod 1,500 mg).¹⁶

Issues that have not been addressed in published clinical trials include the gastroprotective effects of naproxen and misoprostol or a PPI, compared with naproxcinod, the drug interaction profile of naproxcinod with focus on the nitric oxide moiety, and appropriate dosing of naproxcinod in renally impaired patients. Until long-term safety and efficacy data from phase 3 clinical trials become available as published research reports, naproxcinod's place on a formulary remains unclear. However, given its advantageous blood pressure profile, this agent may be a viable option for osteoarthritis patients at high cardiovascular risk, who require NSAID therapy for pain relief.

Ms Diep and Ms Bussard are PharmD candidates at University of the Pacific School of Pharmacy, Stockton, Calif. Dr Song is an associate professor of pharmacy practice at the University of the Pacific School of Pharmacy and PGY1 pharmacy resident coordinator for Santa Clara Valley Medical Center in San Jose, Calif.

Disclosure Information: The authors report no financial disclosures as related to products discussed in this article.

In each issue, the "Focus on" feature reviews a newly approved or investigational drug of interest to pharmacy and therapeutics committee members. The column is coordinated by Robert A. Quercia, MS, RPh, clinical manager, Department of Pharmacy Services, Hartford Hospital, Hartford, Conn, and adjunct associate professor, University of Connecticut School of Pharmacy, Storrs, Conn; and by Craig I. Coleman, PharmD, associate professor of pharmacy practice, University of Connecticut School of Pharmacy, and director, Pharmacoeconomics and Outcomes Studies Group, Hartford Hospital.

EDITORS' NOTE: The clinical information provided in "Focus on" articles is as current as possible. Due to regularly emerging data on developmental or newly approved drug therapies, articles include information published or presented and available to the author up until the time of the manuscript submission.

REFERENCES

1. Berenbaum F. New horizons and perspectives in the treatment of osteoarthritis. Arthritis Res Ther. 2008; 10(suppl 2):S1–S7.

2. Das SK. Osteoarthritis. Best Pract Res Clin Rheumatol. 2008; 22:657–675.

3. Glass GG. Osteoarthritis. Dis Mon. 2006;52:343–362.

4. Zhang W, Nuki G, Moskowitz RW, et al. OARSI recommendations for the management of hip and knee osteoarthritis Part III: changes in evidence following systemic cumulative update of research published through January 2009. Osteoarthritis Cartilage. 2010. Feb 11; [Epub ahead of print].

5. FDA. Acetaminophen Information. 2009. Available at: http://www.fda.gov/Drugs/DrugSafety/informationbydrugclass/ucm165107.htm. Accessed March 22, 2010.

6. MedWatch: The FDA Safety Information and Adverse Event Reporting Program. Vioxx (rofecoxib). 2005. Available at: http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm152743.htm. Accessed March 22, 2010.

7. Chan FKL, Abraham NS, Scheiman JM, et al. Management of patients on nonsteroidal anti-inflammatory drugs: a clinical practice recommendation from the first International Working Party on Gastrointestinal and Cardiovascular Effects of Nonsteroidal Anti-Inflammatory Drugs and Anti-Platelet Agents. Am J Gastroenterol. 2008;103:2908–2918.

8. Ray WA, Varas-Lorenzo C, Chung CP, et al. Cardiovascular risks of nonsteroidal antiinflammatory drugs in patients after hospitalization for serious coronary heart disease. Circ Cardiovasc Qual Outcomes. 2009;2:155–163.

9. Chan CC, Reid CM, Aw TJ, Liew D, Haas SJ, Krum H. Do COX-2 inhibitors raise blood pressure more than nonselective NSAIDs and placebo? An updated meta-analysis. J Hypertens. 2009;27:2332–2341.

10. Fiorucci S. Prevention of nonsteroidal anti-inflammatory drug-induced ulcer: looking to the future. Gastroenterol Clin N Am. 2009;38:315–332.

11. FDA. Joint Meeting of the Arthritis Advisory Committee and the Drug Safety and Risk Management Advisory Committee Meeting Announcement. 2010. Available at: http://www.fda.gov/AdvisoryCommittees/Calendar/ucm203426.htm. Accessed March 22, 2010.

12. Fagerholm U, Bjornsson MA. Clinical pharmacokinetics of the cyclooxygenase inhibiting nitric oxide donator (CINOD) AZD3582. J Pharm Pharmacol. 2005;57:1539–1554.

13. Berndt G, Grosser N, Hoogstraate J, Schröder H. A common pathway of nitric oxide release from AZD3582 and glyceryl trinitrate. Eur J Pharm Sci. 2004;21:331–335.

14. Wallace JL, Viappiani S, Bolla M. Cyclooxygenase inhibiting nitric oxide donators for osteoarthritis. Trends Pharmacol Sci. 2009;30:112–117.

15. Harirforoosh S, Fakhreddin J. Renal adverse effects of nonsteroidal anti-inflammatory drugs. Expert Opin Drug Saf. 2009;8:669–681.

16. Huledal G, Jonzon B, Malmenas M, et al. Renal effects of the cyclooxygenase-inhibiting nitric oxide donator AZD3582 compared with rofecoxib and naproxen during normal and low sodium intake. Clin Pharmacol Ther. 2005;77:437–450.

17. Hawkey CJ, Jones JI, Atherton CT, et al. Gastrointestinal safety of AZD3582, a cyclooxygenase inhibiting nitric oxide donator: proof of concept study in humans. Gut. 2003;52:1537–1542.

18. White WB, Schnitzer TJ, Fleming R, Duquesroix B, Beekman M. Effects of the cyclooxygenase inhibiting nitric oxide donator naproxcinod versus naproxen on systemic blood pressure in patients with osetoarthritis. Am J Cardiol. 2009;104:840–845.

19. National Institutes of Health. Analgesic Efficacy and Safety Study of Naproxcinod in Subjects With Osteoarthritis of the Knee. 2008. Available at: http://clinicaltrials.gov/ct2/show/NCT00542555?term=naproxcinod&rank=8/. Accessed March 29, 2010.

20. National Institutes of Health. Efficacy and Safety Study of Naproxcinod in Subjects With Osteoarthritis of the Knee. 2009. Available at: http://clinicaltrials.gov/ct2/show/NCT00504127?term=naproxcinod&rank=4/. Accessed March 22, 2010.

21. National Institutes of Health. Efficacy and Safety Study of Naproxcinod in Subjects With Osteoarthritis of the Hip. 2009. Available at: http://clinicaltrials.gov/ct2/show/NCT00541489?term=naproxcinod&rank=6/. Accessed March 22, 2010.

22. Schnitzer TJ, Kivitz AJ, Lipetz RS, et al. Comparison of the COX-inhibiting nitric oxide donator AZD3582 and rofecoxib in treating the signs and symptoms of osteoarthritis of the knee. Arthritis Rheum. 2005;53:827–837.

23. Karlsson J, Pivodic A, Aguirre D, Schnitzer TJ. Efficacy, safety, and tolerability of the cyclooxygenase-inhibiting nitric oxide donator naproxcinod in treating osteoarthritis of the hip or knee. J Rheumatol. 2009;36:1290–1297.

24. Lohmander LS, McKeith D, Svensson O, et al. A randomised, placebo controlled, comparative trial of the gastrointestinal safety and efficacy of AZD3582 versus naproxen in osteoarthritis. Ann Rheum Dis. 2005;64:449–456.

25. Naproxen [package insert]. Nutley, NJ: Roche; 2000.